Radio Engineering by Jorma Kekalainen

Lecture notes Radio Engineering by Jorma Kekalainen
RADIO ENGINEERING
Jorma Kekalainen
Contents
Electromagnetic radiation
Radio communication
Electrical Noise
Communication link calculations
Microwaves
Maxwells equations
Application of Maxwells equations
Some passive microwave devices
Signal flow graphs and scattering parameters
Introduction to transmission line theory
Smith chart and impedance matching
Waveguide theory
Radio wave propagation in more or less free space
Microwave link design
2
1
Books
Following references can be useful:
Bennett, Electrical Noise
Freeman, Telecommunication System Engineering
Freeman, Radio System Design for Telecommunications
Forouzan, Data Communications and Networking
Gardiol, Introduction to Microwaves
Goldsmith, Wireless Communications
Krauss, Bostian, Raab, Solid State Radio Engineering
Lee, Statistical Theory of Communication
Ramo, Whinnery, Van Duzer, Fields and Waves in Communication
Electronics
Roddy, Satellite Communications
Skolnik, Introduction to Radar Systems
3
Nature of
electromagnetic radiation
2
Radiation models
Electromagnetic radiation is usually described in one of three
ways:
Ray model
In the classical physics light consisted of rays that could be
reflected and refracted through mirrors and prisms etc.
Although this simple description cannot explain many of the
phenomena we meet in radio and optical communications, ray
model is much-used in telecommunications.
The problem is that when we try to study rays very closely
they start behaving like waves.
In general, the ray model is good when the distances involved
in the device are much larger than the wavelength.
5
Radiation models
Wave model
In the context of telecommunications, the best
way of regarding electromagnetic radiation is
to think of it as an electromagnetic wave.
It is a fair generalization to say that
electromagnetic radiation may be looked on as
a wave in situations where we are studying
transmission or propagation.
3
Radiation models
Particle model
In many contexts, electromagnetic radiation behaves as
though it consists of tiny particles called photons
(quanta).
Electromagnetic radiation may often be regarded as a particle
when we are studying its interactions with matter.
Note. We are used to thinking of electrons as classical
particles but quantum mechanics assigns to them wave
properties, matter waves, as it assigns to the electromagnetic
waves particle like properties. This is wave particle duality.
The fact is that rays, photons and waves are all useful
analogies that help us to understand what electromagnetic
radiation really is from different viewpoints. 7
Electromagnetic waves
In the electromagnetic wave theory, radiation beam is
represented by electromagnetic waves propagating in space.
An electromagnetic wave consists of two fields
an electric field and
a magnetic field.
Both of these fields have
a direction and
a strength (or amplitude).
These fields oscillate in time and space as the beam
propagates.
Within the electromagnetic wave the two fields are oriented at
90 to one another.
8
4
The fields move (by
definition at the speed of
light) in a direction at 90 to
both of them.
In three dimensions, we
could consider the electric
field to be oriented on the x-
axis, and the magnetic field
on the y-axis.
Direction of travelling
wave would then be along
the z-direction.
Travelling wave
In three dimensions, we could consider the electric field to be oriented on
the x-axis, and the magnetic field on the y-axis.
Direction of travelling wave would then be along the z-direction.
The field strength changes periodically.

The rate of oscillation is the frequency of the wave.
The distance travelled during one period of oscillation is the wavelength.
10
5
Polarization of a vector field

The polarization of any vector field is defined by the
geometrical locus of the tip of the vector, as it varies during
one period.
Hence, the polarization describes the behavior of the field (the
direction and amplitude) as a function of time.
In case of electromagnetic wave the polarization of wave is
defined by means of the electric vector, as it varies during one
period.
In an isotropic medium, the direction of oscillation is always
perpendicular to the direction of propagation.
If the direction of oscillation (in the transverse plane) is
random the wave is said to be unpolarized.
11
Note: For all thermal sources in nature, the direction of vibration is random.
Polarization types
A wave is said to be elliptically polarized if the curve traced by the
end point of the electric field vector is an ellipse (in the xy-plane).
The special cases of the elliptically polarized wave are:
circularly polarized wave
Electromagnetic wave is said to be circularly polarized if the electric
field vector uniformly rotates in the xy-plane.
So, wave is circularly polarized when the tip of the electric field vector
describes a circle.
linearly polarized wave
Electromagnetic wave is said to be linearly polarized if the electric
field vector vibrates in a constant direction in the xy-plane (plane
polarized wave).
So, wave is linearly polarized when the tip of the electric field vector
moves along a straight line.
12
6
Electromagnetic spectrum
13
14
7
Wavelength and frequency ranges of

electromagnetic radiation
15
Fractions and multiples
Note: Angstrom is a unit of length used in the study of spectra (spectroscopy). 16
8
Maxwells equations
Maxwells equations govern the propagation of
electromagnetic waves in various media, including
free space,
waveguides,
optical fibers, and
optical crystals.
All
electric,
magnetic,
electromagnetic, and
optical phenomena
are governed by the same fundamental laws of
electromagnetism.
17
Maxwells equations
These laws are written mathematically in terms
of the Maxwells equations:
D
B 0
B
E 0
t
D
H J
t
18
9
Maxwells equations
In these equations, the quantities D and B are called the electric
displacement or the electric flux density vector (in coulombs per square
meter) and magnetic induction or the magnetic flux density vector (in
webers per square meter), respectively.
These two vectors (D ,B) include the effect of the electromagnetic field on
matter.
The quantities E and H are the electric field vector (in volts per meter) and
magnetic field vector (in amperes per meter), respectively.
These two field vectors (E , H) are employed to describe an
electromagnetic field or wave.
The quantities and J are the electric charge density (in coulombs per
cubic meter) and electric current density vector (in amperes per square
meter), respectively.
The electric charge and current may be considered the source of the
electromagnetic radiation, represented by the E and H vector fields.
19
Maxwells equations
These four equations completely determine the electromagnetic field and
are the fundamental equations of the theory of such fields, that is, of
electrodynamics.
Maxwells equations cannot be solved uniquely without the known
relationships between
B and H
E and D .
To obtain a unique determination of the field vectors, Maxwells equations
must be supplemented by the so-called material equations,
D E For anisotropic
material e.g.
B H
where the parameters and are are known as the dielectric tensor (or
permittivity tensor) and the permeability tensor, respectively.
20
10
EM field and matter

When an electromagnetic field is present in matter, the electric field
can perturb the motion of electrons and produce a distribution of
charge separation.
This leads to a dipole polarization per unit volume.
D E 0 E P
where P is the electric polarization and 0 is the dielectric constant of
a vacuum.
Analogously, the magnetic field can also produce a magnetization in
materials.
B H 0 H M
where M is the magnetic polarization and 0 is the magnetic
constant of a vacuum.
21
EM field and matter
Dipole polarization in material
Vacuum
22
11
Notes
Note. The permittivity of a vacuum is constant
0=8.854 x 10-12 F/m.
The constant 0 is the permeability of a vacuum and
has a value of 0=410-7 H/m.
Note. If the material medium is isotropic, both

material parameters and reduce to scalars.
If the medium is isotropic and homogenous, material
parameters reduce to constants.
23
Material parameters
For most applications, material parameters can be assumed to
be independent of the field strengths.
However, if the fields are sufficiently strong, the dependence
of these quantities on E and H must be considered.
These strong fields produce nonlinear electromagnetic effects,
which may be harmful or useful depending on the application.
The fields can be sufficiently strong, for example, if we focus
an intense laser beam on a thin fiber or apply a strong static
electric field to an electro-optic crystal, or transmit very high
power RF beam through the atmosphere.
24
12
Sourceless medium and EM waves

In communications, we often deal with the
propagation of electromagnetic radiation in such
regions of space where both charge density and
current density are zero.
In fact, if we set = 0 and J = 0 in Maxwells
equations, we find that nonzero solutions exist.
This means that an electromagnetic field can exist
even in the absence of any charges or currents.
Electromagnetic fields occurring in media in the
absence of charges are called electromagnetic
waves.
25
Remote sensing
13
Remote sensing
27
Remote sensing
Note various paths
Source to sensor direct?
Source to surface to sensor
Sensor can also be source
radar, lidar, sonar (i.e. active remote sensing)
Reflected and emitted components
Several components of final signal captured at sensor
Note: Radar (RAdio Detection And Ranging) is an object detection system which uses radio
waves to determine the range, altitude, direction, or speed of objects.
Lidar (Light Detection And Ranging) is an optical remote sensing technology that can measure
the distance to, or other properties of, targets by illuminating the target with laser light and
analyzing the backscattered light.
Sonar (originally an acronym for SOund Navigation And Ranging) is a technique that uses sound
propagation (usually underwater, as in submarine navigation) to navigate, communicate with or
detect objects on or under the surface of the water. 28
14
Gathering information about the

environment using EM waves
Electromagnetic radiation is convenient for
information transfer because:
EM radiation travels at speed of light
the fastest possible speed
EM radiation travels along a straight line
EM radiation does not need medium to travel
EM radiation interacts with matter
EM radiation wavelength allows to interact with targets
of various size
EM radiation is everywhere, because every body in
the universe radiates EM energy
29
Electromagnetic environment on
Earth
Mainly determined by solar radiation and properties
of atmosphere
Solar radiation spectrum
30
15
Atmospheric attenuation
31
Atmospheric windows
Radiowave window
Visible window
(clouds are transparent)
32
16
Energy transport
Conduction
transfer of molecular kinetic (motion) energy due to contact
heat energy moves from T1 to T2 where T1 > T2
Convection
movement of hot material from one place to another
e.g. Hot air rises
Radiation
results whenever an electrical charge is accelerated
propagates via EM waves, through vacuum & over long distances
hence of interest for remote sensing
33
Electromagnetic radiation: Wave model
Wave model of EM energy James Clerk Maxwell (1831-1879)
Unified theories of electricity and magnetism (via

Newton, Faraday, Kelvin, Ampre)
Oscillating electric charge produces magnetic field (and
vice versa)
Can be described by four differential equations
Calculated speed of EM wave in a vacuum
34
17
Electromagnetic wave
Electric field (E) perpendicular to magnetic field (M)

Travels at velocity c (3x108 ms-1, in a vacuum)
35
Wave motion
All waves characterized
by:
Wavelength, (m)
Amplitude, a (m)
v
Velocity, v (m/s)
Frequency, f (s-1 or
Hz)
Sometimes period, T
(time for one
oscillation i.e. 1/f)
36
18
Wave motion
Velocity, frequency and wavelength related by
1

f
proportional to 1/ (constant of proportionality is wave
velocity v
v f
37
Sinusoidals
Note angles in radians (rad)

360 = 2 rad, so 1 rad = 360/2 57.3
Rad to deg. (*180/) and deg. to rad (* /180)
38
19
Maxwells equations
Four equations relating vector
E
electric (E) and vector magnetic 0
fields (B)
B 0
0 is permittivity of free space
B
E
0 is permeability of free space t
E
B 0 J 0 0
t
1
0
c 0
2
James Clerk Maxwell (1831-1879) 39
Maxwells equations
1. Gauss law for electricity: The electric flux out of any
closed surface is proportional to the total charge E
0
enclosed within the surface
2. Gauss law for magnetism: The net magnetic flux out B 0
of any closed surface is zero (i.e. magnetic monopoles
do not exist)
3. Faradays Law of Induction: Line integral of electric B

field around a closed loop is equal to negative of rate of E
change of magnetic flux through area enclosed by the loop. t
4. Amperes Law: For a static electric field, the line integral

of the magnetic field around a closed loop is proportional to E
B 0 J 00
t
the electric current flowing through the loop.
Note: is divergence operator and x is curl operator. 40
20
EM spectrum
EM Spectrum
Continuous range of EM radiation
From very short wavelengths (<10-12m)
o high energy
To very long wavelengths (cm, m, km)
o low energy
Energy is related to wavelength (and hence frequency)
41
42
21
Units
EM wavelength is m, but various prefixes
o cm (10-2m)
o mm (10-3m)
o micron or micrometer, m (10-6m)
o Angstrom, (10-8m)
o nanometer, nm (10-9)
f is cycles/second or Hertz (Hz)
Note: Sometimes it is used wavenumber, k = 1/ [m-1]. 43
Energy radiated from sun or active sensor

Energy 1/wavelength (1/)
shorter (higher f) higher energy
longer (lower f) lower energy
44
22
EM spectrum
We will see how energy is related to frequency, f (and hence

inversely proportional to wavelength, ).
When radiation passes from one medium to another, speed of
light (c) and change, hence f stays the same.
45
Electromagnetic spectrum: visible

Visible part
from visible blue (shorter )
to visible red (longer )
~0.4 to ~0.7m
Violet: 0.4 - 0.446 m
Blue: 0.446 - 0.500 m
Green: 0.500 - 0.578 m
Yellow: 0.578 - 0.592 m
Orange: 0.592 - 0.620 m
Red: 0.620 - 0.7 m
46
23
Electromagnetic spectrum: IR
Longer wavelengths (sub-mm)
Lower energy than visible
Arbitrary cutoff
IR regions covers
reflective (shortwave IR SWIR)
and emissive (longwave or thermal
IR TIR)
region just longer than visible
known as near-IR NIR.
47
Electromagnetic spectrum: microwave

Longer wavelengths
RADAR
1 mm to 1 m (300GHz-
300MHz)
various bands used by
RADAR instruments
long so low energy,
hence need to use own
energy source (active
wave)
48
24
Blackbody
All objects above absolute zero (0 K or -273 C)
radiate EM energy (due to vibration of atoms)
We can use concept of a perfect blackbody
o Absorbs and re-radiates all radiation incident upon it at
maximum possible rate per unit area (Wm-2), at each
wavelength, , for a given temperature T (in K)
Energy from a blackbody?
49
Stefan-Boltzmann Law
Total emitted radiation from a blackbody, M, in Wm-2,
described by Stefan-Boltzmann Law
M T 4
T is temperature of the object in K; and = is Stefan-
Boltzmann constant 5.6697x10-8 Wm-2K-4
So energy T4
o Tsun 6000K M,sun 73.5 MWm-2
o TEarth 300K M , Earth 460 Wm-2
50
25
51
Note that peak of suns energy around 0.5 m

o negligible after 4-6m
Peak of Earths radiant energy around 10 m
o negligible before ~ 4m
Total energy in each case is area under curve
52
26
Generalization of Stefan-Boltzmann Law
Radiation emitted from unit area of any plane surface

with emissivity of (<1) can be written
= Tn
where n is a numerical index
For grey surface where is nearly independent of, n
=4.
When radiation emitted predominantly at < m , n > 4.
When radiation emitted predominantly at > m , n < 4.
53
Peak of emitted radiation: Wiens Law

Wien deduced from thermodynamic principles that
energy per unit wavelength E() is function of T and
f (T )
E (
5
At what m is maximum radiant energy emitted?
Comparing blackbodies at different T, note mT is
constant, k 2897mK i.e. m = k/T
o m, sun 0.48m
o m, Earth 9.66m
54
27
Wiens (Displacement) Law
Increase
(displacement) in m
as T reduces
Increasing
Straight line in log-

log space
55
Particle model of radiation

Newton proposed wave theory of light (EMR)
in 1666
o Observation of light separating into spectrum
Einstein explained photoelectric effect by

proposing photon theory of light
o Photons individual packets (quanta) of energy
o Photons possess energy and momentum
Light has both wave- and particle-like

properties
o Wave-particle duality
56
28

EMR intimately related to atomic structure and energy.
Atom: positive (+) charged nucleus (protons +neutrons) &
negative () charged electrons bound in orbits.
o Electron orbits are fixed at certain levels, each level corresponding to
a particular electron energy.
o Change of orbit either requires energy (work done), or releases
energy
o Minimum energy required to move electron up a full energy level.
o Once shifted to higher energy state, atom is excited, and possesses
potential energy.
o Released as electron falls back to lower energy level.
57

As electron falls back, quantum of EMR (photons) emitted.
o Electron energy levels are unevenly spaced and characteristic of a
particular element (basis of spectroscopy)
Bohr and Planck recognized discrete nature of transitions.

Relationship between frequency of radiation (wave theory) of
emitted photon (particle theory)
E hf
E is energy of a quantum in Joules (J); h is Planck constant
(6.626x10-34Js) and f is frequency of radiation
58
29

If we remember that velocity v = f and in this case v is
actually c, speed of light then
hc
E

Energy of emitted radiation is inversely proportional to
o longer (larger) lower energy
o shorter (smaller) higher energy
Implication for remote sensing: harder to detect longer
radiation (e.g. thermal) as it has lower energy.
59
60
30
Plancks Law of blackbody radiation

Planck was able to explain energy spectrum of blackbody
Based on quantum theory rather than classical mechanics
2c 2 h 1
E
5 hc
e kT
1
dE()/d gives constant of Wiens Law
E() over all results in Stefan-Boltzmann relation
Blackbody energy function of , and T
61
Plancks Law
Explains/predicts shape of blackbody curve
Use to predict how much energy lies between given
o Crucial for remote sensing
62
31
Consequences of Plancks Law: Plants

Chlorophyll a, b absorption spectra
Photosynthetic pigments
o Basic driver of nearly all life on
Earth!
o Source of all fossil fuel
63
Applications of Plancks Law

Fractional energy from 0 to i.e. F0 ? Integrate Planck
function.
Note Eb (,T), emissive power of blackbody at , is function of
product T only, so....
Radiant energy from 0 to
E0 , T
F0 , T
T 4
Total radiant energy

for = 0 to =
64
32
Applications of Plancks Law: example

Q: What fraction of the total power radiated by a black body at 5770 K fall, in
the UV (0 < 0.38m)?
o Need table of integral values of F0
o So, T = 0.38m * 5770K = 2193mK l T (mmK x10 3) F 0 l (l T)
(dimensionless)
o Or 2.193x103 mK i.e. between 2 and 3 2 .067
3 .273
o Interpolate between F0 (2x103) and F0 (3x103) 4 .481
5 .634

F00.38 , T F00.38 2 x103

2.193 2
0.193
6
8
.738
.856

F00.38 3x10 F00.38 2 x10
3

3
32
10
12
.914
.945
14 .963
F00.38 , T 0.067
16 .974
18 .981
0.193 20 .986
0.273 0.067
Finally, F00.38 =0.193*(0.273-0.067)+0.067=0.11

i.e. ~11% of total solar energy lies in UV between 0 and 0.38 m
65
Problem
Show that ~38% of total energy radiated
by the sun lies in the visible region
(0.38m < 0.7m) assuming that
T (mK x103) F0(T)
solar T = 5770K (dimensionless)
2 .067
3 .273
o Hint: We already know F(0.38m), so 4 .481
calculate F(0.7m) and interpolate 5
6
.634
.738
8 .856
10 .914
12 .945
14 .963
16 .974
18 .981
20 .986
66
33
Deviation from blackbody assumption
67
Recap
Objects can be approximated as blackbodies
o Radiant energy T4
EM spectrum from sun a continuum peaking at ~0.48m
o ~39% energy between 0.38 and 0.7 in visible region
Plancks Law - shape of power spectrum for given T (Wm-2 m-1)

o Integrate over all to get total radiant power emitted by blackbody per
unit area
Stefan-Boltzmann Law M = T4 (Wm-2)
o Differentiate to get Wiens law
Location of max = k/T where k 2898mK
68
34
Radio communication
Radio communication
system
Invention of radio waves

Hertz proved the existence of the
electromagnetic waves in Karlsruhe Institute
of Technology
Heinrich Hertz: I do not think that the wireless

waves...will have any practical application 70
35
71
Radio engineering
Radio engineering is a technical science that deals with
the electromagnetic waves at the frequencies below the
infrared frequencies (e.g. from 30 Hz up to 3 THz)
Word radio comes from Latin: radius and/or radiare
(= to radiate)
An oscillator generates high-frequency alternating
current which is radiated as radio waves by the
antennas
Millimeter wave frequencies become more important
all the time.
lack of frequencies, higher data rate, improved technology,
new applications
Note: Radio spectrum is only a tiny fraction of electromagnetic spectrum

near dc or zero frequency. 72
36
73
Radio transmission
(a) In the VLF, LF, and MF bands, radio waves follow the
curvature of the earth.
(b) In the HF band, they bounce off the ionosphere.
Microwave transmission uses frequencies above

VHF
74
37
Radio communication system

Transmitters
Frequency stability, power
Antennas
Gain, beam width, sidelobe
level, bandwidth, noise
temperature
Radio channel
Loss, noise, fading, etc.
Receiver
Selectivity, noise
Modulation
Analog, digital
75
Transmitting system
76
38
Transmitter
Stable transmit frequency: stability 10-9 - 10-10 / day
Phase-locked loop (PLL) used to lock the frequency
Reference by using a stable quartz crystal resonator
oscillator
Modern fast way: direct digital synthesis (DDS) => waveform
created with digital-to-analog converter (DAC)
Simple direct conversion transmitter with PLL
oscillator
77
Transmitter
Low or medium power transmitters use semiconductors
Solid-state transmitters can now provide tens of kW for TV
bands
The highest power transmitters use electron tubes
Such as magnetrons in radars because high peak power is
needed
International broadcasting stations at low frequencies could
transmit even 1 MW but modern radio communication
mostly uses much less transmit power
Portable devices work typically with powers up to 1 - 2 W; short
range devices even with much less power
Transmit power is controlled as needed and best for service,
without excessive power (e.g. cell phone systems)
78
39
Receiver
Good filters, an accurate LO frequency, low-noise components
LO can be realized similarly as in transmitters
Superheterodyne receiver uses an intermediate frequency
stage - or two at different frequencies
Received signal level may vary a lot => automatic gain
control
Demodulator extracts information from the IF signal
Typical simple radio communication receiver
79
Receiver specifications
Several parameters determine the ability of a receiver to
successfully demodulate a radio signal.
Receiver sensitivity is the minimum input voltage that
produces a specified signal-to-noise ratio at the output of
the IF section.
Dynamic range is the difference between the largest input
signal that will not become distorted and the smallest signal
that can be discerned, and is measured in dB.
Selectivity specifies a receivers ability to discriminate
against adjacent channel signals.
It is a function of the IF strips BPF.
Noise figure indicates how much the receiver degrades the
input signals signal-to-noise ratio.
80
40
Superheterodyne receiver
81
RF and IF input spectrum
82
41
Double-conversion receiver
83
Direct-conversion receiver
Frequency conversion direct to baseband for data
demodulation
Attractive for realization in integrated circuits (IC)
Characteristics more limited than in complex receiver
designs
84
42
Software radio
Progressively faster digital technology allows also
the use of software defined radio
Typical radio device components used only as antenna
and in receiver front-end and then ADC, followed by
digital processing with software
Digital technology and software-defined
operation is used also for transmitters
After software controlled parts and DAC follow radio
devices: up-converter, power amplifier, filter and
antenna
85
Software defined radio

A software defined radio (SDR) is one that can be configured to any
radio or frequency standard through the use of software.
A software defined radio would be able to work on different broadband
networks and would be able to transfer to another network seamlessly
while traveling outside of the users home network.
A software defined radios best advantage is its great flexibility to be
programmed for emerging wireless standards.
It can be dynamically updated with new software without any changes
in hardware and infrastructure.
Software could just download interface automatically.
This is the job of the packet layer, which will split the data into small
packets.
86
43
Software Defined Radio (SDR)

A conceptual definition of the software defined radio (SDR) is
presented in the following figure
87
Transceiver
Much of radio communication needs a two-way device to
both transmit and receive => transceiver combines these
Basic transceiver device problem: high power TX & low
power RX
How to get necessary isolation when connecting to
common antenna?
If a frequency division duplexing (FDD) system is used => a
duplexer filter separates TX and RX bands (which are different)
If a time division duplexing (TDD) system is used => separate
TX and RX times allow simply a switch to be used only
transmitter or receiver is connected to antenna at any time, not
both simultaneously
For example, cell phone network devices may use both systems
88
44
Signals and waveforms

Communication systems transmit information in the form of
electrical signals that represent
speech,
music,
television pictures,
data, and so forth.
The waveforms of these signals are complex and continually
changing, but the frequency spectrum of the signals is usually
limited to a specified bandwidth either by the nature of the
signal source or by filters in the transmitting equipment.
Since many of these signals occupy a frequency band that
extends downward to a few hertz (bandbase signals), they
cannot be transmitted in their original form over a common
transmission path because it would not be possible to separate
them at the receiving end.
89
Multiplexing
A separate transmission line or separate radio
path for each signal would not be feasible from
either an economic or a practical standpoint.
Hence the overall communication system must
provide a means for simultaneous transmission
of a number of signals either by shifting them
into different parts of the frequency spectrum
or by sending samples of the signals on a time-
shared basis.
90
45
Need for higher frequencies

The wavelength () in meters of a radio wave is given
by c/f, in which c is the velocity of light (3 x 108
meters per second in atmosphere), and f is in hertz.
A radio antenna should have a physical size of one-
half wavelength or more for reasonable efficiency.
Hence, as the transmission frequency is increased, the
physical size and cost of the antenna are reduced and
its efficiency increases.
91
Elements of a radio system
A simplified block
diagram of a radio
transmitter and
receiver illustrates
the signal
processing that
takes place in a
radio system.
92
46
1. Message signal source
The source of the message signal may be

a microphone,
phono pickup,
television camera, or
other device
that transforms the desire information into an
electrical signal.
93
2. Amplifier and 3. Oscillator

The signal is amplified and often passed
through a low-pass filter to limit the
bandwidth.
The RF oscillator establishes the carrier
frequency or some submultiple of it.
Since good frequency stability is required to
keep the transmitter on its assigned frequency,
the oscillator is often controlled by a quartz
crystal.
94
47
4. RF amplifier and frequency

multiplier
One or more amplifier stages increase the
power level of the signal from the oscillator to
that needed for input to the modulator.
Using frequency multiplication the final carrier
frequency can be a multiple of the oscillator
frequency.
95
5. Modulator
The modulator combines the signal and carrier
frequency components to produce one of the varieties
of modulated waves (AM, ...)
In our simplified system the output signal spectrum
lies in the vicinity of the desired RF carrier frequency.
In many transmitters a second oscillator and mixer
(similar to blocks 10 and 11) are inserted between
blocks 5 and 6 in order to shift the modulated wave to
a higher frequency range.
96
Note: The carrier may be a sinucoidal wave or a train of pulses.
48
6. RF amplifier and 7. Transmitting

antenna
Additional amplification may be required after
modulation to bring the power level of the signal to
the desired value for input to the antenna.
The transmitting antenna converts the RF energy into
an electromagnetic wave of the desired polarization.
If a single (fixed) receiver is to be reached, the
antenna is designed to direct as much of the radiated
energy as possible toward the receiving antenna.
For broadcasting service suitable transmitting antenna
is omnidirectional.
97
8. Receiving antenna
The receiving antenna may be omnidirectional for
general service or highly directional for point-to point
communication.
The wave propagated from the transmitter induces a
small voltage in the receiving antenna.
The range of amplitudes of the induced antenna
voltage may be from tens of millivolts to less than 1
microvolt, depending upon a wide variety of
conditions.
98
49
9. RF amplifier
The RF amplifier stage increases the signal power to
a level suitable for input to the mixer and it helps to
isolate the local oscillator from the antenna.
This stage does not have a high degree of frequency
selectivity but does serve to reject signals at
frequencies far from the desired channel.
The increase in signal power level prior to mixing is
desirable because of the noise that is inevitably
introduced in the mixer stage.
99
10. Local oscillator and 11. Mixer

The local oscillator in the receiver is tuned to produce
a frequency fLO that differs from the incoming signal
frequency fRF by the intermediate frequency fIF; that
is, fLO can be equal to fRF + fIF or fRF - fIF .
The mixer is a nonlinear device that shifts the
received signal at fRF to the intermediate frequency
fIF.
Modulation on the received carrier is also
transformed to the intermediate frequency.
100
50
12. IF amplifier
The IF amplifier increases the signal to a level
suitable for detection and provides most of the
frequency selectivity necessary to pass the desired
signal and filter out the undesired signals that are
found in the mixer output.
Because the tuned circuits in blocks 11 and 12 always
operate at a fixed frequency (fIF), they can be
designed to provide good selectivity.
Ceramic or crystal filters are often used.
101
13.-15. Detector, Audio/video

amplifier and Output device
The detector recovers the original message signal
from the modulated IF input.
The audio or video amplifier increases the power
level of the detector output to a value suitable for
driving a loudspeaker or other output device.
The output device converts the signal information
back to its original form (sound waves, picture, etc.).
102
51
Radio transmitter and receiver block

diagram
Note: In practice, there
are so many variations
in transmitter and
receiver systems that no
single block diagram
could even be
considered typical.
103
Note: Electrical noise

In addition to the desired signal that is
processed by the receiver, electrical noise is
added in the transmission path, and is
generated within
the RF amplifier,
local oscillator,
mixer, and so forth.
104
52
Radio communication
Radio transmission
Power gain
Let it be an linear time-invariant (LTI) system whose
input signal has average power Pin.
If the system is distortionless, the average signal

power at the output will be proportional to Pin.
Thus, the systems power gain is
106
53
Power gain in decibels
Systems that include amplification may have

very large values of g, so well find it
convenient to express power gain in decibels
(dB) defined as
107
Power gain in decibels

Since the decibel is a logarithmic unit, it converts
powers of 10 to products of 10.
For instance, g = 10e becomes gdB = e x 10 dB.
Power gain is always positive, of course, but
negative dB values occur when g < 1.0 = 100 and
hence gdB < 0 dB.
Note that 0 dB corresponds to unity gain (g = 1).
Given a value in dB, the antilogarithmic value is
108
54
Absolute units: dBW and dBm
While decibels represent power ratios, signal

power itself may be expressed in dB if you
divide P by one watt or one milliwatt, as
follows:
109
Power gain in dB-units

Rewriting the definition equation of power
gain as (Pout/1 mW) = g(Pin/1 mW) and taking
the logarithm of both sides then yields the dB
equation
110
55
Transfer function
Now consider a system described by its transfer
function H(f).
A sinusoidal input with amplitude Ax produces the
output amplitude Py
Ay = |H(f)|Ax,
and the normalized signal powers are
Px= Ax 2/2 and Py = Ay2/2= |H[f)|2Px.
If the system has the same impedance level at input
and output
Py/ Px = Pout/ Pin = |H[f)|2=g
111
Square of amplitude function

|H[f)|2
In any case |H[f)|2 tells us how the power gain
varies as a function of frequency.
For a useful measure of frequency dependence
in terms of signal power we take
This represents the relative gain in dB.
112
56
Problems
(a) Verify that dBm = dBW + 30 dB.

(b) Show that if |H(f)|dB = -3 dB then
|H(f)|1/2.
113
Solution
(a) Verify that dBm = dBW + 30 dB.
114
57
Solution
(b) Show that if |H(f)|dB = -3 dB then

|H(f)|1/2.
115
Transmission loss
In addition to any signal distortion, a
transmission system also reduces the power
level or strength of the output signal.
This signal-strength reduction is expressed in
terms of transmission power loss.
Although transmission loss can be
compensated by power amplification, the ever-
present electrical noise may prevent successful
signal recovery in the case of large
transmission loss.
116
58
Transmission loss
Any passive transmission medium has power

loss rather than gain, since Pout < Pin
We therefore prefer to work with the
transmission loss or attenuation
Hence
117
Attenuation coefficient
In the case of transmission lines, coaxial cables,
and waveguides, the output power decreases
exponentially with distance.
Well write this relation in the form
where l is the path length between source and

destination and is the attenuation coefficient in
dB per unit length.
Then
118
59
Example
A 40-km cable system has Pin 2 W and a

repeater with 64-dB gain is inserted 24 km
from the input.
The cable sections have = 2.5dB/km.
Use dB equations to find the signal power at:
(a) the repeaters input;
(b) the final output.
119
Solution
Use dB equations to find the signal power at:
(a) the repeaters input;
(b) the final output.
120
60
Radio transmission
Signal transmission by radio wave propagation can
reduce the required number of cable repeaters, and has
the additional advantage of eliminating long cables.
Here we examine the transmission loss for line-of-
sight (LOS) propagation (illustrated in the following
Fig.) where the radio wave travels a direct path from
transmitting to receiving antenna.
This propagation mode is commonly employed for
long-distance communication at frequencies above
100 MHz.
121
Line-of-sight (LOS) radio transmission
122
61
Free-space loss
The free-space loss on a line-of-sight path is due to
spherical dispersion of the radio wave.
This loss is given by
in which is the wavelength, the signal frequency, l

the distance (path length), and c the speed of light.
If we express l in kilometers and f in gigahertz (109 Hz)
then
123
Free-space loss(dB) log(path

length)
We see that LdB increases as the logarithm of the path
length.
Thus, for instance, doubling the path length increases
the loss by only 6 dB. (Prove this as homework!)
We also see that LdB increases as the logarithm of the
frequency.
Thus, for instance, doubling the frequency increases
the loss by 6 dB.
Note: Here LdB is the free-space line-of-sight attenuation and doubling the path length
can violate this free-space line-of-sight assumption. 124
62
Antenna gain
Furthermore, directional antennas have a

focusing effect that acts like amplification in
the sense that
where gT and gR represent the antenna gains at

the transmitter and receiver.
125
Effective aperture area

The maximum transmitting or receiving gain of an antenna
with effective aperture area Ae is
where c 3 x 105 km/s.

The value of Ae for a horn or dish antenna approximately
equals its physical area.
More accurately, the value of Ae for a circular dish antenna
equals its physical area (DA/2)2 multiplied by the aperture
efficiency 0.5 0.70.
Large parabolic dishes may provide gains in excess of 60 dB.
126
63
Example
Satellite relay system shows a transoceanic radio system with a satellite
relay serving as a repeater.
The satellite is about 40000 km from either ground station, and the signal
frequency is 6 GHz.
The satellite has a repeater amplifier with gamp = 80 dB, and the input
power at the ground station is Pin = 100 W .
127
Solution
Free space loss equation gives the uplink and
downlink path loss
Since we have all the data in dB, we can

compute Pout, by adding gains and subtracting
losses in the following tabulation:
128
64
Solution
Note: Very small power levels are typical for satellite systems. 129
Ground reflection
Often, there are multiple
signals arriving to the
receiver due to the
reflections from the
ground, water, building
etc.
Multi-path propagation
For the horizontal polarization the
Very simple case: direct reflection coefficient is -1.
Expression r1+r2-r0 2h1h2/d
wave and one reflected
wave. =2/
130
65
Destructive reflection case

In certain situation, the phase difference between the
direct and reflected wave is 180 degrees and thus the
waves cancel each other out in the receiver.
The total electric field strength is 0.
Condition for that is
Obviously, there is also possible that both waves are

summed together constructively as is the case in the
next example.
131
Example
There is a lake between a transmitter and a receiver (distance
d).
The height of the transmitter antenna is H (H<<d).
Calculate the optimum height of the receiver antenna when the
frequency is f.
The polarization is horizontal and the lake is an ideal reflector.
132
66
Solution
There is a lake between a transmitter and a receiver (distance d).
The height of the transmitter antenna is H (H<<d).
Calculate the optimum height of the receiver antenna when the frequency is
f.
The polarization is horizontal and the lake is an ideal reflector.
The total electric field at the antenna site is
=2/
The maximum field strength is obtained when
Obviously, the smallest height (n=0) is the optimum height.

So
133
Numerical version
There is a lake between a television transmitter
and a receiver (distance 5 km).
The height of the transmitter antenna is 50 m.
Calculate the optimum height of the receiver
antenna when the frequency is 500 MHz.
The polarization is horizontal and the lake is
an ideal reflector.
The optimum height is
h=15 m
134
67
Multi-path propagation
The waves reflect and refract from obstacles and
they arrive to the receiver along different paths.
In the antenna the waves interfere and thus in the
receiver only one signal appear.
The phase difference between the arriving signals affect the
total received signal
When the receiver moves, (e.g. in mobile
communication), the received signal changes
significantly as a function of the distance.
Fast signal fading takes place within the distance of one
wavelength.
135
Antenna parameters
Radiation pattern (Directional pattern )
The directional dependence
of the power density radiated
or received by the antenna
Directivity D
136
68
Antenna parameters
Radiation efficiency rad
Gain G
Effective area Aeff
Area efficiency coefficient for aperture antennas
137
Antenna parameters
Input impedance Zant
Zant(f) = Rrad(f)+ Rloss(f) + j X(f)
When the input impedance Zant of the antenna
differs from the characteristic impedance of the
feed line Z0, part of the power is reflected
Polarization
Is defined as the direction of radiated/received
electric field vector of the antenna.
138
69
Low-frequency antennas
f< SHF
Wire antennas, microstrip antennas
Solid bow tie

139
High-frequency antennas
0.1 mm (3 THz)
Reflectors, horns, dielectric antennas
140
70
Friis formula
Received power
Note:
141
Example: Link budget

In a 58-GHz link radio, the gain of antennas is
34 dBi, the transmitter power is 5 dBm and the
sensitivity of the receiver is -73 dBm. What is
the largest distance that can be implemented
with this link radio?
142
71
Solution
The gain of the antennas
The transmitted and

received powers
The wavelength
The largest distance r can be

solved from Friis formula
143
Non-idealities
It is impossible to implement this theoretical distance in
practice, because there are many non-idealities, so the distance
is significantly shorter due to the following factors:
The attenuation in the air (moisture, rain)
For instance, at 58 GHz the attenuation of the air is significant due to
the resonance mode of oxygen molecules at 60 GHz
Obstacles (buildings, forest, hilly terrain etc.) between the
antennas
Reflections from the ground and other obstacles
Polarization mismatching
Impedance mismatching
Pointing error of the antennas
144
72
Propagation mechanisms of radio

waves
Line-of-sight propagation
Ionospheric propagation
Ground propagation (surface wave)
Reflection (refraction, diffraction)
Atmospheric scattering
145
Propagation mechanisms of radio

waves
Line-of-sight propagation
The most common and
important mechanism
Scattering from the in-
homogeneity of the
atmosphere
communication far beyond
the radio horizon
Reflection from the
ionosphere
Ground wave
146
73
1st Fresnel zone
147
Problem
There is a 45-km radio link from a mainland to
an island. The operation frequency is 6 GHz.
What is the required heights of the link
antennas when the first Fresnel ellipsoid is free
of obstacles?
148
74
Problem
r0=45000m, f=6GHz, K = 4/3, R=6375km
149
Solution
Radius of the 1st Fesnel zone
Radius in middle ( r 1=r2r0/2,

because hF<<r0 ) is
Applying Pythagoras to Fig.

and solving the equation
we get the antenna heights
150
75
Inhomogeneity in the atmosphere cause scattering
We can think that scatterer operates as a isotropic antenna
which radiates power uniformly in all directions, taking its
power from passing by EM wave.
Scattering is always present in the atmosphere but its

attenuation effect for line-of-sight communication
is infinitesimally small in clear atmosphere and minor
in heavy rain up to 10 GHz.
151
Sometimes this weak
scattering effect can be
exploited.
In that case the radio wave
propagates behind an
obstacle or far beyond the
horizon where it would not
propagate otherwise.
In weather radars, the
scattering signals from
hydrometeors (raindrops,
fog, hails or snow flakes) are
measured. 152
76
Summary
A radio communication system is an economic compromise
between information transfer capacity needs, available
frequency bands, propagation and equipment characteristics
(propagation attenuation, size of the antenna ).
Low frequencies (long wavelength)

Small propagation attenuation
Large-sized antennas
Low data rate (a small number of different frequency bands)
High frequencies (short wavelength)
Higher propagation attenuation (and shadow areas caused by terrain)
Small-sized antennas
High data rate
153
Examples
FM radio waves propagate very well because the frequency is
small (about 100 MHz), but the required antenna is large-
sized.
GSM 900 (900 MHz) operate well in rural areas where the
waves can propagate far
GSM 1800 (1800 MHz) operate better in urban areas where
the base stations are close (cell size is small) and a number of
frequency bands can be larger than at 900 MHz.
154
77
Radio communication
Fundamental limitations
Fundamental limitations
There are two general kinds of constraints when designing a
communication system.
On the one hand are the technological problems, including
hardware availability,
economic factors,
federal regulations, and so on.
These are problems can be solved in theory, even though perfect
solutions may not be practical.
On the other hand are the fundamental physical limitations dictating
by the laws of nature.
These limitations ultimately dictate what can or cannot be
accomplished, irrespective of the technological problems.
The fundamental limitations of information transmission by
electrical means are bandwidth and noise.
156
78
Bandwidth (B)
The concept of bandwidth applies to both signals and
systems as a measure of speed.
When a signal changes rapidly with time, its frequency
content, or spectrum, extends over a wide range and
we say that the signal has a large bandwidth.
Similarly, the ability of a system to follow signal
variations is reflected in its usable frequency response
or transmission bandwidth.
All electrical systems contain energy-storage elements,
and stored energy cannot be changed instantaneously.
Consequently, every communication system has a finite
bandwidth B that limits the rate of signal variations.
157
Transmission bandwidth (BT)

Communication under real-time conditions requires
sufficient transmission bandwidth to accommodate the
signal spectrum.
Otherwise, severe distortion will result.
Thus, for example, a bandwidth of several megahertz is
needed for a video signal, while the much slower variations
of a voice signal fit into BT 3 kHz.
For a digital signal with r symbols per second, the
bandwidth must be BT r/2.
In the case of information transmission without a real-time
constraint, the available transmission bandwidth determines
the maximum signal speed.
The time required to transmit a given amount of information
is therefore proportional to 1/BT.
158
79
Noise
Noise imposes a second limitation on information
transmission.
Why is noise unavoidable?
According to kinetic theory at any temperature above
absolute zero, thermal energy causes microscopic
particles to exhibit random motion.
In electrical communication systems the random
motion of charged particles such as electrons generates
random currents or voltages called thermal noise.
There are also other types of noise, but thermal noise
appears in every communication system.
159
SNR
We measure noise relative to an information signal in
terms of the signal-to-noise power ratio (SNR or S/N).
Thermal noise power is ordinarily quite small, and S/N
can be so large that the noise goes unnoticed.
At lower values of S/N, however, noise degrades
fidelity in analog communication and produces bit
errors in digital communication.
These problems become most severe on long-distance
links when the transmission loss reduces the received
signal power down to the noise level.
Amplification at the receiver is then to no use, because
the noise will be amplified along with the signal.
160
80
Hartley-Shannon law
Taking both limitations into account, Shannon
(1948) stated that the rate of information
transmission cannot exceed the channel capacity
C=Blog2(1 +S/N)
with the zero-error probability (pe0).
This relationship, known as the Hartley-Shannon
law, sets an upper limit on the (error-free)
performance of a communication system with a
given bandwidth and signal-to-noise ratio.
161
Ideal communication systems

Every realistic communication system is subject to some
power and bandwidth limitations.
If we exclude radio transmission with fading, careful
system design can largely eliminate all other
contaminations except unavoidable thermal noise.
The AWGN (Additive White Gaussian Noise) channel
serves as a reasonable model under these conditions, and
the Hartley-Shannon law gives the maximum rate for
reliable communication.
Therefore, we here define
An ideal communication system is one that achieve
nearly error-free information transmission at a rate
approaching R=Blog(1+S/N)
162
81
Electrical Noise
Introduction
Noise and interference

Unwanted electric signals come from a variety of sources,
generally classified as human interference or naturally
occurring noise.
Human interference comes from other communication
systems, ignition and commutator sparking, 50 or 60 cycle
hum, and so forth; natural noise-producing phenomena
include atmospheric disturbances, extraterrestrial radiation,
and circuit noise.
By careful engineering, the effects of many unwanted
signals can be reduced or eliminated completely.
But there always remain certain inescapable random
signals, that present a fundamental limit to systems
performance.
164
82
Noise is always here!

Noise is always present in communications systems, but under
normal operating conditions it goes unnoticed because the
signal levels are much higher than the noise levels.
A weak signal accompanied by noise can be amplified if the
associated noise level is low relative to the signal level.
However, if the noise level is close to the signal level,
amplification will be useless because any amplifier will
amplify both the incoming signal and the incoming noise as
well as adding more noise of its own.
This process is evident in any receiver when an incoming
signal fades into the noise or the external noise level rises to
the point that it drowns out the signal.
165
Noise limits performance

Transmission loss usually results in a very weak
signal at the input of a communication receiver.
Consequently, to obtain an adequate signal level
for further processing, the front end of a typical
receiver includes several stages of amplification.
High-gain amplifiers amplify any noise that
accompanies the received signal, and they also
add their own internally generated noise.
An accurate assessment of system performance
must therefore take account of the amplifier noise.
166
83
Circuit and device noise
Noise generated within an electrical circuit

may come from numerous sources involving
several different physical phenomena.
Here well describe major types of device
noise found in communication systems, and
well develop appropriate circuit models and
analysis methods.
167
Signal and noise

To simplify the mathematical details, the
desired signal is assumed to be a sinusoid or a
group of sinusoids that comprise the
transmitted information.
This may be called a deterministic signal.
Noise is defined as any extraneous electrical
disturbance tending to interfere with the
normal reception of the transmitted signal.
168
84
Random noise
Noise can consist of signals from unwanted
sources plus random fluctuations of voltages
and currents caused by physical phenomena.
There are various types of random noise
thermal noise,
shot noise, and
flicker noise

169
Signal power/Noise power

One of the goals of communications system design is to keep the ratio of
average (or peak) signal power to average noise power so large that the
noise has no harmful effects on system performance.
Techniques for doing this include
(1) using powerful transmitters and high-gain antennas to develop
strong signals at the receiver,
(2) designing amplifier and mixer circuits so that they introduce a
minimum amount of additional noise when processing signals, and
(3) using modulation or coding schemes that facilitate the separation of
signaIs from noise.
In the case of man made noise sources (e.g., automobile ignitions) there
exists the fourth option of suppressing the noise at its source by filtering,
bypassing, or redesign.
Attention is usually given to each of these four options, and the mix
selected is determined by such factors as cost, weight, and efficiency.
170
85
Stationarity
The assumption of stationarity is reasonable

for most electrical noise processes, thereby
allowing us to represent noise sources in the
frequency domain.
For convenience, well adopt the common
practice of working entirely with positive
frequency and one-sided frequency functions
as distinguished from the two-sided functions.
171
Example: 2-sided 1-sided spectrum

function
To clarify this distinction, let G(f) be the two-sided available
power spectrum of some noise source.
Since G(f) has even symmetry, the corresponding one-sided
available power density will be defined by
p f pa f 2G f , f 0
and the total available noise power is

N pa f df
0
Likewise, the one-sided mean square voltage and current densities are equal twice the
172
two-sided frequency functions for f0.
86
Stochastic nature of noise

It is desirable to characterize electrical noise as accurately as possible.
A common characteristic of most types of noise, however, is its nondeterministic
nature: that is,
the exact waveform of the noise cannot be predicted.
A measure of the amount of electrical noise can be obtained by connecting a meter
across a noise source to measure the
average,
peak,
rectified-average, or
rms voltage (or current).
Relationships between these quantities are different for different types of noise, that
is, the average value may be zero whereas the others are not.
The rms voltage (or current) can be used to calculate the average noise power
delivered to a resistive load.
As will be shown, the measured value depends upon the spectrum of the noise
source and the frequency response of the measuring instrument.
173
Power spectral density

The frequency-domain characterization of noise can
be given by means of a power spectral density curve
[the units are watts per hertz (W/Hz)].
The simplest kind of noise to work with would be one
whose spectral density is flat over the frequency
range of interest.
A common example of noise with a flat frequency
spectrum is thermal noise associated with Brownian
motion of elections in a conductor.
174
87
Thermal noise
One unavoidable cause of electrical noise is

the thermal motion of electrons in conducting
mediawires, resistors, etc.
As long as communication systems are
constructed from such material, this thermal
noise will be with us.
175
Thermal noise
Available noise power of thermal noise in a 1-Hz bandwidth is
given by
p(f) = kT,
where k is Boltzmanns constant and T is the temperature of
the noise source in kelvin units.
This expression applies from direct current (dc) to high
microwave frequencies.
If the bandwidth were allowed to extend to infinity, the
available power of a thermal noise would appear to be
unlimited also.
However, results from quantum mechanics indicate that
physical thermal noise sources, although extremely wide band,
have a power spectrum that drops to zero at arbitrarily high
frequencies.
176
88
Equivalent thermal noise source

Thermal noise is both the most common type of noise
encountered in radio circuits and the easiest to
analyze.
Other types of noise will be represented by equivalent
thermal noise sources.
Although the common emphasis will be on those
aspects of noise necessary to the design of
communications system that operate under average
conditions, much more can be done with the theory of
noise in terms of developing probability distributions
for system outage time, and etc.
177
Thermal noise in resistors and

networks
As the name implies, thermal noise is due to the random
motion of charge carriers in any conducting medium whose
temperature is above absolute zero.
The velocity of this motion increases with temperature in such
a way that the electrical noise power density produced is
proportional to the resistance of the conductor and to its
absolute temperature, hence the name thermal noise.
It is also called white noise because it has been shown both
theoretically and experimentally to have a uniform spectrum
up to frequencies on the order of 1013 (just as white light is
composed of all colors of the visible spectrum).
178
89
Noise equivalent circuit

A metallic resistor may be considered a thermal noise
source that can be presented by either of the noise
equivalent circuits shown below.
179
The mean-square values

The mean-square noise
voltage (Vn2) and current
(In2) are given by the
following expressions in
which R is the resistance, G
= 1/ R the conductance, T
the temperature of the
resistor in kelvin units, k
Boltzmanns constant
(1.38*10-23J/K), and B the
bandwidth in hertz in which
the noise is observed.
180
90
PN B
The noise power that is transmitted through a
circuit is proportional to the circuit bandwidth.
Consequently, the circuit bandwidth should
never be greater than that necessary to transmit
the desired signal if the maximum output
signal-to-noise ratio (SNR) is to be achieved.
181
Example
Calculate the mean-square noise voltage
produced in a 100-k resistor in a bandwidth
of 106 Hz at room temperature (T=20C = 293
K).
Vn2=4kTRB
Note: The rms voltage is (Vn2)40.3 V 182
91
Comment
If this 100-k resistor were in the input circuit

of an electronic voltmeter that had a bandwidth
of 1 MHz, no amount of gain built into the
voltmeter would enable it to measure signals
below 1 millivolt (mV) with accuracy.
183
Frequency domain models of a noise

source
Thvenin circuit Norton circuit
184
92
Thvenin equivalent circuit

The previous Figure shows the frequency-domain Thvenin
circuit model of a stationary but otherwise arbitrary noise
source with noiseless internal impedance Z(f).
The function vn2(f) represents the open-circuit mean square
voltage density, defined such that
Here p(f) [=pa(f)] is one-sided (f0) available power density
185
Norton equivalent circuit

Converting Thevenin source to a Norton
equivalent circuit gives the source model with
which represents the short-circuit mean square

current density.
This equation also expresses Ohms law in the
form needed for circuit noise analysis.
186
93
Example
Suppose the source in question happens to be a
thermal resistance R at temperature T.
Then the one-sided available power density
p f kT
and setting Z(f) = R yields
4kT
vn2 f 4 RkT and in2 f
R
These constant densities correspond to white
noise, at least up to infrared frequencies.
187
Thevenin equivalent circuit for noise

calculation
A resistive network driven by a voltage source
Circuits containing more
than one resistor may be
analyzed by reducing
them to one (Thvenin)
equivalent resistance
and applying
Thevenin equivalent circuit for noise computation
Vn2=4kTRTB , RTRTh
to obtain the mean-

square noise voltage..
188
94
Note
The noise Thvenin equivalent is different from the signal Thevenin equivalent
of the same circuit, where a signal source is driving a hypothetical noiseless
load resistor Ri (representing the input of an amplifier) through three noisy
resistors R1, R2, and R3.
Thevenin equivalent circuit for signal computation
The signal voltage at the load is found by the conventional voltage division, but
the noise voltage is found from a circuit in which the noise source is related to
the Thevenin resistance of the resistive network rather than the voltage divider
ratio.
189
Note
The process of combining the resistors in a network to obtain
an effective noise resistance is equivalent to combining the
mean-square noise voltages of the resistors themselves to
obtain the net mean-square noise voltage.
This is consistent with the statistical principle that if two or
more independent random processes are combined, the mean-
square value of the resultant is obtained by adding the mean-
square value of each process.
Thus the mean-square noise voltage of a group of resistors
connected in series is the sum of the mean-square noise
voltage across each resistor.
190
95
Effect of reactive elements

Resistance noise occurs in almost all circuits, but
reactive elements may alter the frequency density.
In particular, let Figure be a one-port (two-terminal)
network containing only resistance, capacitance, and
inductance, and having the equivalent impedance Z(f)
= R(f) + jX(f).
191
Nyquists formula
When the resistances are in thermal equilibrium at
temperature T, Nyquists formula states that
vn2 f 4 R f kT
Hence, the mean square voltage density takes the
shape of the equivalent resistance R(f).
Nyquists equation includes the special case of an
all-resistive network whose equivalent resistance
will be independent of f.
192
96
Nyquists formula proof
We prove Nyquists formula by considering

the complete circuit in Fig. below, where the
source has been connected to a load resistance
RL at the same temperature T.
193
Nyquists formula proof

This load resistance, of course, generates
thermal noise with vL2(f) = 4RLkT.
Since the two noise sources are physically
independent, superposition applies and we can
calculate the mean square current densities in
each direction, namely,
194
97
Proof of Nyquists formula

The average power delivered from the source to the load in
any frequency band df is RL iL2(f) df, while the load delivers
R(f)iZ2(f) df back to the source.
Now the net power transfer must be zero at every frequency
for the circuit to be in thermal equilibrium.
Therefore,
and Nyquists formula follows after cancellation.
Note: Nyquists formula does not hold when the resistances are at different
temperatures or the network contains non-thermal sources. However, such cases are
easily analyzed provided that the sources are independent. Using superposition and
summing mean square values the resulting frequency density can be found. 195
Thermal noise by Nyquist

Nyquist has determined the thermal noise output of a
network containing both resistive and reactive
elements.
At a port in such a circuit, the mean-square thermal
noise voltage is given by
in which R(f) is the resistive (real) part of the input

impedance at frequency f and the integration is performed
over the bandwidth (B) of interest.
196
98
Example
The thermal noise generated in the

conductance of a parallel GC circuit is
represented by a current source with
In2(f )=4kTGf
over any small frequency increment f.
197
Example
The admittance of the circuit is
Y G j C
The mean-square voltage at the port is
I n2( f )
V 2
n ( f ) 2
Y
198
99
Example
By integration over infinite bandwidth the total

mean-square noise voltage is obtained by
4kTG
Vn2 df
0 G (2fC ) 2
2
The resistance at the port is
G
R f
G (2fC ) 2
2
199
Example
To complete the calculation of Vn2, it is
convenient to change the variable of
integration to and after some manipulations

kT
Vn2
C
Note: Suprisingly, Vn2depends on C but not on R (or G), even though the noise source is
actually thermal resistance! The explanation of this paradoxical independence from R is
included to the equivalent noise bandwidth (B1/R) of GC-filter. 200
100
Example
Substitution of the half-power frequency of RC

circuit,
G
3dB 2f 3dB
C
into the previous integral yields after

integration the result

Vn2 4kTR f 3dB
2
201
Example
By the comparison of
Vn2 4kTRB
with the previous result,

Vn2 4kTR f 3dB
2
the equivalent noise bandwidth is seen to be
equal to /2 times the half-power bandwidth.
202
101
Problem
Verify that the change in variable from f to

in
4kTG
V
n
2
df
0 G 2 (2fC ) 2
yields

4kT 1
2G 0
V
n
2
2
d
C
1
G 203
Problem
Then introduce the substitution given by
G
3dB 2f 3dB
C
and change the variable of integration to
x = /3dB.
Show that the value of the definite integral that
results yields
kT
Vn2 4kTR f 3 dB
2 C 204
102
Noise in receiving antennas

The resistance measured at the terminals of an antenna is on
the order of 70 ohms (half-wave dipole) to 300 ohms (folded
dipole).
This resistance value is primarily the radiation resistance,
which accounts for power that is radiated from the antenna.
The ohmic resistance contributed by the resistance of the
antenna conductors is usually negligible in comparison with
the radiation resistance.
Note: An exception occurs in low-frequency antennas in which the dlmensions are a

small fraction of a wavelength. Here the ohmic resistance may predominate,
205
Antenna noise sources

A receiving antenna exhibits noise at its
terminals from two sources:
(1) the thermal noise generated in its ohmic
resistance (usually negligible), and
(2) the noise received from external sources.
(Any body with temperature greater than 0 K
radiates noise energy.)
206
103
Antenna noise temperature

The received noise is represented as though it were
thermal noise generated in a fictitious resistance equal
to the radiation resistance, at a temperature TA that
would account for the noise actually measured.
This is called the noise temperature of the antenna.
207
Example
Suppose that a 200-ohm antenna exhibits an
rms noise voltage of 0.1 V at its terminals,
when measured in a bandwidth B = 10 kHz.
By the use of the equation
Vn2 4kTA RB
Vn2 1014
TA 90.6 K
4kRB 4 1.38 10 23 200 104
208
104
Equivalent noise temperature
Thus the noise at the antenna terminals is

equivalent to that from a 200-ohm resistor at a
temperature of 90.6 K.
Note that other parts of a receiving system can
also be characterized by equivalent noise
temperatures in order to simplify the
computation of signal-to-noise ratio at the
output of a receiver.
209
Shot noise
The most common type of non-thermal noise in
electrical circuits is shot noise.
This phenomenon occurs whenever charged
particles cross a potential barrier as in
semiconductor junctions or vacuum tubes.
Small variations of kinetic energy among the
individual particles cause random fluctuations of
the total current.
Schottky first studied this effect in a vacuum-tube
diode operated under temperature-limited
conditions.
210
105
Shot noise
Schottky found that shot noise could be
represented by a current source with
in which q is the electronic charge (1.6 x 10-19 C)

and I is the DC current.
Since this simple equation is independent of both
frequency and temperature, shot noise seems to be
non-thermal white noise.
Later investigations have shown that in2(f)
actually decreases at frequencies above f1/ ,
where denotes the average particle transit time.
211
Shot noise in junction diode

Schottkys result also holds for the
semiconductor junction diode
The net DC current consists of two

components, given by the diode equation
212
106
Shot noise in junction diode

The net DC current consists of two components,
given by the diode equation
where V is the junction voltage and Is is the

reverse saturation current.
The two current components produce statistically
independent shot noise, so the total mean square
noise current density becomes
213
Semiconductor junction diodes noise

model
Figure below shows the complete noise source model,
including the diodes dynamic resistance
Unlike ohmic resistance, dynamic resistance is noiseless since

it does not correspond to any power dissipation.
Semiconductor junction diodes noise model
214
107
Non-white noise
Transistors, vacuum tubes, resistors, and other devices exhibit a low-
frequency phenomenon known as flicker noiseoften called one-
over-f noise because the mean square density is proportional to 1/fn
with n1.
Some semiconductor devices also produce burst or popcorn noise,
whose waveform resembles the random telegraph wave.
Flicker and burst noise pose serious problems for low-frequency
applications, but they usually can be ignored at frequencies above a
few kilohertz.
At much higher frequencies, capacitive coupling and various other
effects tend to increase noise in electronic devices.
Non-white noise may or may not be significant, depending on the
device and the application.
215
Semiconductor noise
Figure illustrates the frequency variation of
semiconductor noise caused by flicker, burst,
and white noise.
216
108
Noise in multi-terminal devices

Because resistors and antennas are two-terminal devices, it is easy to
describe their noise characteristics in terms of a noise temperature or an
equivalent noise resistance.
The situation is more complicated for transistors and other multiterminal
circuit elements because their internally generated noise depends upon
temperature,
operating point, and
input and output terminations.
For noise calculations in transistor circuits, the transistors are conveniently
represented as black boxes with specified noise figures and the physical
causes of the transistor noise are represented by equivalent noise sources.
Semiconductor manufacturers give data-sheet information on the measured noise

figures for a variety of operating conditions. 217
Electrical Noise
Noise terms
109
Definitions of noise terms
Various terms are used to define and compare

the relative amounts of noise produced in
electrical systems.
The following definitions and discussion will
provide the basis for understanding the
nomenclature and for computing the overall
effect of noise in a system.
219
Signal-to-Noise Ratio (SNR)

In a specified bandwidth, the signal-to-noise
ratio is defined as the ratio of signal power to
noise power at a port.
PS Vs2
SNR
PN Vn2
where Vs and Vn are the rms signal and noise
voltages, respectively.
In decibels, P
SNRdB 10 log10 S
PN 220
110
Minimum SNR
The larger the SNR, the less the signal is

corrupted by the noise.
The lowest permissible value of SNR depends
upon the application.
Approximate minimum value is 10 dB at the
detector input of a receiver.
221
Noise sources
Note that as a signal passes through a cascade of
amplifier stages, the SNR continually decreases
because each stage adds additional noise.
In most systems, however, the amplified output noise
is due primarily to
(1) the noise present along with the input signal, and
(2) the noise contributed by the first two stages
(such as the RF amplifier and mixer stages in a
receiver).
222
111
Noise bandwidth
The most common sources of noise (thermal

and shot noise) have an essentially uniform
spectral distribution so that the noise
transmitted through an amplifier is determined
by the bandwidth of the amplifier.
Shot noise is generated in semiconductors. It arises because the current crossing

junctions is made up of charge carriers that are emitted randomly; the number of these
carriers fluctuates statistically from instant to instant. Shot noise has essentially a flat
223
spectral distribution and is treated in the same manner as thermal noise.
Noise bandwidth
If the amplifier had a Constant gain characteristic, with
cutoff at fc
constant gain Av up
to some frequency fc
and zero gain
thereafter, the noise
bandwidth B would Filter with voltage gain Av (f), power
clearly be equal to fc. gain proportional to Av(f)2.
224
112
Limited frequency response
Generally, however, the frequency response is

limited (by the shunt capacitance or by a tuned
circuit) so that an abrupt cutoff of the
frequency response is not achieved.
Thus a more sophisticated determination of
noise bandwidth is required.
225
Filter
Consider a filter, as shown in figure, that has
voltage gain Av(f)=V2/V1.
Filter with voltage gain Av (f), power
gain proportional to |Av(f)|2.
226
113
Frequency response
Since the noise power is related to the mean-

square voltage, it is convenient to plot the
frequency response in terms of lAv(f)I2, as
shown by the solid line in Fig., where lAml2 is
the maximum value of this curve.
Illustration of
noise-equivalent
bandwidth B,
defined by equal
areas under the
dashed and solid
curves. 227
Filter output
If the input to this filter is white noise with mean-
square voltage v1n2/Hz, the corresponding mean-
square output voltage in a 1-hertz interval at
frequency f is
Addition of all such increments over the frequency

band yields
228
114
Noise-equivalent bandwidth
The value of the integral on the right side is the area
under the solid-line curve of |Av(f)l2 in Fig..
The dashed line shows a rectangular spectrum of the
same maximum height lAm|2 and with bandwidth B.
Illustration of
noise-equivalent
bandwidth B,
defined by equal
areas under the
dashed and solid
curves. 229
Noise-equivalent bandwidth (B or BN)
The noise-equivalent bandwidth B is the value

that gives equal areas under the solid- and
dashed-line curves such that
or
230
115
Comment
This is the value of B that should be used in equations
earlier, and obviously it must be evaluated for the
particular system being analyzed.
The integral is not always easy to evaluate.
However, in many RF amplifiers, the bandwidth is
established by tuned RLC circuits for which the noise
bandwidth is /2 times the 3-dB bandwidth of the
circuit.
231
Available noise power

The available power Pa of a source is the
maximum power that can be drawn from the
source.
If the source has internal impedance Zs = R +
jX, the maximum power will be delivered to a
conjugate-matched load (ZL = R - jX).
232
116
If the open-circuit voltage of the source is V,

the maximum power transfer theorem yields
If R is a source of thermal noise, V2=4kTRB,

and
233

The available noise power in a bandwidth of 1
Hz is
Since Pa is independent of R, the available

noise power from all nonzero finite resistances
is the same.
234
117
Amplifier noise
A detailed circuit model showing all the
individual noise sources within an amplifier
would be very complicated and of little practical
value.
Consequently, alternative methods have been
devised for the analysis of noise in amplifiers.
Two particularly useful measures of amplifier
noise are the effective noise temperature and the
noise figure.
Both of these measures involve the concept of
available power gain.
235
Model of a general two-port network.
A general two-port network that could serve as

a model for an amplifier stage, filter, or other
network.
236
118
Model of a general two-port network.
The source has open-circuit voltage Vs and internal

impedance Zs.
The two-port has input impedance Z1, output
impedance Z2, and open-circuit output voltage Vo.
The load is an impedance ZL.
The open-circuit voltage gain of the two-port will be

called H(f)
237
Available signal power

The available signal power from the source, as
defined previously, is
Similarly, the available signal power at the

output port is
238
Note: Rs and R2 are the real parts of corresponding impedances.
119
Available power gain
The available power gain Ga of the network is

defined as the ratio Pao/Pas, that is,
Note that this expression does not account for

impedance mismatches at the input and output
ports.
239
Voltage relations
Examination of the network yields the
following relations:
Z1 Z s 1 Z1 1
Vs V1
Z1 Vs Z 1 Z s V1
Construction of Vo/Vs:
Vo 1 Z1 V Z1
Vo o H f
Vs Vs Z1 Z s V1 Z1 Z s
240
120
Available power gain

Substitution of the expressions
into
gives the following relation for available power gain

at frequency f
241
Remark
The two-port model applies only if the network is
unilateral (no reverse transmission).
For this case it is worth emphasizing that Ga given by
the previous expression is independent of the value of
ZL.
The available gain of several cascaded unilateral
networks is equal to the product of the Ga values of
the individual networks.
Note that Ga is not the actual gain of the network the value of Ga would be obtained in
242
practice only with ideal matched conditions (Z1 = Zs*and ZL = Z2*) at both ports.
121
Definition: Equivalent noise temperature
The noise temperature at any port of a network

is defined as follows:
A noise source that has an available power Pa in a
small frequency interval f has an equivalent noise
temperature equal to
Te = Pa/kf.
If the noise power spectrum of the source is not
flat, Pa and Te are frequency-dependent.
Note:
243
Excess noise temperature (Tx)

Noise generators used for amplifier testing are
often calibrated in terms of excess noise
temperature,
in which T is the noise temperature of the

source and T0=290 K is the standard reference
temperature.
244
122
Noiseless network
If a thermal noise source of temperature T is

connected to a noiseless network with small
bandwidth f and available gain Ga(f), the
available noise power from the source is
and the available noise power at the output of

the network is
245
Model of a noisy network

If the network is noisy, it will produce additional
noise power, Pne, at the output.
With the same input noise as before, the output noise
power will be
246
123
Noiseless network arrangement
Now replace the noisy network with a

noiseless network having the same available
power gain Ga(f), and account for the output
noise Pne by means of an extra noise source on
the input side.
247
Effective input noise temperature
The temperature Te of this extra source is

adjusted to produce Pne at the output,
and
The value Te is called the effective input noise

temperature of the network.
Note: This way of representing network noise is very useful in the determination of
248
overall signal-to-noise ratios of cascaded amplifiers.
124
Noise Figure (NF or F)

It seems to be the variety of ways in which the
relative noisiness of a device or system can be
expressed.
Fortunately, a single index called the noise figure can
be used to compare noise performance.
The noise figure (NF) of a two-port network gives a
measure of the degradation of the SNR between the
input and output ports.
Note: Watch out for several variations of the term with subtle differences that can be
249
confusing.
I/O-powers
Figure shows a noisy network with input
signal and noise powers and Psi and Pni
respectively, and corresponding output signal
and noise powers Pso and Pno.
Signal and noise powers at the input

and output of a two-port network.
250
125
Definion: Noise Figure (Noise Factor)
The noise figure is defined over a specified

bandwidth as
251
NFdB (or FdB)

The value of NF is often expressed in decibels
through the relation
For a noise-free network, the input and output

SNRs will be equal and NF = 1 or NFdB=0.
Practical circuits always have larger noise
figures than this.
252
126
Additional specifications for N F

Equation
provides a conceptual definition of the noise figure, but it

requires further qualification to make it more precise.
The ratio Psi/Pso is equal to 1/Ga(f), where Ga(f) is the frequency-
dependent available power gain of the network.
The variation of Ga(f) with frequency must be taken into account.
Furthermore, the input power from the signal source is a function of
temperature.
In order to obtain a standard value for NF, the source temperature must be assumed
to be 290 K.
These considerations have led to the following definitions.
253
Spot noise figure

At a selected input frequency, the spot noise
figure is the ratio of
(1) the total available noise power per unit
bandwidth at the output port to
(2) the portion produced at the input
frequency by the input termination, whose
noise temperature is 290 K.
254
127
Spot noise figure
If the network provides a conjugate match at

the input port, the available power from the
standard-temperature source in a 1-Hz
bandwidth is equal to Pa(1Hz)=kT0.
Hence, the spot noise figure is given by
255
Note
In practice, the value of Pno is measured over a
small bandwidth f that is more than 1 Hz due
to the practical limitation on filter bandwidth.
The equation for NF then becomes
256
128
Average noise figure

Over wide bandwidths in which Ga(f) varies
appreciably, the average noise figure NF is
given by
in which Pno is the total noise power delivered

to the output termination in noise bandwidth B,
and Gmax is the maximum value of |Ga(f)|.
257
Overall NF of cascaded networks

Of primary interest in the evaluation of noise
performance of multistage amplifiers (such as
radio receivers) is the overall noise figure of
the system.
In general, the overall noise figure is evaluated
for a bandwidth B that is the bandwidth of the
overall system.
In a radio receiver the IF stages are narrow-band
and they determine the amount of noise that
reaches the detector.
258
129
Output noise power
From
if the source temperature T is assumed to be T0

(required for the standard definition of NF),
the output noise power of a single stage in a
small frequency band is
259
Relation between NF and Te

Substitution of the previous equation into
gives
Thus the noise figure referred to a standard-temperature

source can be expressed in terms of the equivalent noise
temperature of the network, and vice versa.
260
130
Two networks in cascade
Figure shows two cascaded networks with

their respective available gains, effective noise
temperatures, and noise figures.
A noise source at standard temperature is input
to the system.
261
Available output noise power of two

stages
Small bandwidth f is assumed for all parts of the system.
The available output noise power of two cascaded
network is found by the use of the output noise power of a
single stage
and some regrouping of terms
262
131
Overall effective (input) noise

temperature
A comparison of this relationship with the equations
leads to an expression for the effective input

temperature of the two networks in cascade
Note that the input source temperature does not appear in the expression.
Pne is a symbol for additional noise power at the output caused by noisy network. 263
Overall effective (input) noise

temperature
Here Te1,2 is the effective input temperature
that accounts for all of the output noise
introduced by the noisy networks.
For n networks in cascade, the corresponding
expression is
264
132
Overall Noise Figure

Introduction of the network noise figures by the use
of
in
leads to the following expression for the overall noise

figure
265
Conclusion
It is apparent from this relation that the NF of the first

stage in a system has the predominant effect on the
overall NF, unless Ga1 is small or NF2 is large.
Therefore, the system designer should always try to
minimize the noise produced in the first stage by the
choice of low-noise transistors and the selection of
operating conditions that minimize noise.
266
133
If Ts T0=290K
By definition, the standard noise figure is the
ratio of the input SNR to the output SNR with
the input source at the standard 290 K noise
temperature.
In many practical cases, however, the input
noise temperature is not 290 K and the
standard noise temperature does not describe
accurately the SNR degradation from input to
output of a system.
267
Actual noise figure

A true measure of the SNR degradation is the
actual noise figure defined by
in which TsT0.
This value is related to the standard noise
figure evaluated for Ts=T0 by the relation
268
134
Example
Calculate effective temperatures and noise figures for the
receiver front end shown in below. The antenna has a
(radiation) resistance of 70 ohms and an effective temperature
of 20 K due primarily to external radiation.
The noise contributed by the receiver local oscillator is assumed to be negligible (which
269
is not always the case).
Solution
Noise figures and gains for the RF amplifier
and mixer are given in decibels, and must be
converted to actual values for use in the
computation.
For the RF amplifier, NF1=2, and Ga1 = 10.
For the mixer, NF2 = 4.47 and Ga2 = 7.94.
270
135
Solution
By the use of
the effective noise temperatures of the two

units are found to be
Te1=290 K and
Te2=1006 K.
271
Solution
Then from
the effective input temperature of the receiver

(excluding the antenna) is found as
Note: Here Tr Te1,2 272
136
Solution
The overall noise figure can be calculated from
either
or
273
Comment
In the preceding example, the value of NF was found
to be 2.35 or 3.7 dB.
However, since the antenna temperature was 20 K,
the use of
to calculate the actual noise figure yields

NFact = 1 + (2.35-1)(290/20) = 20.43 or 13.1 dB.
By definition, the standard noise figure is the ratio of the input SNR to the output SNR
with the input source at the standard 290 K noise temperature. Here, however, the input
noise temperature is not 290 K and the standard noise temperature does not describe
274
accurately the SNR degradation from input to output of a system.
137
Amplifier noise considerations

For calculations of NF or Te of a system
comprising one or more stages, knowledge is
required of
(1) the noise delivered from the signal source,
(2) the noise-equivalent bandwidth B,
(3) the thermal noise generated in various resistances
in the circuit, and
(4) the noise generated within the solid-state devices.
Noise produced within semiconductors can be predicted by the use of

equivalent circuits. However, because there are so many variables to be taken
into account that such predictions are best made by the use of analysis
275
programs.
Electrical Noise
Noise calculations
138
Example
Verify that the change in variable from f to

in
4kTG
V
n
2
df
0 G 2 (2fC ) 2
yields

4kT 1
2G 0
V
n
2
2
d
C
1
G 277
Solution
d
f df
2 2
1 4kTG
Vn2
2 G (C ) 2
0 2
d
4kTG d
2 0

2G C 2
1 ( )
G
4kT d

2G 0 1 ( C ) 2
G
278
139
Evaluation of the previous integral

4kT d G
Vn2
2G 0 1 ( C ) 2
, 3dB , (G || C - circuit)
C
G
4kT d G

2G 1 ( ) 2
0
,x
3dB
d 3dB dx dx
C
3dB
4kT d 4kTG dx
2G 0 1 ( ) 2 2GC 0 1 x 2

3dB
4kT kT
arctan x0

2C C

0
2
279
Example
Show that
4kTG
V
n
2
df
0 G 2 (2fC ) 2
yields
1 G
Vn2 4kTR f 3dB , when f 3dB
2 2 C
280
140
Solution
4kTG 4kTG df
Vn2 df
0 G 2 ( 2fC ) 2 G 2 0 2fC 2
1 ( )
G
4kT df 1 2C

G 0 f
,
f 3dB

G
1 ( )2
f 3dB
4kTf 3dB dx f

G 0 1 x2
,x
f 3dB
df f 3dB dx
1
G
4kTf 3dB
arctan x 0 4kTR f 3dB
R

G 2
0
2
281
Example
According to quantum mechanics the spectral density of
thermal noise can be expressed in form
hf V2
Gnoise f hf
Hz
e kT
1
where k=1.38*10-23 J/K
h=6.62*10-34 Js
is the proportionality symbol
a) Show that Gnoise(f) is essentially constant in the region of the radio

spectrum.
b) Determine an expression for Gnoise(0).
Note: 0/0 is not the correct answer.
282
141
Spectral density function
Thermal noise level [dBW] T=290K
Frequency [GHz]
283
Solution a)
Of course, we can approximate with series expansion of e
2
hf
hf
hf kT
e kT
1
kT 2
hf hf
, 1
kT kT
Now we get
hf
G f kT , 0 1
kT
284
142
Solution a)
Lets select the standard temperature T=T0=290K and a
rough but conservative estimate
hf upper kT0
0.1 1 f upper 0.1
kT0 h
Substituting numerical values kT0 4 10 21 J
and h 6.62 10 -34 Js
f upper 1012 Hz
This upper limit is in the infrared portion of the
electromagnetic spectrum.
285
LHospitals rule
If lim M f lim N f 0
f 0 f 0
then we can calculate Mf

lim
f 0 N f
using LHospitals rule
Mf M f
lim N f lim N f
f 0 f 0
whenever the limit on the right can be found.
286
143
Gnoise(0)
Lets denote Mf hf V2
Gnoise f
N f hf
Hz
e kT
1
d
hf
Mf M f df
lim
f 0 N f
lim
f 0 N f
lim
f 0 d
hf

e kT 1
df

h kT
lim hf
lim hf
kT
f 0 h kT f 0
e e kT
kT
Gnoise(0) kT 287
Example
Reduce the circuit
to the Thevenin equivalent
288
144
Solution
289
Solution
Note: Removing the voltage source means that we replace the source voltage with
short-circuit.
Note: In circuit noise analysis we simply use the thermal noise source model
vn2(f)=4kTRTh , not the signal source model based on the voltage division as in this
problem. Circuits containing more than one resistor may be analyzed by reducing them
to one (Thvenin) equivalent resistance RTh and applying vn2 (f)=4kTRTh to obtain the
mean-square noise voltage density. So the noise Thevenin equivalent of such a circuit
is then a voltage source with this mean-square voltage density in series with an ideal
(noiseless) resistor RTh. 290
145
Example
In the network below R1 and R2 are thermal resistances
at different temperatures (T1T2). Let R2 = R1 = R and L
= R/(2). Obtain expressions for vn2(f) and p(f).
291
Solution
Now resistances are at different temperatures
(T1T2) Nyquists formula
vn2 f 4 R f kT
is not valid.
However, the sources are independent.
Thus, we can use superposition and sum mean square
values to find resulting frequency densities.
292
146
Solution
Applying superposition the total short-circuit mean square
current density is
vn21 f vn22 f
in21 f in22 f 2
2
Z1 Z2
where vn21 f 4 RkT1 and vn22 f 4 RkT2
2
R
R jfR R fR
2 2
R j 2f 2 2
Z1
2
2
Z2 R2
4k T1
in2 f T2
R 1 f 2

293
Solution
Z Th f ( R jfR) || R
R( R jfR) R1 jf

R ( R jfR) 2 jf
Z Th f
2 R1 jf
2

R2 1 f 2
2 jf
2
4 f 2
According to Ohm' s law
v f Z Th f
2
i f
2 2
4kR T1 1 f 2 T2
n n
4 f 2
294
147
Solution
R1 jf
Z Th f
2 jf
ReZ Th f Re
R1 jf 2 jf R 2 f 2
2 jf 2 jf 4 f 2

p f
vn2 f
k

T1 1 f 2 T2
4 ReZ Th f 2 f 2
T1 T2 T
Note : p f kT
295
Note
When the resistances are in thermal equilibrium at
temperature T, Nyquists formula states that
vn2 f 4 RTh f kT , RTh f ReZTh f
Z Th f RTh f jX Th f
Hence, the mean square voltage density takes the
shape of the equivalent resistance RTh(f).
Nyquists equation includes the special case of an all-
resistive network whose equivalent resistance will be
independent of f.
296
148
Note
If in the previous problem (T1=T2=T) Nyquists

formula is valid and then
vn2 f 4 RTh f kT , RTh f ReZ Th f

2f2
4kTR and
4f2
vn2 f
p f kT
4 ReZ Th f
297
Example
Lets study the circuit model of a noiseless amplifier inserted
between a source and a load.
For simplicity, we have omitted any reactances that might be
associated with the source, amplifier, or load impedances.
In this analysis we use spectral density functions
298
149
The available power density

The amplifier is characterized by an input resistance ri,
output resistance ro, and voltage transfer function H(f).
The source generates a mean-square voltage density
vs2(f) representing noise or an information signal or
bothand the available power density from the source is
pas(f) = vs2(f)/4Rs.
The available power density at the output of the amplifier
is
299
Amplifiers available power gain

We define the amplifiers available power gain
ga(f) as the ratio of these available power
densities, i.e.,
The actual power gain of an amplifier equals

the available power gain when impedances are
matched to obtain maximum power transfer at
input and output.
300
150
Noisy amplifier
Well assume from now on that the source generates
white noise, thermal or non-thermal, with noise
temperature Ts.
Then pas(f) = kTs and the available noise power
density at the output of a noiseless amplifier will be
pao f g a f pas f g a f kTs
But a noisy amplifier contributes additional internally

generated noise pint(f).
301
Available output power density

Since the internal noise is independent of the source noise, we
write
where pint(f) stands for the available power density of the

internal noise seen at the output.
Figure depicts the previous equation in the form of a block
diagram.
302
151
Total available output noise power
Integration then yields the total available output noise power

pao f df kTs g a f df pint f df
0 0 0
an expression that calls for some simplifications.
Most amplifiers in a communication system have a frequency-
selective response, with maximum power gain g and noise
equivalent bandwidth BN.
These parameters are related to ga(f) by
303
Effective noise temperature

So the first term of No reduces to kTs gBN.
Next, to simplify the second term, we define the
effective noise temperature of the amplifier to be
1
Te pint f df
gkBN 0
Hence, the total output noise power becomes

N o kTs gBN gkBN Te gk Ts Te BN
It is diagrammatically portrayed in the following
figure.
304
152
Total output noise power
This diagram brings out the fact that Te

represents the internal noise referred to the
input and thereby expedites calculations of
signal-to-noise ratios.
305
Block-diagram representation of a
noisy amplifier
306
153
Calculations of signal-to-noise ratios

The figure represents a noisy amplifier with signal plus white noise
at the input.
The available signal power from the source is Ss and the signal
spectrum falls within the passband of the amplifier so the available
signal power at the output will be
So = gSs.
307
Output signal-to-noise ratio

Thus, using the two previous equations, the
output signal-to-noise ratio is
S gS Ss
s
N o N o k Ts Te BN
Note that the gain g has canceled out in
numerator and denominator.
308
154
Note
Although the source noise does not necessarily have a defined
bandwidth, the source signal-to-noise ratio is taken by convention to
be
S def . S s

N s kTs BN
which just corresponds to the signal-to-noise ratio produced by an
ideal noiseless filter with unit gain and bandwidth BN.
309
(S/N)o expressed with (S/N)s

However, reforming (S/N)o expression of the noisy
amplifier produces
S gS Ss Ss
s
N o N o k Ts Te BN 1 Te kT B
Ts s N
1 S

1 Te N s
Ts

S def . S s
where
N s kTs BN 310
155
SNR comments
Note that in case of a noisy amplifier
(S/N)o< (S/N)s.
We also see that the degradation of signal-to-
noise ratio due to a noisy amplifier depends upon
the value of effective noise temperature relative to
the source noise temperature.
In particular, if Te<<Ts then (S/N)o (S/N)s
Meaning that under this condition the internal
noise has little effect and the amplifier appears to
be noiseless.
311
Mismatching comments
When impedances are not matched at the input or
output all signal and noise powers will be less
than the available powers.
Nonetheless, the previous SNR calculations are
still valid because they express power ratios
measured at specific points, so the impedance
mismatch factor cancels out along with the gain.
The effective noise temperature is therefore a
significant parameter, irrespective of impedance
matching.
312
156
Noise figure (F)
Another measure of amplifier noise is the

noise figure F, defined such that
when Ts T0
Since So= gSs and (S/N)o = Ss/(kT0 BN) when

Ts=T0
No T
F 1 e
gkT0 BN T0
313
F vs. Te
Conversely
Te F 1T0
A very noisy amplifier has Te>>T0 , and F>>

1 in which case we generally express the
value of F in decibels.
A low-noise amplifier has Te < T0 and 1 < F <
2 in which case we usually work with the
effective noise temperature.
314
157
Noise parameters of some

microwave amplifiers
315
Spot noise
Equation
No T
F 1 e
gkT0 BN T0
defines the average or integrated noise figure in the sense that
No involves the integral of p0(f) over all frequency.
But sometimes we need to know how the internal noise varies
with frequency.
The spot noise figure F(f) contains this information in the form
po f
F f , when Ts T0
kTs g a f 316
158
Amplifier noise measurement
Measuring absolute noise power is a difficult ,

so clever techniques have been developed for
amplifier noise measurement with a relative
power meter connected at the output.
One technique utilizes a calibrated source of
white noise, such as diode noise generator,
impedance-matched to the input of the
amplifier.
317
Noise measurement procedure

1 Set the noise source temperature at Ts=T0 and record the power
meter reading N1. This value corresponds to
N1 CN o Cgk T0 Te BN
where the proportionality constant C includes any impedance
mismatch factor at the output.
2 Increase the source temperature to Ts=T0+Tx such that the
power meter reading has doubled
N 2 Cgk T0 Tx Te BN 2N1
318
159
Noise measurement procedure
Then form
N 2 Tx T0 Te
2 Te Tx T0
N1 T0 Te
The final result follows substituting Te
to noise figure expression
No T T T T
F 1 e 1 x 0 x
gkT0 BN T0 T0 T0
Note: We dont need to know g, BN, or the constant C. 319
Example
An amplifier with g = 60 dB and BN =2 MHz has No = 40 nW
when the source noise is at room temperature.
(a) Find the effective noise temperature Te and noise figure F.
(b) Calculate the increased source temperature Ts needed for
the second step of the previous measurement procedure.
Note: For computational purposes, the quantity kT can be

rewritten as
T T
kT kT0 4 10 21 J
T0 T0
T0 290K is the standard room temperature 320
160
Solution
(a) Find the effective noise temperature Te and noise figure F.
The total output noise power
N o gk T0 Te BN 106 k T0 Te 2 10 6
T0 Te
2 1012 4 10 21 40 10 9
T0
T0 Te
So 5 Te 5T0 T0 4T0
T0
Te 4T
F 1 1 0 5
T0 T0
321
Solution
(b) Calculate the increased source temperature Ts needed
for the second step of the previous measurement procedure
From a)
T
F x 5 Tx 5T0
T0
and the increased source temperature
Ts =T0+Tx=6T0=6*290=1740K
322
161
Lossy two-port networks

Lossy two-port networks are such as transmission lines and
connecting cables.
Power loss implies dissipation by internal resistance.
Consequently, the internal noise is thermal noise at the
ambient temperature Tamb and pint(f)=kTamb.
However, we cannot use the previous noisy amplifier model
presented in the following slice because lossy two-ports are
bilateral, meaning that a portion of the internal noise flows
back to the input.
323
Block-diagram representation of a
noisy amplifier

324
162
Bilateral lossy two-port

When impedances are matched, a bilateral lossy two-port has
constant gain g < 1 in both directions, so gpint(f) flows back to
the input while (1-g) pint(f) goes to the output.
The total available noise power in bandwidth BN at the output
thus becomes
N o gkTs BN 1 g kTamb BN
1
gk Ts L 1Tamb BN , L
g
where L is the transmission loss or attenuation.
325

Comparing our expression for No with Eq.
N o gkTs BN 1 g kTamb BN
1
gk Ts L 1Tamb BN , L
g
we obtain the effective noise temperature
Te L 1Tamb
326
163

Substituting the effective noise temperature
expression
Te L 1Tamb
to the noise figure equation
No T
F 1 e
gkT0 BN T0
we obtain
Tamb
F 1 L 1
T0
Note: If a lossy two-port is at room temperature, then Tamb = T0 and F = L. 327
System noise calculations

Here we take up the analysis of cascade-
connected systems that include amplifiers and
other noisy or lossy two-port networks.
Our objective is to develop expressions for the
overall performance of the system in terms of
the parameters of the individual stages.
328
164
Example: Two noisy two-ports

Examine the cascade of two noisy two-ports, where
subscripts identify the maximum power gain, noise
bandwidth, and effective noise temperature of each
stage.
We assume that both stages are linear and time

invariant (LTI).
329
BN B2B1 g = g1g2
We further assume that the passband of the second stage falls
within the passband of the first stage, so B2 B1 and the
overall noise bandwidth is
BN B2.
This condition reflects the sensible strategy of designing the
last stage to mop up any remaining noise that falls outside the
signal band.
The overall power gain then equals the product
g = g1g2
since the first stage amplifies everything passed by the second
stage.
330
165
The total output noise power consists of three

terms:
source noise amplified by both stages;
internal noise from the first stage, amplified by the
second stage; and
internal noise from the second stage.
Thus,
N o gkTs BN g 2 g1kT1 BN g 2 kT2 BN
T
gk Ts T1 2 BN
g1 331
Overall effective noise temperature

and noise figure
The overall effective noise temperature is
T2
Te T1
g1
The overall noise figure is
T1 T F 1
F 1 2 F1 2
T0 g1T0 g1
which follows from the general relationship
Te
F 1 .
T0
332
166
General case
The foregoing analysis readily generalizes to the case of three or
more cascaded LTI two-ports.
The overall effective noise temperature and noise figure are given
by
T2 T
Te T1 3
g1 g1 g 2
F2 1 F3 1
F F1
g1 g1 g 2
Both expressions bring out the fact that the first stage plays a critical
role and must be given careful attention in system design.
333
Example: Lossy 1. stage

Suppose the first stage happens to be a connecting cable or any other lossy
two-port.
From the noise viewpoint, the attenuation has a double harmful effect since
g1 = 1/L1 < 1 and
T1 L1 1Tamb Te L 1Tamb
The overall noise temperature thus becomes
T2 T
Te T1 3
g1 g1 g 2
L1T3
L1 1Tamb L1T2
g2
Note: L1 (>1) multiplies the noise temperatures of all subsequent stages. 334
167
Example: Preamplification principle

Suppose the first stage is a preamplifier with sufficiently large gain
g1 that the overall noise temperature reduces
T2 T
Te T1 3
g1 g1 g 2
T1 , g1
The system noise is then determined primarily by the preamplifier.
The remaining stages provide additional amplification and filtering,
amplifying the signal and noise without appreciably changing the
signal-to-noise ratio.
The design of low-noise receivers is usually based on this
preamplification principle.
335
Complete receiver
Here a complete communications receiver has been
divided into two major parts, a predetection unit
followed by a detector.
The detector processes the amplified signal plus noise
and carries out a nonlinear operation, i.e., analog
demodulation or digital regeneration.
It is the reasonable assumption that the detector
introduces negligible noise compared to the amplified
noise coming from the predetection unit.
We are concerned here with the predetection signal-to-
noise ratio denoted by (S/N)R.
Note: When the predetection unit includes a frequency converter, as in a
superheterodyne receiver, its conversion gain takes the place of available power gain. 336
168
Communications receiver
(a) Communications receiver; (b) noise model of predetection unit.
BT is the transmission
bandwidth required
for the signal
337
Predection unit
The predetection portion of a receiver is a cascade of
noisy amplifiers and other functional blocks that act
as LTI two-ports under the usual small-signal
conditions.
Hence, as indicated in the previous block diagram a),
the entire predetection unit can be characterized by its
overall effective noise temperature calculated from
the general equation
T2 T
Te T1 3
g1 g1 g 2
338
169
Predetection signal-to-noise ratio

(S/N)R
For a well-designed receiver, the predetection noise bandwidth
essentially equals the transmission bandwidth BT required for
the signal.
If the available signal power at the receiver input is SR and the
accompanying noise has temperature TR then the previously
presented SNR equation becomes
S S SR S
R gS s Ss

N R pBT k TR Te BT
N N o k Ts Te BN
o
where p k TR Te kTN
339
System noise temperature

The sum
TN TR Te
is called the system noise temperature, and p
represents the total noise power density referred to
the input of an equivalent noiseless receiver
corresponding to the diagram below.
340
170
Example
The signal received at a satellite ground station is
extremely weak.
Fortunately, the accompanying noise comes
primarily from cold atmospheric phenomena
and has a very low temperature.
Minimizing the receiver noise is therefore
essential.
In contrast, a receiving antenna pointed at or
below the horizon picks up blackbody radiation
from the hot earth; then TRT0 and receiver
noise will be relatively less important.
341
Example
Calculate a) the overall effective noise temperature Te
b) the system noise temperature
c) the predetection signal-to-noise ratio(S/N)R
Figure depicts an illustrative low-noise microwave receiver for a satellite signal

with frequency modulation. The waveguide is part of the antenna feed structure
and introduces a small loss. Two preamplifiers are employed to mitigate the
noise of the high-gain FM receiver. 342
171
Solution a)
Calculate a) the overall effective noise temperature Te
b) the system noise temperature
c) the predetection signal-to-noise ratio (S/N)R
T1 L1 1Tamb Te L 1Tamb
T2 T
Te T1 3
g1 g1 g 2
L1T3
L1 1Tamb L1T2
g2
1.05 170 1.05 1860
Te (1.05 1)290 1.05 9
100 100 10
14.5 9.5 1.8 2.0 27.8 K
Note: The waveguide loss accounts for half of Te, while the noise from the FM
receiver has been nearly washed out by the preamplification gain. 343
Solution b)
b) Calculate the system noise temperature
TN TR Te 30 K 27.8 K 57.8K
0.2T0
344
172
c) Predetection (S/N)R
(a) Communications receiver; (b) noise model of predetection unit.
BT is the transmission
bandwidth required
for the signal
S S SR 10 12
R 50 17 dB
N R pBT k TR Te BT 0.2 4 10 25 10
- 21 6
where p k TR Te kTN k 57.8 k 0.2T0 0.2 kT0 0.2 4 10 21
Note: This SNR would be too low for analog communication without the further
345
improvement caused by FM demodulation.
Example
Suppose the parametric amplifier in the figure below could be
mounted directly on the antenna, ahead of the waveguide.
Find the system noise temperature TN with and without the
FET preamplifier.
346
173
Solution
Find the system noise temperature TN with and without the
FET preamplifier.
With FET: 14.5

Te 9 1.8 2.0 12.9 K
100
TN TR Te 30 12.9 42.9 K
Without FET: 14.5 1.05 1860

Te 9 28.7 K
100 100
TN TR Te 30 28.7 58.7 K
347
Note: FET increases (S/N)R by 58.7/42.9=1.371.4 dB
Electrical Noise
Recap
174
Noise
Thermal noise: electron collisions with the lattice structure in
resistive materials
For a resistor, we have a noise voltage
Shot noise: charge is not a continuous quantity

1/f-noise at low frequencies
Quantum noise: power in quanta (sub-mm )
Noise is characterized by noise temperature or noise factor
Noise factor describes the decrease of signal-to-noise ratio
caused by the system noise
At radio frequencies, the noise mainly considered is so called
white noise noise which has flat frequency spectrum 349
Receiver noise
Characterization of receiver noise (as well as noise of other devices)
relies on some main points and parameters to be remembered
For use as a noise reference:
Available noise power of a matched load
For use as parameters of the receiver or device:
Noise factor (noise figure when in dB)
Noise temperature
Noise temperature is applicable also e.g. to antenna then it is
called antenna noise temperature
Also noise bandwidth is a useful and effective concept
Idea: All of these simplify handling noise very complex subject
otherwise
350
175
Reference noise
Available noise power of a matched load
Resistive loads are used as reference noise sources in

calculations but also in practice (matched resistive
terminations in device inputs)
Available noise power suitably depends on considered
bandwidth and the physical temperature of the load
Physical temperature can be measured quite accurately => a
good resistive termination is a good absolute reference for noise
when we are characterizing receivers and their sensitivity
Physical temperature is useful alone for defining noise power
density
351
Reference noise temperature T0

For definitions and related calculations, a commonly
defined true reference is always needed
In radio engineering we use standard temperature T0 = 290
K (= exact constant!) as the assumed physical temperature
of an absolute noise reference termination
Thus we get e.g. noise factor results without any definition
dependent errors or guessing
But as a result of the choice 290 K, it is often
convenient that T0 anyway happens to be near typical
room temperature
Conclusion: T0 = 290 K is actually constant - not
variable such as any room temperature or physical
temperature can be
352
176
Noise factor
Noise factor (F) indicates how many times larger the output
noise power of a device is compared to a noiseless device
when both have a 290 K resistive termination in input
Nout includes the noise added by the device as well as the

amplified input noise NinG => F > 1
F has to be unambiguous for a device => F is specified with
using known defined Nin = kT0B
Note: Here noise temperature has to be reference value T0 = 290 K, not some
freely chosen temperature 353
F =SNRin/SNRout
Noise factor also provides a way to connect

signal-to-noise ratios (S/N) in input and output
of a two-port
We get the S/N relation in the noise factor
equation introducing input signal power Sin
354
177
Noise figure
Noise figure is noise factor in dB:

F [dB] = 10 log10(F)
Typically amplifier datasheets give noise
figure specifications
Note: Some authors use these two concepts just the other way round. 355
Effective noise temperature

For receiver devices (and others) we can define noise also by
using effective noise temperature
This is always calculated as in the input of the device so that
we can use a well-defined equivalent source to help in noise
calculations
Otherwise noise in a device would be a very complex matter to
handle
Definition: Effective noise temperature Te of a device =
physical temperature of a matched load connected to a similar
noiseless device to obtain output power equal to the noise
power of the device
356
178
Relation between Te and F
Relating Te to the noise factor F:
357
Derivation of the previous relation
Note: Te is not any physical temperature when linked to the device

as a parameter. 358
179
Noise bandwidth
Gain of a receiver is usually not constant as a

function of frequency
Thats why for calculations of total noise we
need to define an effective noise bandwidth
to use with maximum gain:
Equal
areas
Note: Scale of G is in absolute values, not dB. 359
Note
Through this way we can use Gt,max and Bn as
simple parameters in noise power calculations
of device.
360
180
Example
Picture shows a G-curve of a receiver low-noise amplifier
(LNA) with 13 dB of maximum gain.
Define the noise bandwidth Bn of the LNA.
G(f)
361
Noise of an attenuator
Noise temperature of a resistive attenuator
Note that waveguides and all kinds of wires

and cables have loss and thus can be
considered as attenuators adding noise =>
Attenuator noise has practical importance
especially in sensitive receivers
362
181
Noise factor of attenuator

Noise factor of a resistive attenuator is
If we assume that Tphys T0 as might be in practice =>

we get for attenuator noise factor in this simple case:
Thus in practice about at room temperature, the noise

factor and loss of an attenuator are more or less equal
in value!
363
Receiver noise
Receivers are chains of stages - some noise is added
and also amplified in them - how to calculate it all?
Noise factor and noise temperature of a chain of
stages
Friis formulas for

matched stages
364
182
Comment on receiver noise

Note: If G1 is large, the first stage dominates
Basically this is good because we can focus on

getting a low-noise first stage with gain
First stage noise dominance is very understandable:
this noise besides input noise - is amplified most to
the final output in a chain of amplifiers, i.e., by the
product of all gains: G1 G2 G3
In fact, it follows that the noise added by the final
stage is only a negligible part of the total noise in the
output!
365
Comparing amplifiers
When amplifiers are compared, the noise
measure counts:
Smaller M is better =>a low-noise amplifier with a

small noise factor but with also low gain may not
be a desirable choice
Friis formula can always be used for final check
366
183
Example
Lets consider a simple example of two amplifiers in
chain
How about combined noise?
First amplifier (LNA stage): T1 = 100 K, G1 = 13 dB (=>
G1 = 20)
Second amplifier (gain stage): T2 = 300 K, G2 = 26 dB
(=> G2 = 400)
367
Comment on example
We notice that of the noise added by amplifiers,
100 K/115 K 87 % is from the 1st and only 15
K/115 K 13 % from the 2nd amplifier => we
have a typical dominating first stage
If we use the amplifier chain with a 290 K input
termination, then 290 K /(115 K + 290 K) 72 %
of noise in the chain output is amplified noise
from the input termination while amplifiers
contribute 25 % and 4 %, respectively
In practice, of course, input noise depends on
application
368
184
Noise temperatures of receivers

Background noise sources as a reference
Molecular
resonances of water
vapour and oxygen:
loss as in an
attenuator => also
noise
Background
noise of the
Big Bang
Note: The HEMT or High Electron Mobility Transistor is a form of field effect transistor,
FET, that is used to provide very high levels of performance at microwave frequencies. 369
Antenna noise temperature

Brightness of black surface
Noise power received by an antenna
Note: Noise power received by an antenna is constant although brightness B

increases as 1/ 2 or f 2 . This is because antenna beam solid angle changes as
2 or 1/f 2 and compensates. 370
185
Noise temperature of the sky
371
Radio link
Signal-to-noise ratio is a critical parameter for a radio link
For example, frequency modulation requires SNR 10 dB
Usually also some margin is needed due to fading etc.
SNR depends on received signal power and noise power of the
receiving system
By calculating SNR we get a link budget where the effects of
different factors are easily seen
372
186
Radio link
Received signal power Input noise power of
receiver system
Figure of merit G/T

So SNR is
373
Example: Signal bandwidths
374
187
Example: Satellite-TV connection at 12 GHz
Link budget factors are often considered just like in a budget

of dB values
Budget calculation result: Necessary receiving system

parameters

375
Example
If for a circular dish antenna DA = 0.4 m Gr

= 31.8 dB (using aperture efficiency 0.6)
max. TS = (31.8-2.4) dBK
Thus TS 870 K TR 720 K (when TA =
150 K)
376
188
Note: Effective aperture area

The maximum transmitting or receiving gain
of an antenna with effective aperture area Ae is
where c 3 x 105 km/s.

The value of Ae for a circular dish antenna
equals its physical area (DA/2)2 multiplied by
the aperture efficiency 0.5 0.70.
377
Note: Figure of merit

Gr/Ts is the figure of merit for a receiving
system.
Gr is the receiving antenna gain.
Ts is the system noise temperature.
System noise temperature = antenna noise
temperature + receiver noise temperature
(LNA)
378
189
Communication link
calculations
Communication link
All communications links comprise
a transmitter,
a channel and
a receiver.
A channel might be
a radio or optical link through the atmosphere,
an optical fiber,
a coaxial cable,
a twisted pair cable, or
a telephone wire on a pole.
Links have performance attributes throughput/capacity,
error rates, bandwidth etc.
The type of channel determines the design of the
transmitter and the receiver.
380
190
Shannons model
381
Elements of a communication link
Send messages or information from a

source to a destination
382
191
Classes of transmission media
383
Transmission media
There are basically two types of media, wired or
wireless.
Here, we are dealing with the wireless medium
only.
Electromagnetic waves travel through the earths
atmosphere in the following four ways:
Ground wave propagation
Tropospheric scattering
Line of sight
384
192
Ground wave propagation

Frequencies below 30 MHz propagate along the earths
curvature guided by the surface and are called ground waves.
This guided wave has two main components:
The direct wave, and
The reflected component
This mode is also called ducting.
385
Frequencies between 3 and 50 MHz are
reflected by the ionosphere and can travel
much further than ground waves.
386
193
Tropospheric scattering
Above VHF, we see a phenomena where the signals are
scattered by the troposphere.
The scattered waves are very weak, but they can be received
and demodulated.
This mode of media behavior is called tropospheric scattering.
387
Line of sight
This is the mode for satellite communications and also for
terrestrial microwave links.
For frequencies above 3 GHz to about 10 GHz, the earths
atmosphere offers practically no degradation.
Frequencies above 10 GHz suffer from oxygen and water
vapor absorption and cause degradation.
388
194
Link budgets
The designer hopes that the signal will complete its trip
through link chain with just enough power to be decoded at
the receiver with the desired signal quality.
With digital signals, the quality is measured by the Bit Error
Rate (BER).
If we want our signal to have a low BER, we would start it out
with higher power and then make sure that along the way it
has enough power available at every stage to maintain this
BER.
The signal can get extra power infusion along the way from
intermediate amplifiers such as microwave repeaters for
telephone links or from satellite transponders for satellite links.
389
How we specify the quality of signal

transmission
The BER, as a measure of the signal quality, is the
most important figure of merits in all link budgets.
The BER is a function of a quantity called Eb/N0, the
bit energy per noise-density of the signal.
For a QPSK signal in an Additive White-Gaussian-
Noise (AWGN) channel, the BER is given by
390
195
Erfc
According to the previous formula the BER of
any signal is related to its Eb/N0 by the
complementary error function
Erfc-function is found tabulated in most

telecommunications textbooks and is available
as a built-in function in most math programs.
391
The above equation when plotted has a classic

waterfall shape when plotted on a log-log
scale.
The BER is inversely related to Eb/N0.
Higher Eb/N0 means better quality or smaller
BER.
392
196
BER of a signal
393
Signal power vs. noise power

Instead of monitoring just the signal power we
will actually monitor a ratio of the signal power to
the noise power injected along the way.
This means that in doing link budgets, we will
keep track of not just the signal power and how it
is getting attenuated, but also where the noise is
entering into the link and how much.
This dual role makes things complicated because
there are numerous sources of noise.
394
197
Eb/N0 - a measure for digital links
Eb/N0 is the most common parameter used to

compare communication systems even when
they have differing bit rates, modulations, and
media.
Lets take a closer look at the Eb/N0.
The quantity Eb is a measure of the bit energy.
The energy is the capacity to do work and
energy expended per time is called power.
395
Bit energy
To compute Eb, we divide the average signal power by
its bit rate
This makes sense because the average power is the

energy per unit time, and the bit rate is the number of
bits per unit time.
So the division removes the units of time leaving
energy per bit.
We can also write the above equation in an alternate
amplitude-squared form with the amplitude-squared
representing the Pavg
396
198
Noise density
In the denominator of Eb/N0, the quantity N0 is called
the noise density.
It is the total noise power in the frequency band of the
signal divided by the bandwidth of the signal.
Noise density is measured as Watts/Hz and is the
noise power in one Hz of bandwidth.
where PN = noise power and BN = noise bandwidth.

The units of N0 are Joules.
397
C/N and C/N0 - a measure of analog

links
For analog signals, we use a quantity called C/N0 in the
same way as Eb/N0, where C is the signal power.
C and Eb are related by the bit rate.
So you will typically see C/N0 specified for the analog
portions (or the passband signals) of the link and Eb/N0
for the digital (or the baseband) portions.
C in the expression (C/N) is simply the carrier power in
the whole useable bandwidth, where C in the
expression (C/N0) is carrier power per unit bandwidth.
398
199
Relations between Eb/N0 and C/N0

and C/N
Previously, we know that
C = Energy per bit x bit rate
= Eb x Rb, from which we get
Note both C/N0 and Eb/N0 are densities so we do not need to

specify the bandwidth of the signal.
But to convert (C/N0) to (C/N), need to divide by the signal
bandwidth
N=N0B
399
Link and link equation

A link consists of three parts.
1. Transmitter
2. Receiver
3. Transmission media
The very simplest form of a link equation is
written as
Preceived = Power of the transmitter + Gain of
the transmitting antenna + Gain
of the receiving antenna - Sum of
all losses
400
200
Transmitter
A transmitter receives baseband data,

modulates onto a higher frequency carrier,
amplifies it and broadcasts it via an antenna.
The two main items that are associated with
transmitters are
1. Flux density
2. EIRP
401
Flux density
The flux density is a measure of energy that is available
for gathering from a particular source.
It is sometimes called the Radio Power of a Source.
The flux density is defined by
where G = gain of the transmitting antenna and P =

transmitter power in watts.
The flux density is a measure of the amount of energy that
is received at a distance r from a transmitter of gain G and
transmit power P watts.
402
201
Amplifier + Antenna = Transmitter

A very basic transmitter consists of an amplifier and an antenna
The amplifier puts out a certain amount of power and the antenna is
said to have a particular gain that further amplifies this power.
This combination is here called the transmitter.
Usually, lossy elements such as wires connect these two components
in the preferred direction of radiation of the antenna.
These losses are included in the quoted EIRP figure for the
transmitter.
403
EIRP Effective Isotropic Radiated

Power
EIRP is closely related to the flux density.
Where flux density is energy as measured a distance
away from the source, EIRP is a measure only of the
transmitted power.
For a transmitter EIRP is defined as the combination
of
EIRP = Power of transmitter x Gain of the
antenna
= Pamp x Gantenna
404
Note: EIRP is the numerator of the flux density equation.
202
Note
There are two hidden assumption in EIRP.
First is that the transmitter is putting out the
maximum power that it can, and second, that
the EIRP figure is delivered at the antennas
boresight.
So if you happen to have your antenna pointed
not quite straight into the boresight of the
transmitting antenna then you will not get the
quoted EIRP.
405
Received power
EIRP and the flux density both tells us something
about a transmitter but nothing about what is
actually received.
To compute power received by a receiver at a
distance r from the source, we need to multiply
the flux density with the receiving antennas area.
The flux density is energy per unit area per unit
time.
The only useable part of this energy is what is
accepted by the receiving antenna.
406
203
Effective area
So the power received is equal to the flux density
times the receiving area
The effective receiving area (not actually a physical

area but strongly related to it) of any antenna is
defined by
where GR is the gain of the receiving antenna and is

the wavelength.
407
Power equation
Now we can write the expression for computing the received
power as
This equation says that if we know the gain of the receiving

antenna, the EIRP (= GTPT) of the transmitter, the operating
frequency, and the distance between the two, then we can
calculate the received power.
We can rewrite the above in dB as
408
204
Free Space Loss (FSL)
The last portion of the expression
containing the ratio of the distance r to the

wavelength , i.e. the number of wavelengths
in the distance, is called the Free Space Loss
(FSL).
409
Noise
This topic causes a lot of confusion, particularly when dealing
with link budgets.
As we can see, our important parameters Eb/N0, C/N, C/N0 all
have this noise term on the denominator.
All objects not at absolute zero emit electromagnetic radiation.
The band of frequencies emitted are a function of the
temperature of the object.
E.g. a light bulb emits many different frequencies.
Most of its radiation is in the range of infra-red light and
ultraviolet frequencies which we can see and feel.
410
205
Noise
The sun puts out visible noise in the light wave
frequencies among many others that we can
not see such as X-rays and infra-red.
The noise coming to us from the galaxies is
typically in microwave frequencies.
The statistics of this noise is well described by
quantum physics.
The black body radiation problem was first
solved by Max Planck in 1901.
411
Spectral radiance
According Planck's law
hf
G f , T hf
e kT
1
where G(f,T) is the spectral radiance or the energy per unit
time (or the power) radiated per unit area of emitting surface in
the normal direction per unit solid angle per unit frequency by
a black body at temperature T; h is the Planck constant; k is the
Boltzmann constant; f is the frequency of the electromagnetic
radiation; and T is the absolute temperature of the body.
412
206
hf vs. kT
For radio and microwave frequencies, the factor hf is quite
small relative to the factor kT in the nominal range of room
temperatures, say 290 degrees in Kelvin.
Thus the exponential function in the expression can be
approximated by the first two terms.
413
Thermal noise power

Now we get
We can interpret that this is the power at frequency f, provided

f is small enough such that hf << kT.
Thus it seems appropriate to refer to the low-frequency noise
as thermal noise (it is proportional to the temperature) and the
high-frequency noise, from somewhere sub-optical on up, as
the quantum noise, since it is proportional to Plancks
constant h.
414
207
Note
415
Thermal noise
416
208
Note: Quantum noise

When f becomes large enough that this kT
approximation no longer holds, the frequencies
are in the generalized optical (i.e., infrared,
visual optical, ultraviolet, and X-ray, )
range.
In this frequency range, the higher order hf
terms eventually dominate the frequency-
dependent noise called quantum noise.
417
Noise definition 1
The thermal noise power in a bandwidth B is
kTB.
We can define noise density simply as the
power of the noise signal divided by its
bandwidth
418
209
Noise definition 2
We can define noise density
This definition is consistent with widespread usage,

where T is considered to be the system temperature,
not the ambient or room temperature.
In the radio and microwave bands, the spectral
density is taken as N0 for a one-sided spectrum, and
as N0/2 for a two-sided spectrum.
The noise power will be N0B in all cases where this
convention is applied.
419
Noise power
Lets now set these two definitions for noise
density N0 equal
So, we can write
where k = Boltzmanns constant 1.38 x 10-23

Joules/Kelvin, T in Kelvins and BN in Hertz, and
PN in Watts.
420
210
Noise bandwidth
There is no confusion when talking about bit rates but
unfortunately there are many different ways of defining
bandwidth.
The most common definition of bandwidth is the distance
from one pass-band edge to another where the edge is
defined as the 3 dB below the maximum.
What is the noise bandwidth in the equation for noise?
In simple terms, it is the noise that is allowed to enter into
the system by the receive filter.
The idea of noise bandwidth is to cover all power passed-in
by the filter.
Generally, the noise bandwidth is approximately 1.12 times
the 3 dB bandwidth of the signal.
421
G/T
G/T is a very important parameter of receivers.
The T is the thermal noise temperature of the receiver and it
impacts the ability of the receiver to see a signal in the noise.
Just as we characterize a transmitter by its EIRP, we use G/T in
a similar way to specify receivers.
In dB, G/T is the difference in the gain of the receiving
antenna gain and its noise temperature
where GR = Gain of the receiving antenna and

T = Thermal noise temperature of the receiver
422
211
Note
This handy G/T variable allows us to compare
receiving systems of all kinds.
A G/T of 15 dB is better than 10 dB.
This parameter is usually given for earth stations
as well as for satellite receivers, and does not
need to be calculated.
In doing link budgets, we will assume that it is
given.
Actually calculating the G/T or the Noise Figure
of a receiver is a topic in itself and worthy of the
own story.
423
Losses experienced in LOS links

The losses experienced by the signal fall into
these categories
Free Space Loss (FSL)
Rain
Antenna misalignment
Gaseous absorption
424
212
Computing free space loss

For line of sight links, this loss is a function of the square of
the distance.
For a signal going from ground to the satellite, the free space
loss is largest of all other types of losses.
It can be simplified and written as
where r is the distance and f the frequency.
425
Rain
Signal attenuation due to rain is the second most
significant after free space loss.
It is particularly significant for higher SHF frequencies.
We have to deal with rain losses for both uplinks and
downlinks.
It also varies a great deal from location to location since
it is a function of the rain rate.
The attenuation can vary from 0.1 dB in California to
over 10 dB in Seattle.
Providing for this large attenuation in satellite links
results in over design of the system for areas which
have little rain.
426
213
Countermeasures for rain attenuation
Accommodations are made by providing ground or

site diversity, which just means that there are two or
more receivers instead of one which may or may not be
geographically separated.
Ground diversity such as having another ground station
located a few kilometers away in a rainy region can
reduce the rain attenuation by more than half.
Other ways to accommodate for location-specific rain
attenuation is to allow higher power for the
transmitters and variable error correcting codes and
variable data rates.
427
Rain models
There many popular rain models that help us
compute the rain loss.
E.g., these are
NASA Rain Attenuation Model
CCIR Rain Attenuation Model
The result from each can vary by 1-2 dB
depending on how the regions are defined and
estimate of rain rate is made.
428
214
The antenna gain plays an important role in the link
calculations and we have assumed that the receiving
antenna and the transmitting antenna are oriented perfectly
so that the maximum gain of the receiving antenna is
aligned with the uplink.
The gain varies a great deal off the bore sight and as shown
in the figure below, unless we have perfect alignment, we
are going to have losses associated with this.
There are two parts to this loss.
One is at the transmitter, if its antenna is not pointed to deliver
maximum gain and
the second is at the receiving antenna is not pointed to receive
the maximum gain.
Antenna pointing is an important operation and calibrations
are performed when the system is set up.
429
To receive maximum EIRP, both receiving and
transmitting antenna have to be aligned
perfectly
430
215
Gaseous absorption
The attenuation caused by clouds and fog due

to the phenomena of gaseous absorption has to
be considered particularly for frequencies
above 10 GHz.
These effects are primarily due to amplitude
reduction which reduces signal power, but
water can also affect the polarization of
signals.
431
Combining link power and noise

Now we write the link equation in terms of
C/N.
Carrier power (C) is the sum of the EIRP of
the transmitter, the gain of the receiver and any
associated implementation losses all expressed
in dB
C = EIRP + G Losses
So that is all the power available to the signal.
432
216
(C/N)dB and (C/N0)dB

Remember that
PN N = k T BN =N0BN
In dB form we can write
(C/N)dB = EIRP + G/T - Losses - k - BN
If we convert C/N to C/N0 form
C/N=C/(N0BN) C/N0=(C/N)BN
(C/N0)dB = EIRP + G/T - Losses - k
Note: The bandwidth has dropped out. C/N0 is then a figure that is independent
of bandwidth. However, you will hear much more often about C/N (also referred
to as SNR) that C/N0. 433
Eb/N0
C/N is used for analog signals, and Eb/N0 for

digital data signals.
Conversion to Eb/N0 requires knowledge of the
symbol/bit rate and C/N the knowledge of the
bandwidth.
Summarizing these conversions in dB form
Eb/N0 = C/N0 - Rb
C/N = C/N0 - BN
434
217
Symbol rate and bit rate

Nowadays nearly all the BER curves are given as a function
of the bit energy or Eb/N0.
However, the modem part of the communications systems
operate on the basis of symbol rates and not bit rates.
In a binary case, a symbol consists of just one bit, so the bit
rate of a binary system is equal to the symbol rate.
In a QPSK system, one symbol represents two bits, so here
the bit rate is twice symbol rate.
In 8-PSK system, one symbol stands for three bits, so now
the bit rate is three times the symbol rate.
The advantage is of these higher order modulations is that
higher bit rates can be contained in the same bandwidth as
the binary signal.
435
Higher order modulations

We can make a symbol stand for as many bits
as we want but since we find that the BER
increases faster than the bit rate, which
requires more power to overcome.
If the system is bandwidth-limited and but has
plenty of power, then these higher order
modulations make sense but e.g. for satellites
which are power limited (due to their being
weight-limited), they are not used.
436
218
Coding gain
E.g. in bank transactions over the line to an another
bank, an error-free transmission is wanted.
Typically a data transmission at about 10-12 BER is
considered nearly error free or quasi error-free using
the official term.
To get a picture of 10-12 BER, this is equivalent to about
one error per hundred million pages of text transmitted.
A typical satellite link can provide, for a reasonable
power level, only about 10-3 BER.
This is equal to about ten errors per page.
This may be OK if this a voice signal or music but data
at this error rate is quite unacceptable.
437
BER vs. Eb/N0
438
219
Error correction codes

Nowadays error correction codes (ERC) is nearly
always used on all types of links.
There are many different varieties of ERC
suitable for all different types of link.
In all cases, the use of coding is a trade-off
between power and payload.
The bits used to convey coding cannot be used for
information, effectively reducing the useable bit
rate for a given bandwidth and power level.
439
Simple coding example

Consider e.g. a group of 7 bits.
We assign the last bit as a parity bit.
This means that when the blocks of bits is checked at the
receiver, if the first six bits are even (equal number of 0 and 1)
then, the parity bit is set to 0, and if odd then it is set to 1.
Now we can tell if a received block is erroneous.
So by using a parity bit, we now have a way to tell if the block
has a bit error.
If the first six bit of the received block have an even number of
1s, as in the following figure, we know there has been an error
somewhere.
440
220
Simple coding example

This type of code is called a block code because
it deals with a group of bits at one time.
It is said to have a rate of 6/7, as the code uses 1
out of 7 bits for coding purposes and only 6 out 7
bits can be used for information.
In return for an Eb/N0 of 10 dB, it gives us an
approximate BER of 10-9 instead of the 10-5 we
would have gotten otherwise.
So in return for a loss of about 15% of the bit rate
to coding overhead, we get coarsely 5 dB
improvement in Eb/N0.
441
BER curves
Lets take a look at the
following BER curves.
The right most curve is
for an uncoded QPSK
signal, the other two for
coded links. BER
First thing we see is that
the use of valid coding
moves the BER curve to
the left.
Otherwise there would be
no point to use error Eb/N0(dB)
correcting codes.
442
221
BER curves
These curves say that
we can get the same
BER but for a lesser
Eb/N0.
Of course, our effective BER
data rate is reduced.
But we wont worry
about the reduced data
rate, if we can live with
it.
Eb/N0(dB)
443
Example
We see that for a desired
BER of 10-5, an
uncoded QPSK signal
requires an Eb/N0 of 9.6
dB. BER
Now if we add coding,

for the first code we
need only 4.4 dB and
the second code requires
Eb/N0(dB)
only 3.4 dB.
444
222
Coding gain
Coding gain is the difference between Eb/N0 needed for
an uncoded system and Eb/N0 needed for a coded
system.
We define the coding gain as
Coding Gain = Uncoded Eb/N0 - Coded Eb/N0
for any particular BER.
The coding gain is a function of the BER level chosen.
So before the coding gain calculation we need to know
what BER level is desired for the link.
E.g., a BER of 10-5 can be chosen for voice and 10-10
10-12 for data links.
445
Coding gain
The coded BER curves
shown beside are usually
produced by testing or
simulation.
In scientific literature there
can be found BER curves for
BER
various types of codes.
In most cases this loss in
useful bit rate is accepted for
reduction in transmit power.
This is true now for all types
of links including wireless Eb/N0 (dB)
phones as well.
446
223
Link margin
The goal of the link budget exercise is to have

a reasonable margin for the chosen data rate,
bandwidth and EIRP and the G/T figures.
Often some adjusting and readjusting is needed
to get the desired link margin.
Margin
447
Note
Other factors such as interference from
adjacent channels and adjacent links will affect
the link budgets.
448
224
Microwaves
General description and

definitions
Applications
Microwaves are conventionally encountered in
three kinds of applications:
1. radar, used for detection and measurements;
2. radiocommunications, for point-to-point links,
most particularly for satellite and space
communications;
3. heating, drying, cooking of many different types
of materials.
450
225
Applications
The majority of applications of microwaves are related
to radar and communication systems.
Microwave communication systems handle a large
fraction of the worlds international and other long-haul
telephone, data, and television transmissions.
Most of the current wireless telecommunications
systems, such as terrestrial radio links, satellite
communications , wireless local and personal or body
area networks (WLAN, WPAN, WBAN), cellular
systems, and global positioning satellite (GPS) systems,
operate in the UHF EHF range, and thus rely on
microwave technology.
451
Applications
Information applications are in the field of information
acquisition and transfer.
In radar, microwaves permit acquisition of certain
information, which is contained within the echo signal from a
target.
The location and characteristics of the latter are determined by
comparing the signal received after reflection to the emitted
signal.
In communications, microwaves provide the support to the
information which is to be transmitted from very small to very
large distances.
452
226
Applications
Energy applications utilize a most remarkable

property of microwaves, that is in-depth
heating.
This is an energy transfer application, which
has found a number of uses in everyday life,
from the kitchen to the medical treatment of
various ailments.
453
Microwave field
The microwave field covers quite a wide range of
applications of very diverse nature.
Many different problems are found in the field.
On the fundamental level, basic theories quickly lead
to very complex mathematics.
In rather sharp contrast, many practical applications
are still based on the traditional cut and try
experimental approach or/and simulation.
454
227
Theoretical models
The same variety is encountered in the use of models at different levels of
complexity.
The study of waveguides and cavities can be done by considering the
electrical and magnetic fields, described in terms of Maxwells
equations.
The same macroscopic model is also adequate to consider the interactions
between electromagnetic fields and an electron beam, as they occur in most
generating and amplifying vacuum tubes.
For other active components, such as the solid state sources, the use of
quantum physics is required.
The characterization and measurement of components is done in terms of
circuit theory, using the scattering matrix formalism and Kirchhoffs
laws.
455
What are microwaves?

If you consult Webster you find the word
microwave defined as a very short
electromagnetic wave.
The key word in this definition is short, which
implies length (a wavelength) is involved.
The wavelength () of an electromagnetic
wave is equal to the velocity (v) of that wave
divided by its frequency (f) or =v/f.
456
228
Therefore, the wavelength of an electro-

magnetic wave is inversely proportional to its
frequency; that is, the higher the frequency, the
shorter the wavelength.
So what are microwaves?
In simple terms, they are radio waves of a very
high frequency.
457

Exactly where the microwave spectrum begins
is a matter of some discussion.
There are those who say it begins at 300 MHz
( = 1 m in air) and those who maintain it
starts at 1.0 GHz ( = 30 cm in air).
458
229
Compromise definition
The microwave spectrum should begin where

conventional components (lumped capacitors,
resistors, and coils) no longer exhibit their
proper parameters and where it becomes
practical and advantageous to build
(distributed) components in stripline and
microstrip, as well as in coaxial, form.
459
Skin effect
The origin of the microwave spectrum should be where
conventional components cease to display normal parameters.
But why would such a change occur?
One key reason is a phenomenon known as the skin effect.
At higher frequencies the energy of a wave tends to use only
the outside skin of a conductor for the transferring of energy
because larger and larger flux are set up around a conductor,
increasing its inductance.
By once the inductive reactance (XL) increases when the
frequency does, thereby offering a larger resistance to the
energy.
Thus the high frequency component searches for the path of
lowest resistance on the outer skin of the conductor.
460
230
Skin effect
Because of this phenomenon the energy travels only on the outside
edge of the conductors; lumped components thus become very
unpredictable and impractical at microwave frequencies.
It has been observed that some capacitors actually look inductive at
high frequencies.
The capacitance set up in coils and wire-wound resistors when used
at microwave frequencies completely cancel out the intended
functions of each of these components.
Special materials, usually of a ceramic base, are used in the
production of capacitors and resistors for use at the higher
frequencies.
Coils are usually made of high impedance lines cut to a specific
length.
461
Official definition: Microwave region
The frequency range extending from 300 MHz up

to 300 GHz is generally known as microwaves,
which thus characterize signals having between
300 million and 300 billion periods per second
(from 3 108 to 3 1011 Hz).
These limits are to some extent arbitrary: they
permit to position the microwave domain between
the waves used for conventional radio and TV
diffusion (lower ranges) and infrared rays (higher
ranges).
462
231
UHF, SHF, and EHF decades

A finer designation of the frequency bands divides
them into decades.
The microwave range more or less covers the
three decades called
ultra-high frequencies (UHF),
super-high frequencies (SFH), and
extra-high frequencies (EHF).
The upper ranges of electromagnetic radiation are
known as infrared, visible rays, and ultraviolets.
Microwaves are divided into still narrower bands,
corresponding to specific wave-guide sizes.
463
US MICROWAVE BAND
464
232
Energy of a microwave photon

In quantum physics, electromagnetic radiation
is considered to be a flow of photons of energy
hf, where f is the frequency and h is Plancks
constant ( 6.6310-24 Js).
A microwave photon thus has an energy
located in the range between roughly 1.2 10-6
and 1.2 10-3 eV.
465
1 eV=1e*1V=1.6*10-19 As*1V=1.6*10-19J
Magnitude of the periods and

wavelengths
The period T of a microwave signal, defined as
the inverse of the frequency f is located
between 3 ns (nanoseconds) and 3 ps
(picoseconds).
The wavelength of a microwave signal for an
electromagnetic wave propagating across
space at the velocity of light c0 = 2.997925 ...
108 m/s is defined by = c0/f= c0T,
somewhere between 1 m and 1 mm.
466
233
Remark
The term microwaves, used to define a range
of frequencies, denotes the smallness of the
wavelength encountered, as compared to those
utilized for conventional radio and television.
The use of the comparative prefix micro may
lead to confusions here too: one would expect
microwaves to denote wavelengths in the
micrometer range and not, as is actually the
case, in the millimeter to meter ranges.
467
Subdivisions of the EM spectrum.
468
234
Alternate names: Decimeter,

centimeter, and millimeter waves
The three decade ranges can also be called

decimeter waves (UHF),
centimeter waves (SHF), and
millimeter waves (EHF).
469
Division, nomenclature, and allocation of microwaves
470
235
Frequency allocation
Use of radio frequency bands is regulated by governments.
ITU works on standards for frequency allocation.
471
Dimensional comment
The wavelength of a microwave signal is of the same
order of magnitude as the devices used to produce it
and to transmit it.
It is not possible to assume that devices are merely
dimensionless points in space, as is done in circuit
theory approximations.
Also, the term voltage is not defined in a unique way,
since the electric field does not derive from a scalar
potential.
On the other hand, it is neither possible to assume that
devices become large with respect to wavelength, as is
the case for geometrical optics.
472
236
Wave theory
Microwave problems must be considered in

terms of electric and magnetic fields, as
defined in Maxwells model.
Microwaves actually helped to demonstrate the
validity of this model.
The study and utilization of the microwave
domain required the development of specific
new techniques and devices.
473
Comparable field: Acoustics

Since the velocity of sound in air is of the order of 300 m/s,
roughly one million times smaller than the velocity of light,
wavelengths between 1 mm and 1 m correspond, in
acoustics, to frequencies between 300 Hz and 300 kHz.
This range contains most known acoustic sources, in
particular the human voice and music.
A similarity does therefore exist between microwaves and
acoustics, two fields in which devices have the same size as
wavelengths.
Certain methods developed for acoustical applications can
be directly transposed to microwaves and vice versa.
474
237
Microwaves
Properties of microwaves
Bandwidth
The very wide frequency bands available at microwaves are most
favorable for radio communications.
The rate of transmission of a channel being directly proportional to
its bandwidth, a simple calculation shows that over the 300 MHz to
300 GHz frequency range, 999 times more information can be
transmitted over a specified time period than in all the lower
frequency bands taken together.
As a result, the use of microwaves permits meeting the increasing
need for communications channels.
This particular property is directly related to the signal frequency.
Following the same line of reasoning, even much larger amounts of
information could, in principle, be transmitted over the infrared and
visible spectra, using lasers and optical fiber systems.
476
238
Transparency of the ionosphere
The ionosphere forms a group of ionized

layers of electron plasma, which surround the
earth at altitudes ranging from 50 to 10,000
km.
These ionized layers are produced by the
impact of solar radiation, so that their
parameters (density, height) fluctuate widely
between day and night, exhibiting seasonal
variations and changes related to the solar
activity. 477
Transparency of the ionosphere

The electromagnetic propagation within the ionosphere is similar to that in
a waveguide.
Signals at frequencies lower than 10 - 40 MHz (cut-off frequency) are
partially or totally reflected.
This property is used to realize multiple reflection links at short waves.
Higher frequency signals travel across the ionosphere, but experience
distortion, which decreases with frequency.
Microwave signals, well above the ionospheric cut-off, are hardly affected
at all, at sufficiently small power levels.
For these reasons, microwaves are utilized for satellite communications and
space transmissions.
Radioastronomy deals with microwave radiation emanating from distant
stars, galaxies, and quasars.
A quasi-stellar radio source ("quasar") is a very energetic, distant, and compact
region in the center of a massive galaxy surrounding its central supermassive black
478
hole.
239
Partial transparency of the

atmosphere
The various atmospheric components (oxygen,
nitrogen, water vapor, carbon dioxide) and the
many elements in suspension (water droplets,
ice crystals, dust, smoke) do not significantly
affect signals having frequencies smaller than
about 10 GHz.
Higher frequency signals experience several
unwanted effects: absorption, depolarization,
and scintillation.
479
Electromagnetic noise
The noise picked up by an antenna directed skyward, in the absence of
signal, goes through a relatively flat minimum over the 1 to 10 GHz
frequency range.
The received noise power is the product of the equivalent noise temperature
(in Kelvin) by Boltzmanns constant (kB =1.3804 10-23 J/K) and by the
receivers bandwidth.
Over the 1-10 GHz band, the noise temperature decreases below 10 Kelvin.
In practical terms, this means that it is within this frequency range that
signals of the lowest amplitude can be detected, and thus the most sensitive
receivers can be designed.
For instance, signals of extremely low levels received after transmission
across planetary space are often in the neighborhood of 3 GHz.
A further requirement for the receiver is not to produce too much noise,
which would degrade the input signal-to-noise ratio.
480
240
Directivity of antennas
The width of the beam radiated by an antenna is directly
proportional to the ratio of the wavelength to the antennas
largest dimension.
When transmitting a signal from one point to another
(microwave link), or when determining the origin of a
reflection (radar), a narrow beamwidth is required.
It is then either necessary to have a large size antenna, which is
often not convenient for mechanical reasons, or to utilize a
signal of high frequency.
Microwaves are well suited for such applications.
481
Note
Even narrower beamwidths are obtained using
signals in the visible range (laser).
As the beam is then extremely narrow, it must
be directed quite accurately towards the
detector, and pointing problems may become
significant.
482
241
Reflection on targets
The effective reflection area of an object depends in a
very sensitive manner on the ratio of the objects size
to the wavelength.
When the reflecting element is much smaller than the
wavelength, the reflection becomes vanishingly
small.
On the other hand, when the wavelength becomes
much smaller than the object, the effective reflection
area for a metallic element is approximately its cross
section transverse to the beam.
483
Remark
Centimeter waves thus detect objects of meter
size, but are not affected by raindrops.
The latter can, on the other hand, perturb the
detection at millimeter waves.
484
242
Interaction with matter
When an electromagnetic wave impinges on a

material sample, it is preferentially absorbed at
particular frequencies, which are the resonant
frequencies of the material.
The resonances observed at microwaves
depend on the molecular composition of
matter.
This effect is put to good use in chemical and
physical analyses.
485
Interaction with matter

In particular, water strongly absorbs all
microwaves, a specific property which leads to
two types of applications:
1. microwave heating, utilized for the cooking of
food, the drying and thermal processing of
numerous materials, and the medical treatment of a
number of ailments by hyperthermia;
2. the detection and measurement of moisture
contained within materials.
486
243
Non-ionizing radiation
The energy of a molecular bond is larger, by

several orders of magnitude, than the energy
belonging to a microwave photon.
This means that a photon, at microwave
frequencies, does not possess sufficient energy
to break a chemical link, for instance by
photoelectric effect.
Microwaves are thus a non-ionizing form of
radiation.
487
Note
At frequencies far above microwaves, a photon
can have enough energy to extract an electron
and produce ionization.
This happens in the visible spectrum,
ultraviolet (suntan), X and gamma rays.
488
244
Stable oscillation frequencies
The most stable known atomic oscillators,

hydrogen, cesium, and rubidium exhibit
extremely stable oscillations within the
microwave range.
As a result, all atomic clocks and frequency
standards make use of microwaves.
489
Note
The set of properties listed previously makes
microwaves a privileged field for a large
number of applications, such as satellite
communications and radar, but also heating
and measurement.
Some of these applications simply could not
exist without microwaves.
For others, microwaves provide the best
compromise among various requirements to be
satisfied simultaneously.
490
245
Microwaves
Historical landmarks
First experiments in radio

communications
The spark generators used by Hertz and
Marconi for their first radio transmissions in
1888 and 1890 were non-coherent emitters,
generating a very broad spectrum of noise
extending well into the microwave range.
492
246
First waveguide
In 1894, Oliver Lodge surrounded a spark

generator with a metal tube and noted that the
resulting radiated signal exhibited directional
properties.
This effect was at the time considered a
laboratory curiosity, and it did not lead to
practical applications.
With the advent of vacuum tubes, the further
development of radio was hence oriented
towards lower frequencies. 493
Communications experiments
The first practical experiments carried out with
the purpose of using microwaves for the
transmission of information are credited to
George Southworth, at the Bell Telephone
Laboratories, in the US during the 1920s and
1930s.
Wave propagation along copper water pipes
was investigated.
Since then, users of microwaves have often been
nicknamed plumbers.
494
247
Development of radar
The advent of microwave techniques is quite

closely linked to the development of radar
during World War II.
The basic principle, already outlined by Robert
Watson-Watt around 1930, makes use of the
electromagnetic echo produced by a target.
495
A short signal pulse is launched, the time
between its departure and the return of a
reflected signal is measured.
To ensure an adequate detection of targets,
signals of increasingly higher frequencies were
found to be needed.
The magnetron was developed for this
purpose, being the first tube having a high
power generating capability at microwaves.
496
248
The design and the industrial fabrication of radar
systems started around 1940.
Radar played a significant role during World War II,
among others in the well known Battle of Britain.
Some of the research and development work was de-
classified at the end of the war.
Particularly worth mentioning is the 25-volume set
published by the Massachusetts Institute of Technology
(Radiation Laboratory Series), covering all aspects of radar
design, including the foundations of microwaves.
497
Development of microwave tubes

Microwave generators were at first vacuum
tubes, specifically designed for use within
radar systems.
The magnetron is based on the interactions of
an electron flow within crossed electric and
magnetic fields, an effect studied since the
1920s.
It became operative at the beginning of World
War II.
498
249
Development of microwave tubes
The klystron was invented in 1935 by Russell

and Sigurd Varian.
Many different kinds of microwave generators
were developed later on, the most significant
ones being
the backward-wave oscillator (BWO) and
the traveling wave tube (TWT).
499
Ferrite devices
The first non-reciprocal linear passive device
appeared in 1956, a gyrator.
The many isolators and circulators developed since
then are mostly devices for protection, decoupling,
and control.
They are nowadays found in most microwave
systems.
An important property of a gyrator is that it inverts the current-voltage characteristic of
an electrical component or network.
An isolator is a two-port device that transmits microwave or radio frequency power in
one direction only.
A circulator is a passive non-reciprocal three - or four-port device, in which a microwave
or radio frequency signal entering any port is transmitted to the next port in rotation only.
500
250
Satellite communications
The first satellite, Sputnik I, was placed in orbit on
October 4 1957.
In 1962 Telstar was launched, the first
communications satellite placed in a low earth orbit.
Three years later, in 1965, the satellite Early Bird was
placed in a geostationary orbit (remaining over a
fixed location on the equator).
Since then, successive generations of satellites are
playing an important role in communications, mostly
for international links, but also within domestic
networks.
501
Microwave solid state two-ports

Active semiconductor devices appeared on the market
in the 1960s, replacing vacuum tubes as microwave
sources for low and medium power levels.
The first device of this kind developed was the Gunn
diode, based on a physical phenomenon discovered
by J.B Gunn in 1962.
Diode generators and amplifiers were increasingly
being replaced, since the end of the 1970s, by
transistors, either bipolar or MESFETs
(MetalsemiconductorFET).
502
251
Microwave printed circuits
Over the same time span, metallic waveguides

are often superseded by more compact circuits
in microstrip, slot line, or coplanar line,
realized by means of printed circuit
techniques.
503
Maxwells equations
Local and global form

versions
252
Maxwells equations
Maxwells equations govern the propagation of
electromagnetic waves in various media, including
free space,
waveguides,
optical fibers, and
optical crystals.
All
electric,
magnetic,
electromagnetic, and
optical phenomena
are governed by the same fundamental laws of
electromagnetism.
505
Gauss' Law (M I)
Gauss' Law says how the electric field behaves around electric charges.
Gauss' Law can be written in terms of the electric flux density (D) and the
electric charge density (v) as:
The symbol is the divergence operator.

This local form equation is valid at any point in space.
Thus, if there exists electric charge somewhere, then the divergence of D at
that point is nonzero, otherwise it is equal to zero.
The divergence of the D field over any region (volume) of space is exactly
equal to the net amount of charge in that region.
506
253
Note: Divergence ()
Divergence at a point (x,y,z) is the measure of the vector flow out of a
surface surrounding that point.
Divergence is a specific measure of how fast the vector field is changing in
the x, y, and z directions.
If a vector function A is given by:
Then the divergence of A is the sum of how fast the vector function is
changing:
507
Example
Consider the vector function A
Thus divergence of A at any point (x,y,z) in the space is
If we want to know the divergence at (x,y,z)=(1,2,3) then we can calculate

the divergence of A at this point is 2+6*3 = 20.
We can find the divergence at any point in space because we know the
functions defining the vector A, and then calculated the derivates (the rate
of changes).
508
254
M I (Gauss Law)
Gauss' law says that electric field lines diverge

away from electric charges.
This means that positive charge acts as a source of
electric field.
Gauss' law means that negative charge acts as a
sink for electric field.
This all means electric field lines start and stop on
electric charges.
509
M II (Gauss Law for magnetism)

Gauss' magnetism law states that the divergence of
the magnetic flux density (B) is zero.
Since B and the magnetic field H are related by the

permeability , we note that the divergence of the
magnetic field is also zero.
The divergence of the B or H field is always zero

through any volume.
510
255
M II (Gauss Law for magnetism)
The M II says that magnetic charge (magnetic

monopoles) do not exist.
This equation states that the magnetic field
tends to wrap around things - since the
divergence is zero the fields tend to form
closed loops.
511
M III (Faradays Law)
Faraday's law says that a time-varying

magnetic field will generate a circulating
electric field.
This means we have two ways of generating
electric fields
from electric charges (or flowing electric charge,
current) or
from a time-varying (changing) magnetic field.
512
256
Note: Curl ()
The curl is a measure of the rotation of a vector field.
Let us take a vector field, A(x,y,z), we can write A as:
The curl of A is defined to be:
513
M IV (generalized Amperes Law)

Ampere's law says that a flowing electric current gives rise to
a magnetic field that circles the wire.
In addition to this, the generalized Amperes law also says that
a time-varying electric field gives rise to a magnetic field that
encircles the electric field - this is the displacement current
term (also called Maxwells addition to Amperes law).
This means there are two ways to generate a circulating
(solenoidal) magnetic field
a flowing electric current or
a changing electric field.
514
257
Local (point) form versions

Maxwell's equations can be rewritten with only E and H
present.
This means we express D, B and J using E, H and E.
We can do this by using the material equations:
515
Global form version

If the local form of Maxwell's equations are true at
every point, then we can integrate them over any volume
(V) or through any surface and they will still be true.
Total charge inside the
volume V with the (closed)
boundary surface S
Elementary vector dS is ortogonal to the surface S and |dS| is the area element.
516
Elementary vector dL is parallel to the closed loop L and |dL| the length element.
258
Maxwells equations
517
Note
The global form version can be derived from
the local form Maxwells equations using the
following theorems:
Divergence
theorem
Stokes
theorem
Note: These theorems are valid with an arbitrary vector field A. 518
259
Example
Derive the global form version from the local

form Gauss laws (M I and M II) using the
Divergence theorem:
A surface integral

A dS

S
519
Solution
Derive the global form version (surface integrals) from the
local form Gauss laws (M I and M II) using the Divergence
theorem (Gauss theorem):
Thus
520
260
Solution
Divergence theorem (Gauss theorem):
Thus
521
Example
Derive the global form version (line integrals) from
the local form Faradays law and Amperes law (M III
and M IV) using the Stokes theorem:
522
261
Solution
Stokes theorem:
523
Solution
Stokes theorem:
524
262
Let's look at the M III and M IV equations that govern all of
electric and magnetic field propagation:
Faraday's law says that a time-varying magnetic field will

give rise to a circulating electric field.
In practice, fields do not continually grow or continually
shrink - they oscillate about an average.
So the magnetic field oscillates - which means that also the
electric field oscillates.
525
Ampere's law says that a time-varying electric field generates a magnetic
field that encircles the electric field.
In the same way, the electric field will be oscillating in time, the encircling
magnetic field will be oscillating in time as well.
So, a time-varying electric field generates a time-varying magnetic field
and vice versa.
This continuous coupling between these two fields produces the
propagation of electromagnetic waves.
All electromagnetic waves (radiation) radio waves, visible light, x-rays
, consist solely of propagating electromagnetic fields.
From these two basic Maxwells equations (M III, M IV) we can determine
how electromagnetic waves propagate through medium and
how they interact with materials.
526
263
Origin of an electromagnetic wave
527
Maxwells equations
Complex formalism and

phasors
264
Linear systems and

eigenfunctions
In electromagnetics, we often deals with monochromatic
waves, which serves as a carrier of information in
electromagnetic communication.
The field vectors of monochromatic waves are sinusoids of
time.
More complex waveforms can always be developed over a
finite or infinite sum of sine waves.
The selection of sine waves as the basis for decomposition
results from the fact that, in linear systems the eigenfunctions
are sine waves, which move along within the system without
deformation, even though their amplitude and phase do vary
with position.
529
Phasor
Nonsidusoidal signals change shape while propagating in
dispersive systems, either through amplitude distortion or
through phase distortion.
For simplifying the algebraic manipulations, sine waves are
often represented in complex notation by means of phasors
and phasor-vectors.
To illustrate this complex formalism, we consider simple
functions
where is the angular frequency (in units of radian per second), |X|
and X are the amplitudes, and is the phase.
530
265
Phasor
If we define the complex amplitude of x(t) by
then we can write
In complex-function formalism, we will use the complex

function
instead of the real-function
531
Phasor transformation
When this happens it is always understood that we mean the
real part of x(t), which is x(t).
In linear operations the replacement of the real form x(t) by the
complex form x(t) poses no problems, when it is understood
that the final result is the real part of the complex form result.
Actually, we can reduce algebraic manipulations with phasor
transformation:
532
266
Example
Some transformation pairs:
533
Note
x(t) is called an instantaneous value and it is
time-function
Its complex counterpart x(t) is also time-

function.
However, phasor X is not time-function.
534
267
Phasor-vector
Correspondingly to the previous phasor

representation, the actual physical vector field
X(r,t), function of space and time, is related to
the corresponding phasor-vector X(r) by the
relationship
where vector r represents the position in space, t

the time and the angular frequency.
The modulus of the phasor-vector X(r) is the peak
value of the corresponding physical field.
535
Non-linear operation
In operations that involve product of
sinusoidal time-functions or monochromatic
field vectors such as powers, energies, energy
densities, we must use the real form of
functions and vectors, because generally
536
268
Example
The product of two sinusoidal functions

x(t)y(t) can be calculated in real form or using
complex time-functions or using the special
expression of phasors
where the asterisk superscript indicates the

complex conjugate.
537
Time-average
Occasionally in electromagnetics, we have to deal with waves
with extremely short periods (T=/c<10-12s).
Most detectors are unable to respond instantaneously at such a
rapid change, but they can find the time-averaged values.
It is often necessary to find the time average of the product of
two sinusoidal functions of the same frequency:
538
269
Example
Convert Telegraphers equations
to phasor form.
539
Solution
Let us replace u(x,t) with a corresponding
rotating phasor u(x,t) in both equations
Lets do the same operation for i(x,t)
540
270
Solution
Lets take partial derivates from rotating

phasors u(x,t) and i(x,t)
541
Solution
Lets substitute phasor partial derivates to the
original Telegraphers equations
542
271
Note
If we choose a sine function as a base

function
543
Sinusoidal waves, phasor-vectors

The theoretical developments presented here
consider exclusively time-dependent
sinusoidal wave signals, represented in
complex notation by means of phasor-vectors.
More complex signal patterns can, in principle,
always be developed over a finite or infinite
sum of sinusoidal waves (Fourier series,
Fourier transform).
544
272
Why sinusoidals?
The selection of sinusoidal waves as the basis for
decomposition results from the fact that, in all
systems of linear equations the eigenfunctions are
sine waves, which move along within the system
without deformation, even though their amplitude and
phase do vary with position.
In every dispersive system, nonsinusoidal signals, on
the contrary, change shape while propagating, either
through amplitude distortion or through phase
distortion.
545
X(r,t) and X(r)

The actual physical field X(r,t), function of space and
time, is related to the corresponding phasor-vector
X(r) by the relationship
where vector r represents the position in space, t

the time and =2 the angular frequency
corresponding to frequency .
Note: Underlining means that a quantity is complex (phasor or phasor-vector). Bold face
italics denote vectors. The modulus of the phasor-vector X(r) is the effective value of the
corresponding physical field. 546
273
Maxwells equations in complex

notation
Maxwells equations, in term of phasor-vectors in a
linear, homogeneous, possibly lossy, isotropic
medium are given by
where
547
Phasor-quantities
548
274
Note
The phasor form or the time-harmonic form

of Maxwell's equations is perfectly
legitimate, because
this form tells us how the waves behave if they are
oscillating at frequency f, and
all waves can be decomposed into the sum of
simple oscillating waves.
549
Maxwells equations
Material parameters and

boundary conditions
275
Parameters of the medium
The terms r and r are respectively the

relative permittivity and permeability of the
medium.
The loss tangent tan = / characterizes the
attenuation of the electric field in a lossy
material.
551
Perfect electric conductor

In many metals and semiconductors, the current
density J is many orders of magnitude larger than the
displacement current D/t.
It is then often assumed that, in the first
approximation, , the conductor being then a
perfect electric conductor (pec).
However, for physical reasons the current density J
cannot become infinite: the above condition means
that the electric field E must vanish within a pec.
552
276
Perfect magnetic conductor
Similarly, in ferromagnetic materials the

permeability becomes quite large, and is
often assumed to be infinite, as a first-order
approximation, the conductor being then a
perfect magnetic conductor (pmc).
In this instance, the induction field B cannot
become infinite, so that the magnetic field H
must vanish within a pmc.
553
Boundary conditions
On the interface separating two different materials,
none of which is a perfect electric conductor, the
tangential components of the electric and of the
magnetic field are continuous, as expressed by
where n is the unit vector normal to the interface,

directed from medium 2 towards medium 1, and
where indices 1 and 2 specify, respectively, the
medium within which the field is defined.
554
277
Boundary conditions
When 0, the boundary conditions for the

normal components of the phasor-vectors are
automatically satisfied when equations
are met.
555
Perfect electric conductor

On the surface of a perfect electric conductor
(pec, =), the electric phasor-vector must
meet the condition
This condition is approximately satisfied on a

metallic surface (short-circuit).
556
278
Perfect magnetic conductor
On the surface of a perfect magnetic conductor

(pmc, =) without surface current, the
magnetic phasor-vector must meet the
condition
Magnetic materials no longer have, at

microwaves, a sufficiently large permeability
to satisfy the above condition.
557
Boundary conditions on the

normal components
Boundary conditions on the normal components of the electric
field yield:
Surface charges s can only be found on the interface between

two materials when at least one of the two has a non-zero
conductivity .
Similarly, the magnetic field must satisfy
558
279
Note: Magnetic field in a perfect

electric conductor
Since the electric field E must vanish inside of
a perfect electric conductor, it follows from
that the magnetic field must also vanish there

when 0.
559
Constants for vacuum

In vacuum, and approximately in air, the electric
displacement D is related to the electric field E by the
electric constant 0
D 1 As
0 109
E 36 Vm
560
280
Constants for vacuum

Similarly, the magnetic constant 0 is the ratio
of the magnetic induction B by the magnetic
field H in all nonmagnetic materials
561
The velocity of light

The velocity of light in vacuum c0 is related to
the product of those two constants by
562
281
The characteristic impedance
The characteristic impedance of vacuum Z0 is

defined as:
563
Continuity for fields at boundary

At first let us consider the Faradays law (M III) in the integral
form
applied to a path formed by moving distance l along one side

of the boundary between any two materials, and returning on
the other side, an infinitesimal distance into the second
medium.
564
282
Tangential components of electric

field
The line integral of electric field is
Since the path (infinitesimal loop) is an infinitesimal distance

on either side of the boundary, its area is zero and therefore the
contribution from changing magnetic flux is zero so long as
rate of change of magnetic flux density is finite.
When
565
Tangential components of electric

field
Consequently,
Thus tangential (parallel with the boundary)

components of electric field are continuous across the
boundary surface or
where n is unit normal (vector) to boundary surface.
566
283
Tangential components of magnetic

field
Similarly, the generalized Amperes law in global form may be
applied to a like path with its two sides on the two sides of the
boundary as in the previous case.
The line integral of magnetic field (M IV) is

When
Since the path (infinitesimal loop) is an infinitesimal distance

on either side of the boundary, its (loop) area is zero and
therefore the contribution from changing magnetic flux is zero
so long as rate of change of magnetic flux density is finite.
567
Tangential components of magnetic

field
Thus
((Providing that J
vector is finite.))
Thus tangential components of electric and magnetic

field must be equal on the two sides of any boundary
between physically real media.
568
284
Normal components
If two very small elements of area S are considered, one on either side of
the boundary between any two materials, with a surface charge density p
existing on the boundary, the application of Gauss law (M I)
to this elemental volume
569
Normal components (cont.)

gives
That is, for a charge-free boundary, normal components of electric flux

density are continuous; for a boundary with charges, they are discontinuous
by the amount of the surface charge density p.
With no magnetic charge term on the RHS of M II
a development corresponding to the above shows that always the magnetic

flux density is continuous:
Bn1-Bn2=0 570
285
Boundary between different

materials
571
Isotropy
Inside the great majority of materials utilized
in practical applications, phasor vectors D and
E are parallel to one another, or colinear.
The same occurs for B and H.
Such a material is then called isotropic.
The permittivity and the permeability are
both scalar quantities.
572
286
Anisotropy
For a family of magnetic dielectrics called ferrites,
the phasor-vectors B and H of a microwave signal are
not parallel.
These are anisotropic materials, for which the
permeability becomes a tensor, represented by a 3
3 matrix.
11 12 13 H x
B H 21 22 23 H y
31 32 33 H z
Within certain anisotropic materials, the reciprocity

is no longer applicable.
573
Note: Reciprocal networks

If a network does not contain non-reciprocal devices or materials* (i.e.
ferrites, or active devices), then the network is reciprocal.
Zij Z ji Y ij Y ji Note: The inverse

of a symmetric
matrix is symmetric.
Z and Y are symmetric
A reciprocal material is one that has reciprocal permittivity and permeability

tensors.
A reciprocal device is one that is made from reciprocal materials.
Example of a nonreciprocal material: ferrite
This material is very useful for making isolators and circulators.
574
287
Example: Reciprocal networks

We can show that the equivalent circuits for reciprocal two-port
networks are:
Z11 Z 21 Z 22 Z 21
T-equivalent Z 21
Y21
-equivalent Y11 Y21 Y22 Y21
575
Note: Reciprocal materials

D E
B
Dx xx xy xz Ex Bx xx xy xz x
D
yy yz E y B
y yx yy yz y
y yx
Dz zx zy zz Ez Bz zx zy zz z
Reciprocal: ij ji , ij ji
j 0
0
Ferrite: 0 j is not symmetric!
0 0 1
576
288
577
Example: Sourceless, homogeneous,

isotropic and linear dielectric region
578
289
Example: Eliminating magnetic field

vector
579
Example: Wave equation
580
290
Example: Cartesian form
581
Example: Cartesian component form
582
291
Example: One-dimensional wave

equation
583
Note
584
292
Maxwells equations
Dynamics Quasistatics
Maxwells equations in free space
586
293
Physical quantities and Units

Quantity Unit/Value
587
Electrostatics and magnetostatics

Suppose we restrict ourselves to time-independent situations (i.e.
nothing is varying with time everything is static).
We get two sets of equations for electric and magnetic fields that are
completely independent and uncoupled:
Equations of Electrostatics Equations of Magnetostatics
Electric fields are produced by Magnetic fields are produced by

only electric charges. only electric currents.
In electrostatics problems one In magnetostatics problems one
needs to determine electric field needs to determine magnetic field
given some charge distribution. given some current distribution.
588
294
Slow changes
The restriction to completely time-independent situations is too
limiting and often unnecessary.
What if things are changing slowly in time.
Allowing for slow time variations, one often uses the equations of
electroquasistatics and magnetoquasistatics.
Equations of Electroquasistatics Equations of Magnetoquasistatics
Electric fields are produced by Magnetic fields are produced by

only electric charges. only electric currents.
Once the electric field is Once the magnetic field is
determined, the magnetic field can determined, the electric field can be
be found by the last equation. found by the last equation.
589
How slow is slow enough ?

Electromagnetic waves and signals move at the
speed c (speed of light)
Time variations are considered slow if the time
scales over which things are changing are
much longer compared to the time taken by
light to cover distances equal to the length
scales of the problem.
590
295
Example
Lets take an amplifier chip (max. dimension=3 cm) operating
from 100 MHz to 10 GHz. Is quasistatics analysis valid ?
100 MHz operation
Time scale of the problem = 1/(100 MHz) = 10 ns
Length scale of the problem = 3 cm
Time taken by light to travel 3 cm = 0.1 ns
Since 10 ns >> 0.1 ns, quasistatics is a valid means of analysis
at 100 MHz
10 GHz operation
Time scale of the problem = 1/(10 GHz) = 0.1 ns
Length scale of the problem = 3 cm
Time taken by light to travel 3 cm = 0.1 ns
Quasistatics is not a valid means of analysis at 10 GHz 591
Validity scale of quasistatic

analysis
Electromagnetic wave frequency f and wavelength are related to the
speed of the wave c by the relation:
f=c
Let: L = length scale of the problem
T = time scale of the problem 1/f
Condition for quasitatic analysis to be valid:
T>>L/c
cT>>L
c/f >>L
>>L
Quasistatic analysis is valid if the wavelength of electromagnetic wave at
the frequency of interest is much longer than the length scales involved in
the problem.
592
296
Polarization states of plane

waves
Linear, circular, and
elliptical polarization
Polarization of a field
The polarization of any vector field is defined by the
geometrical locus of the tip of the vector, as it varies during
one period T.
Equation
may be written in the following form:
This is actually the equation of an ellipse having two conjugate

half-axes X(r,0) and X(r,T/4), which defines the plane of the
ellipse as well as its dimension.
594
297
In the most general case, therefore, a field is

elliptically polarized.
There are two limiting cases:
1. when the two vectors are directed parallel to
each other, or when one of the two vanishes, the
ellipse shrinks to a straight line.
The field is then linearly polarized, and the condition
is satisfied.
595
The asterisk * denotes the complex conjugate.
2. when the two vectors have the same length and
are perpendicular to each other, the ellipse
becomes a circle, and the field is circularly
polarized.
This occurs when
and the field may be described by
where e0 and eT/4 are unit vectors along the two

directions X(r,0), and X(r,T/4), respectively.
596
298
An elliptical A linear
polarization polarization
A circular
polarization
597
Comment
Any field may be expressed by the sum of two
linearly polarized fields: this is actually done in
Any field may also be decomposed into two

circularly polarized fields, rotating in opposite
directions.
The long half-axis of the ellipse is then the sum of the two
radii, and the short axis their difference (which vanishes for
a linear polarization).
598
299
Example: Circularly polarized plane waves
Note: The x- and y-components of the E-field have the same amplitude but
599
are 90-degrees out of phase.
Note: The E-field at z = 0 never goes to zero but keeps rotating in a

600
circular trajectory.
300
So what is happening in space when wave proceeds?

If one takes a snapshot of a circularly polarized wave at any instant, then he
will see the picture shown below
The E-field vector does not change in magnitude but its direction twists
in space.
An observer sitting in the path of the wave will see the E-field vector rotate
in a circular trajectory at his location as the wave passes by.
Note: The H-field at each point is orthogonal to the E-field. 601
Example: Linearly vs. circularly polarized

plane waves
602
301
Handedness of circularly polarized plane

waves
Right-hand circular polarization Left-hand circular polarization
The wave is right-hand circularly The wave is left-hand circularly
polarized if at any location in space polarized if at any location in space
when the thumb of the right hand when the thumb of the left hand
points in the direction of points in the direction of
propagation, the fingers curl in the propagation, the fingers curl in the
direction of rotation of the E-field direction of rotation of the E-field
vector in a plane perpendicular to vector in a plane perpendicular to
the direction of propagation. the direction of propagation.
603
Example: Elliptically polarized plane waves
604
302
Example: Special cases
605
606
303
607
Example: Linear and/or circular?

A circularly polarized wave is a linear superposition of two linearly
polarized waves
Similarly, a linearly polarized wave is a linear superposition of two

circularly polarized waves
Any arbitrary polarization state can be expressed as a linear superposition

of appropriate basis states chosen from either linear or circular basis.
608
304
Comment
609
Application of Maxwells
equations
Wave propagation
305
Maxwells equations
Let us consider a region in space containing no sources so
there will be no charge and current terms in Maxwells
equations.
Arbitrary time variations of the fields will be permitted and the
medium is to be considered homogeneous, isotropic, linear,
and with zero conductivity.
So, we have equations in form
611
Separation of fields
To attempt a solution of a group of simultaneous equations, it is usually a
good plan to separate the various functions of position, such as D and B
here, to arrive at equations that give the distributions of each.
Thus, lets try to separate E and H to the own expressions.
First, taking the curl of M III, we see that
Time and space partial

derivatives may be taken
in any order.
612
306
Wave equation
The left side may be expanded, using a vector identity, to give
A source-free region has been specified.

Thus
which is the general form of the wave equation for the

medium considered.
Note: This form applies as well to the magnetic field, as may seen by a similar derivation started
613
by taking the curl of M IV.
Wave equation
From this simple special case many of the characteristics of
electromagnetic waves can be found.
Recalling that in rectangular coordinates the Laplacian of a
vector can be separated into the vector sum of the Laplacian of
the components, we have in Cartesian form
614
307
Solution of the one-dimensional

wave equation
For the case of variation in the z direction only, Laplacian
reduces to the form of the one-dimensional wave equation.
It was shown there that the equation has a general solution in

the form
Where F1 and F2 are arbitrary functions, which have

corresponding derivates (F, F) and
615
Uniform plane waves

The first term of the solution represents the
wave or function F1 traveling with velocity v
and unchanging form in the positive z
direction.
The second term represents the wave or
function F2 traveling with velocity v and
unchanging form in the negative z direction.
These functions F1 and F2 are called uniform
plane waves.
Note: Radio waves at some distance from the antenna and the ground are approximately of this
616
simple form with space variations in one direction only.
308
Equations
Poyntings theorem and
energy relations in an
electromagnetic field
Energy transfer through

electromagnetic fields
The simple transmission line waves can transfer energy
from one point to another.
Energy transfer may also be accomplished through more
general types of electromagnetic waves, the amount of the
energy depending on the magnitudes, distribution, and
phases of the electric and magnetic fields of the wave.
Let us take a region in which permittivity and
permeability may be functions of position but not of time.
Maxwells equations, written in terms of the total fields,
currents, and charges of a region, describe the
electromagnetic behavior of the region.
618
309
Derivation of Poyntings theorem

Let us take a
Using a general equivalence of vector operations, we get
Next we substitute M III and M IV
to the previous equivalence
619
Derivation of Poyntings theorem
Now we integrate the previous equation over the volume V
From the Gauss theorem, the volume integral of div(E x H)

equals the surface integral of (E x H) over the boundary.
620
Note: All points of the closed surface A are on the boundary surface of the volume V.
310
Note: Gauss theorem (Divergence

theorem)
A surface integral

A dS

S
621
Poyntings theorem
This form is valid for general media, but in the common case
of linear, isotropic, time-invariant materials, material
parameters are scalars independent of time.
Then
B H

D E
622
311
Interpretation of Poyntings theorem
The term H2/2 represents the energy storage per unit

volume for a static magnetic field.
Energy density
of magnetic field
If this interpretation is extended by definition to any

magnetic field, the first term of volume integral
represents the time rate of increase of the stored
energy in the magnetic fields of the region.
Vs
Henry Vs
A
m m Am 623
The term E2/2 represents the energy storage

per unit volume for a static electric field.
Energy density
of electric field
Similarly thinking as above, the second term of

volume integral represents the time rate of
increase of the stored energy in the electric
fields of the region.
As
Faraday V As

m m Vm 624
312
The third term is the usual ohmic term and so represents

energy dissipated in heat per unit time in medium.
All the net energy term must have been supplied externally.
Thus the term on the right (surface integral) represents the
energy flow into the volume per unit time.
Changing sign, the rate of energy flow out through the
enclosing surface is
V A W J
S E H m m m 2
sm 2
where

P S A W J
m2 W
m2 s
is called Poyntings vector.
625
Poyntings vector
It is convenient to think of the Poyntings
vector
as the vector giving direction and magnitude of

energy flow at any point in space.
V A W J
S E H m m m 2
sm 2

Note:
P S A W m 2 W J

m2 s
626
313
Poyntings vector
If E and H
are phasors (complex vectors), then
An average power flow is expressed by
627
Example
Let it be
where A unit vector parallel with the

direction of propagation
628
314
equations
Plane waves
Plane wave approximation

Plane waves are good approximations to real waves in many
practical situations.
Radio waves at large distances from the transmitter, or from
diffracting objects, have negligible curvature and are well
represented by plane waves.
Much of optics utilizes the plane wave approximation.
More complicated electromagnetic wave patterns can be
considered as superposition of plane waves, so in this sense
the plane waves are basic building blocks for all wave
problems.
Even when that approach is not followed, the basic ideas of
propagation, reflection, and refraction, which are met simply
here, help the understanding of other wave problems.
630
315
Uniform plane waves in a perfect

dielectric (free-space)
We restrict our attention to media for which and are scalar
constants.
For a uniform plane wave, variations in two directions, say x
and y, are assumed to be zero.
Maxwells curl equation in rectangular coordinates
631
Maxwells curl equation in rectangular

coordinates
Maxwells curl equation in rectangular coordinates reduce to
We see first from (3) and (6) that Ez and Hz must both be zero, except
possibly for constant (static) parts which are not of interest to us in the
wave solution.
That is, the electric and magnetic fields of this simple wave are both
transverse to the direction of propagation.
Note:
632
316
Wave equation and its solution

Next, if (2) is differentiated with respect to z, (4) with respect
to t, and the two results combined, we obtain the one-
dimensional form of the wave equation in Ex
The solution represents wave traveling with velocity v in the

positive z direction and another traveling with the same
velocity in the negative z direction.
633
Wave traveling positive z-

direction
For free space
Lets concentrate on the wave traveling with velocity v in the

positive z direction
From (2) we may find a relation for H in the positive wave:
634
317
Wave impedance
Integrating, and ignoring (static) constants of integration, we

have
The wave impedance of the medium is
635
Wave impedance
The quantity may also be considered a

constant of the medium, and it will be a useful
parameter in the analysis of electromagnetic
waves.
It is also known as the intrinsic impedance of
the medium.
Thus, its dimension is ohm ().
For vacuum (free space),
636
318
equations
A plane wave solution of
the wave equation in a
conducting medium
Conductivity
In the previous section we considered a uniform plane wave
propagating in free space.
Here we consider a uniform plane wave propagation in a
conductor.
The term conductor will be applied to those materials in which
collisions determine the movement of the charges when an
electric field is applied.
Let it be (>0) the constant conductivity of the homogenous
medium.
Thus the current density resulting from movement of the
charges in conducting medium is given by Ohms law:
638
319
Maxwells curl equations in time-

harmonic case
Now Maxwells curl equations can expressed in the form
Applying curl operation to Faradays law (M III) we can

eliminate magnetic field
639
Note: Sinusoidal fields

A real physical field can be expressed in complex phasor
form
So
Now the field quantities

in this formula collection
are not instantaneous
values but complex
phasors.
640
320
Wave equation in conductive

medium
Taking divergence from (M IV) and using assumed
homogeneity of the medium (, ) and generally valid vector
identities
641
Note: The divergence of the curl of any vector is zero.
Solution of the wave equation

If we mark
The wave equation reduces to the form
in the case of plane wave where an electric field has only x-

component.
A solution of this wave equation is
where is a complex propagation coefficient and C1 and C2

are some constants.
642
321
Propagation coefficient
The complex propagation coefficient can be written
Using boundary conditions arbitrary constants can

be determined.
643
Attenuating waveform
Here we assume
And
Finally, the inverse transform gives
644
322
Problem
Show that the expressions of

attenuation coefficient () and
phase coefficient () for the
electro-magnetic wave in
lossy medium can be
presented in the following
forms a) and b).
645
Solution
646
323
Solution
647
Solution
648
324
Problem
Calculate the attenuation (dB/m) in the
(saltless) lake water (r=80, =10 mS/m)
and in the sea water (r=80, =5 S/m) for an
electro-magnetic wave with frequency
a) f=10 kHz
b) f=10 MHz
c) f=10 GHz
and assuming that a field strength varies
according the distance in the form
E exp(-z), where z is a distance
649
Solution
Attenuation
expressed in dB
650
325
Solution (r=80, =10 mS/m)
651
Solution (r=80, =10 mS/m)
652
326
Solution (r=80, =5 S/m)
653
Solution (r=80, =5 S/m)
654
327
Problem
Show that the propagation velocity (v) of the electromagnetic
wave in lossy medium can be presented in the following forms
a) and b).
655
Solution
656
328
Solution
657
Problem
Calculate the propagation velocity of the

electromagnetic wave
a) in the air (r = 1, 0)
b) in the lake water (r = 80, = 10 mS/m)
c) in the sea water (r = 80, = 5 S/m)
when corresponding vacuum values are
08.85410-12 As/(Vm), c02.998 108 m/s and
f=100 MHz.
658
329
Solution
659
Solution
660
330
Solution
661
equations
Penetration of
electromagnetic fields
into a good conductor
331
Plane conductor
The simplest case to solve is a plane conductor

of infinite depth, and with no field variations
along the width or length dimension.
This case is frequently taken as that a
conductor filling the half space z > 0 in a
rectangular coordinate system with the x-y
plane coinciding with the conductor surface.
663
Good conductor
Lets look at penetration of uniform field into good
conductor filling half-space 0<z<.
If we assume that at z=0 an electric field has value
Ex(z=0)=E0 and propagation direction is z, then we
can use previously derived plane wave equation and
its solution.
So
For any good conductor the following relation is

valid
664
332
Amperes law for good conductors
Now, we can write (M IV) in phasor notation
Note: For all but the poorest conductors (such as earth) the displacement current term is
665
completely negligible compared with conduction current term for radio frequencies at least.
Approximation of propagation
coefficient
Lets return to the previous solution of the wave equation
where
Now
Proofing as homework!
666
333
Skin depth or depth of

penetration
Substituting the previous expression to Ex we get
where
Skin depth or depth of penetration
The magnetic field and the current density are governed by

the same form differential equations as the electric field.
Thus
667

penetration
It is evident from that the magnitudes of the
fields and current density decrease
exponentially with penetration into the
conductor, and z=s has the significance of the
depth at which the magnitudes have decreased
to 1/e (about 36.8 per cent) of their values at
the surface (E0, H0, J0).
The quantity s is called depth of penetration
or skin depth.
Note: symbol is also used as a symbol of skin depth or depth of penetration.668
334

penetration
According to

We see that s decreases, when increase
f

E.g. for aluminium s1 mm, f=10kHz

s1 m, f=10GHz
669
Note
Depth of penetration is smaller, the higher the
conductivity, the higher the permeability, and the higher
the frequency, since it is inversely proportional to the
square root of each of these.
The fields and current penetrate deeper than the depth
s; this s is merely the point at which they have
decreased to 1/e of their values at the surface.
The concept as stated here applies strictly to plane
conductors but it may be extended to conductors of
other shapes so long as the value of s calculated is
much smaller than any curvatures of the surfaces or
curvature of surface
s
thickness of conductor 670
335
Comment
In spite of the infinite depth requirement, the analysis
of this case is practically important to many conductors
of finite extent, and with curved surfaces, because at
high frequencies significant fields are concentrated very
near the surface, so that radius of curvature and
conductor depth may be taken as infinite in this
comparison.
Moreover, any field variations along the length or width
dimension due to curvature, edge effects, or variations
along a wavelength are ordinarily so small compared
with the variations into the conductor that they may be
neglected.
671
Problem
The surface value of the field is E0. Determine
the magnitude of field penetrated s meters
deep into the good conductor.
672
336
Problem
The surface value of the field is E0. Determine

the magnitude of field penetrated s meters
deep into the good conductor.
Let
The magnitude of the field is
0.368 E0
673
Problem
Show that the unit of s defined previously is
meter.
1
s
f
674
337
Problem
Show that the unit of s defined previously is

meter. 1
s
f
Vs A
1 B m 2 Vs J m 2 A
s , ,
f H A Am E V Vm
m m
1
s m
1 Vs A
s Am Vm
675
Problem
Defining wavelength as = v/f =1/(f) show

that, for a good conductor, depth of penetration
is always a very small quantity compared with
waveIength or s<<.
676
338
Solution
1
s , for a good conductor
f
1 1 1
s2
f v 1

1 v 2
v v
2f 2 f 2 2
s 1 1
0.23 s
2 4.4
677
Problem
Show that
when
678
339
Solution
679
equations
Field theoretic problems
340
Problem 1
Write Maxwells equations
in Cartesian (xyz) coordinates, when
681
Solution
Write Maxwells equations in Cartesian (xyz)
coordinates.
682
341
Solution
683
Solution
684
342
Solution
685
Solution
686
343
Problem 2
Derive the global form version from the local

form Gauss laws (M I and M II) using the
Divergence theorem:
A surface integral

A dS

S
687
Solution
Derive the global form version (surface integrals) from the
local form Gauss laws (M I and M II) using the Divergence
theorem (Gauss theorem):
Thus
688
344
Solution
Divergence theorem (Gauss theorem):
Thus
689
Problem 3
Derive the global form version (line integrals) from the local
form Faradays law and Amperes law (M III and M IV) using
the Stokes theorem:
690
345
Solution
Stokes theorem:
691
Solution
Stokes theorem:
692
346
Problem 4
693
Solution
Now Maxwells equations are
Operations:
M IV and D E
M III E B H
B H

t t
694
347
Solution
On the other side we have well-known vector

identity
695
Problem 5
696
348
Solution 5
697
Solution 5
698
349
Problem 6
699
Solution 6
Let us derivate the solution candidate Ex-
Let us substitute the derivates to the

wave equation Ex- is a solution of
the wave equation
700
350
Problem 7
701
Solution 7
702
351
Solution 7
703
Solution 7
704
352
Solution 7
705
Solution 7
706
353
Problem 8
707
Solution 8
708
354
Solution 8
709
Solution 8
710
355
Problem 9
711
Solution 9
Now
712
356
Solution 9
713
Solution 9
714
357
Solution 9
715
Problem 10
Write the local form Maxwells equation in the
following special cases:
1 for static fields
2 for sourceless fields
3 for sinusoidal fields in phasor form
716
358
Solution 10
717
Solution 10
718
359
Solution 10
719
Solution 10
720
360
Solution 10
721
Note
722
361
Problem 11
723
Solution 11
724
362
Solution 11
725
Note
726
363
Problem 12
727
Solution 12
728
364
Problem 13
729
Solution 13
730
365
Solution 13
731
Problem 14
732
366
Solution 14
733
Solution 14
734
367
Solution 14
735
Solution 14
736
368
Solution 14
737
Notes
738
369
Problem 15
739
Solution 15
740
370
Solution 15
741
Problem 16
742
371
Solution 16
743
Problem 17
744
372
Solution 17
745
Problem 18
746
373
Solution 18
747
Note
748
374
Problem 19
749
Solution 19
750
375
Problem 20
751
Solution 20
752
376
Passive microwave devices
Definitions and classification

of transmission lines and
waveguides
Uniformity along the direction of

propagation
The propagation characteristics of a
transmission line or a waveguide can be
determined from the study of the
electromagnetic fields within the general
structure depicted in the figure below.
Waveguide of arbitrary
cross section, uniform
along the direction of
propagation.
754
377
Uniform transmission line or

waveguide
The cross section of this structure is arbitrary.
It can contain several different propagation media (air,
dielectrics, ferrites), in which case it is called inhomogeneous.
It can also contain, or be surrounded by, metallic conductors.
The direction of propagation is straightforward: the
longitudinal coordinate axis z is located along this direction.
A displacement along this axis does not modify either the
geometry, nor the material properties: these quantities are thus
independent of the longitudinal coordinate z (translation-
invariance).
The transmission line or waveguide is then called uniform.
755
Selection of a coordinate system

The system of coordinates used to study the propagation must
be well adapted to the transverse geometry of the line.
Boundaries and interfaces should be described by simple
geometrical expressions, so that boundary conditions for the
fields can be satisfied.
In all uniform structures, the coordinate system contains the
coordinate z, defined in the previous section, and two
coordinates located in the transverse plane.
For problems exhibiting a symmetry of revolution, the
(circular) cylindrical coordinate system is the obvious choice.
A rectangular waveguide is best studied in a rectangular
reference base.
756
378
Example: Cartesian coordinate

system
All mean exactly the same

thing but just a different
notation for the unit vectors.
757
Example: Cylindrical coordinate

system
Both mean exactly the same thing

but just a different notation for the
unit vectors.
758
379
Example: Spherical coordinate

system
Both mean exactly the same thing

but just a different notation for the
unit vectors.
759
Example: Vector field
In engineering terms, a vector field implies a

vector associated with every point is space.
Electric field: E(x,y,z,t ) or E(r ,t )
Magnetic field: H(x,y,z,t ) or H(r ,t )
760
380
Separation
Transverse planes are always perpendicular to

the direction of propagation.
Transverse and longitudinal dependencies of
the fields are thus independent of each other,
and the differential operator del or nabla can
be expressed by the sum of a transverse part
(index t) and a longitudinal one:
761
Rectangular coordinates
In the rectangular coordinate system, the transverse part of the
operator can be further separated, in terms of the coordinates x
and y:
In all other coordinate systems, the transverse operator cannot

be separated in this manner!
Similarly, the phasor-vectors have transverse and longitudinal
components:
The same is true for the vector r, which indicates position:
762
381
Consequence of separation
Within every system of cylindrical coordinates
(which correspond to uniform lines), Maxwells
equations are partially separable.
The general solution for the electromagnetic fields is
then the sum of several terms, each term being the
product of a function of the transverse coordinate rt
by a function of the longitudinal variable z.
The transverse and longitudinal dependencies of the
fields can then be studied independently of each
other.
763
Longitudinal dependence of the

fields: Wave equation
Applying the method of separation of variables, and
then grouping the resulting terms, a wave equation
involving the direction of propagation is obtained for
every field component in an isotropic structure
The complex constant 2 results from the separation

process.
Its value depends on frequency, material properties
and the particular solution considered for the fields.
764
382
General solution
Wave equation is satisfied by the general form

of solution:
The two terms correspond, respectively, to the

forward and to the backward waves, functions
Xi+ (rt) and Xi- (rt) defining their transverse
dependencies in the plane z= 0.
765
Propagation factor and related

quantities
The meaning of the propagation factor
= +j :
The real part , expressed in Neper per meter
(Np/m), is called attenuation per unit length.
It represents the damping of the wave traveling along
the transmission line.
Its inverse = 1/ is the skin depth, the distance over
which the signal amplitude decreases to 1/e of its initial
value.
766
383
Propagation factor and related

quantities
The meaning of the propagation factor
= +j :
The imaginary part , measured in radians per
meter (rad/m), is the phaseshift per unit length.
It indicates the phase variation of the wave along the
direction of propagation.
The wavelength along the line or waveguide, denoted
by g, is inversely proportional to :
767
Propagation velocities
The two propagation velocities, the phase and group
velocities, are respectively defined by the relations
An observer following a zero of the field moves at the

phase velocity.
An observer following the envelope of a modulated
signal at the group velocity.
768
384
Guided waves
Theoretically the simplest case is to deal with waves
that propagated in infinite size media.
For many applications it is desirable to have
electromagnetic energy be guided in much the same
way as water flow is guided by having it flow in
pipes.
Transmission lines are the simplest structures that
guide electromagnetic waves.
Transmission line would be any two arbitrary shaped
metal conductors that are long and uniform in at least
one dimension.
769
Definition: Transmission line

Transmission line is the preferred term to
describe transmission systems with two or
more metallic conductors electrically insulated
from one another
for instance the two-wire line and the coaxial line.
770
385
Some transmission line structures

Capacitance and inductance per unit length for each structure
771
E-fields and H-fields for the

previous transmission lines
772
386
Some transmission line structures

Capacitance and inductance per unit length for each structure
773
E-fields and H-fields for the

previous transmission lines
774
387
Definition: Waveguide
The propagation in a waveguide is generally
ensured by successive reflections on the guide
boundaries.
These are conducting walls in the case of metallic
waveguides.
Dielectric waveguides and optical fibers
utilize the total reflection on the interface
between two dielectric materials.
775
Definition: Modes of propagation

The solution of Maxwells equations, subject
to the transverse boundary conditions in the
structure of transmission line or waveguide is
an eigenvalue problem.
A set of solutions is obtained, which are called
the modes of propagation.
They are the eigenvectors of the problem, each
one associated with an eigenvalue (propagation
factor, transverse wavenumber).
776
388
Modes of propagation
The modes of propagation form an infinite set,
which is discrete for a closed structure
(surrounded by a metal boundary),
[to which must be added a continuous spectrum of
radiating modes when the structure is open (fields
extending to infinity)].
Each mode possesses specific propagation
properties:
attenuation and phase-shift per unit length,
propagation velocities,
cut-off frequency. 777
Multimode propagation
When propagation of a signal takes place at the same time over several
modes, the difference in the propagation velocities may produce
distortions of the signal.
To avoid this unwanted effect in radar and in communications, the shape
and the dimensions of the line are adjusted so that only one mode can
propagate at the signal frequency.
The possible existence of several propagating modes limits the available
frequency band in all transmission lines.
The length of the line or the bit rate of the signal are reduced to maintain
distortion within acceptable limits.
The presence (or absence) of a particular mode of propagation on a
transmission line depends on the excitation, i.e., on the boundary conditions
at both ends of the line (generator and load).
Note: In fiber optics, also multimode operation (thick fiber) is used in short
distance communication. 778
389
Classification of the modes of

propagation
The presence or the absence of longitudinal
field components affects the propagation
behavior of the modes.
Four mode categories can exist, as shown
below.
779
Classification of lines and

waveguides
A large number of different structures can be used to
transmit electromagnetic signals:
1. either open (radiation can take place) or closed (fields
entirely enclosed within a conducting envelope);
2. either homogeneous (one single propagation medium
without transverse dependence of the material properties),
or inhomogeneous (several different propagating media, or
a single medium having a continuous variation of the
material properties in the transverse plane);
3. either conductorless, or possessing one or more
conductors.
780
390
Classification of lines and

waveguides
Note: When a waveguide or a transmission line (for instance a coaxial line) is partially
781
filled with dielectric, they are respectively called loaded waveguide or loaded line.
Cross sections of the transmission

lines
782
391
Cross sections of the waveguides
783
Comment
Early microwave systems relied on waveguide and coaxial lines for
transmission line media.
Waveguide has the advantage of high power-handling capability and low
loss but is bulky and expensive.
Coaxial line has very high bandwidth and is convenient for test
applications, but is a difficult medium to fabricate complex microwave
components.
Planar transmission lines provide an alternative, in the form of stripline,
microstrip, slotline, coplanar waveguide, and many other types of related
geometries.
Such transmission lines are compact, low in cost, and are capable of being
easily integrated with active devices such as diodes and transistors to form
microwave integrated circuits.
784
392
Characteristics of some
transmission media
785
Notes
Wire pair is distance limited. The shorter the pair length, the higher the
bit rate.
LOS microwave is limited by statute (legal limitations), meaning by the
ITU Radio Regulations and the national regulatory authority.
Satellite communication faces the same legal limitations. Geostationary
orbit (GEO) satellites have long delays, which could affect interactive data
systems. Only one GEO satellite relay allowed for a voice connectivity.
The limits of fiber optics are still being explored. All terrestrial buried and
aerial cable systems are vulnerable to severing by natural disaster or by
man.
Coaxial cable is limited by amplitude-frequency response characteristics.
In nearly every instance, fiber optic cable connectivity is preferred.
786
393
Twisted pair
(a) Category 3 UTP (Unshielded Twisted Pair), 16 Mbps.

(b) Category 5 UTP, 100 Mbps.
Category 6 and 7, 250 600 Mbps.
787
Coaxial cable
50 ohm and 75 ohm 1 GHz
788
394
Optical fiber
(a) Three examples of a light ray from inside a silica fiber

impinging on the air/silica boundary at different angles.
(b) Light trapped by total internal reflection.
Multimode uses total reflection.

Single mode 50 Gbps over 100 km.
789
Transmission of light through fiber
Attenuation = 10*log10(transmitted power/received power)

f = c/ f = 300,000,000/1.3*10^-6 = 2.31*10^14 = 231 THz.
Bandwidth = 300,000,000*(1/1.22 1/1.37)*10^6 = 0.27*10^14 = 27000 GHz 790
395
Fiber cables
(a) Side view of a single fiber.

(b) End view of a sheath with three
fibers.
791
Fiber vs. copper
Repeaters: copper <5 km vs. fiber >50 km.

1000 twisted pairs 1 km long weights 8 tons
vs. 2 fibers with more capacity weight 100
kg.
Fiber doesnt leak the light excellent
security.
792
396
Example: Bit rate

Coax 750 MHz bandwidth
If we assume that it is possible to code 8 bps/Hz of
bandwidth.
6 Gbps.
Fiber:
f = c/ f = |c
For = 1.3 m band = 0.17 m that gives the

bandwidth f = 40 THz
To get 40 Tbps coding 1 bps/Hz of bandwidth is
needed. 793
Transmission lines and

waveguides
Recap
397
Guiding structures
A guiding structure is one that carries a signal

(or power) from one point to another.
There are three common types:

Transmission lines
Fiber-optic guides
Waveguides
Note: An alternative to guiding structures is wireless transmission in

more or less free-space using antennas. 795
795
Transmission line properties
Has two conductors running parallel.

Can propagate a signal at any frequency (in theory).
Becomes lossy at high frequency.
Can handle low or moderate amounts of power.
Does not have signal distortion, unless there is loss.
May or may not be immune to interference.
Does not have Ez or Hz components of the fields (TEMz).
796
398
Example: Transmission line

CAT 5 cable (twisted pair)
Coaxial cable
The two wires of the transmission line are twisted to reduce

interference and radiation from discontinuities. 797
Symmetric transmission line
The twin lead is an open type of transmission

line the fields extend out to infinity.
+ -
The extended fields may cause
interference with nearby
objects. (This may be improved
by using twisted pair.)
Note: Having fields that extend to infinity is not the same thing as
having radiation, however. 798
399
Example: Transmission lines

Transmission lines on printed-circuit boards
w
r
h
r h w
Microstrip Stripline
w w w
r h r h
Coplanar strips Coplanar waveguide (CPW)

799
RF waveguide properties
Is a single hollow metal pipe.
Can propagate a signal only at high frequency: > c
The width must be at least one-half of a wavelength.
Has signal distortion, even in the lossless case.
Immune to interference.
Can handle large amounts of power.
Has low loss (compared with a transmission line).
Has either Ez or Hz component of the fields (TMz or
TEz).
800
400
Fiber-optic guide
Properties
Uses a dielectric rod
Can be made very low loss
Has minimal signal distortion
Very immune to interference
Not suitable for high power
Has both Ez and Hz
components of the fields
801
Two types of fiber-optic guides
1) Single-mode fiber
Carries a single mode, as with the mode on a
transmission line or waveguide.
Requires the fiber diameter to be small relative to a
wavelength.
2) Multi-mode fiber
Has a fiber diameter that is large relative to a
wavelength.
Its operation can be described using simple ray
theory and the principle of total internal
reflection (critical angle effect).
802
401
Optical fiber
Higher index core region
803
Transmission line theory

Lumped circuit elements: resistors, capacitors,
inductors
neglect time delays (phase)
Distributed circuit elements: transmission lines
account for propagation and

time delays (phase change)
We need transmission line theory whenever the

length of a line is significant compared with a
wavelength.
804
402
Passive microwave
components
Directional couplers
Components
Each individual component has a specific
purpose.
Its function may be to couple, mix, filter, isolate,
divide, add, act on a systems phase, convert RF
to DC, attenuate, or any combination of the
above.
Basic questions for component:
How does it work?
What does a data sheet or catalog say about that
component?
806
403
Directional coupler
One of the most widely used components in microwaves is the
directional coupler.
It may be employed by itself to monitor, sample, or attenuate; or it
may be used as part of another component, such as a mixer.
The basic definition of couple, from which the word coupler is
derived, is to bring two electrical circuits into such close proximity
as to permit mutual influence.
In other words, two circuits are side-by-side so that they have an
effect on one another.
The circuits could be two pieces of waveguide, two coaxial lines,
or two copper lines in a stripline coupler.
One of the circuits is the main line of the coupler; the other circuit is
the coupled line.
807
Basic directional coupler
808
404
Directional couplers
a coaxial coupler
a stripline coupler
a waveguide coupler
809
Directional coupler
Figure helps in our explanation of the term directional.
If energy is applied to port 1 in (a), most of the energy appears
at port 2, while a fraction appears at port 3.
The actual amount at port 3 depends on the coupling value;
standard couplers are of 6, 10, 20, and 30 dB.
Note: A 3 dB coupler, called a quadrature hybrid, is also available. 810
405
Dual-directional coupler
One method to expand the basic directional coupler is
through the construction of a dual-directional coupler.
Just as the name implies, this procedure involves two
couplers put back-to-back.
The one great advantage of this type of coupler is that it
allows you to monitor both forward and reflected power
at the same time.
It aids in the obtaining of data from a device which
indicates the input match and, thus, its VSWR.
811
Specifications
Below are sample specifications for a directional coupler as
they could appear on a data sheet or in a catalog.
812
406
Specifications
Connectors This portion of coupler

specifications is very important since it tells
you if you either need adapters to fit the
coupler into your system.
Note that additional adapters mean higher
overall losses in a system.
Connectors may be type N, BNC, SMA, APC-
7, or any special variety required.
813
3- and 4 port coupler
814
407
Specifications
Coupled Line VSWR This specification
indicates what sort of match the coupled port
(port 3) offers to any external load.
A low VSWR would mean a good match and
would cause little or no problems to any
external circuit.
815
Specifications
Coupling The ratio of the power available
at the coupled port to the power at the input
port. (Port 3 to port 1).
It is the amount of attenuation of the input
power as a result of the coupling structure.
This term is expressed in dB and has standard
values of 6, 10, 20, and 30.
816
408
Specifications
Coupling Deviation This measurement

indicates how flat the coupling is over a
specified band of operation; it is expressed as a
(+) or () dB value.
If we rely heavily on a specified value of
coupling over a wide range of frequencies,
coupling deviation should be low (e.g. 1 dB).
817
Specifications
Directivity This figure indicates how
accurately you are able to measure parameters in
your system.
It is the difference between the desired and
undesired couplings.
If a coupler has low directivity, the forward and
reverse powers in the coupler interfere with one
another and cause great inaccuracies to occur.
Ports 1 and 4 in are the undesired coupling; 1 and 3 are
the desired combination.
818
409
Specifications
Frequency This term indicates the range of

frequencies for which all the other
specifications are true.
Above and below this range, none of the
specifications apply.
Most couplers are designed to operate over an
octave band (1-2 GHz, 2-4 GHz, etc.).
819
Specifications
Impedance Impedance is an RF
resistance reading which ensures that any
component with a similar impedance value
connected to the coupler will operate properly.
The most common impedance is 50 ohms,
which is considered to be a standard in most
systems.
820
410
Specifications
Input VSWR This
specification is identical to
the coupled line VSWR in
that it is an indication of
what sort of match is offered
by the input port of the
coupler;
it is usually of low-value,
1.15:1 for example.
In Figure (A) it is the match
at port 1; in (B) it is either
port 1 or 2. 821
Specifications
Insertion Loss The insertion loss indicates the power lost
in the main line of the coupler primarily through dissipation.
In coaxial coupler it is simply a loss which occurs in a conductor as
energy is passed through it;
in a stripline coupler, the copper losses of the conductors; and
in a waveguide coupler, a loss within the guide with the only
contributing factor usually the length of the guide.
In Figures (A) and (B), the insertion loss is that from either
ports 1 and 2 or 2 and 1.
In lower value couplers (3 dB or 6 dB, for example) the
coupled energy due to the coupled port also is included in the
total through-line insertion loss, thereby increasing the overall
value.
822
411
Specifications
Model Number The model number is

simply a way of referring to and identifying
the coupler.
In most cases the last two numerals indicate
the coupling value.
823
Specifications
Power Rating This specification refers to
the amount of both CW and peak power that
the coupler is capable of handling.
Any power level greater than this value may
cause arcing or a deterioration of performance.
This term is one of the most important of
which to take care if you will be operating at a
high power level.
824
412
Specifications
Reverse Power Rating This term applies only to a
3-port coupler; it is usually a lower value than that of
the forward power rating because the fourth port is
internally terminated and so, generally, can not handle
any amount of power.
This specification is also quoted as a CW and peak
power number; it is the power handling capability of
the coupler if power is applied in a reverse direction.
At port 2 instead of port 1 in Figure (A).
825
Specifications
Size the actual physical size of the coupler.
This term is of a secondary nature except when
you are required to put the coupler in a restricted
space.
Weight Weight is self-explanatory term
which only comes into play when an overall
weight restriction is placed on a system, such
as with airborne equipment.
826
413
Directional coupler symbols
827
Applications of directional couplers
The most obvious use of a directional coupler is to

monitor an output power.
Its directional properties ensure that only the forward
power is coupled; its property of circuit isolation
leaves the main line power undisturbed.
Figure shows the orientation of a directional coupler
when used as a power monitor.
828
414
A scheme similar to the monitoring technique described earlier

is shown in figure below.
The big difference in such a setup is that a diode detector is placed at
the coupled port to provide a dc voltage proportional to any main line
power variations.
That voltage operates leveling circuitry which corrects the RF level and
eliminates differences in the output power over a certain specified
frequency band.
This technique is used widely where swept frequency measurements
are taken.
829
Figure below shows an application which takes advantage of

the directional couplers ability to handle higher CW power.
It is an actual example of a system used to check a 100 W power
amplifier.
The only components available were two-watt attenuators, five-watt
terminations, and the directional coupler.
It can be seen that a more adequate setup was created by utilizing the
couplers 200 watt capability and directivity of 25-30 dB.
None of the components used had its power rating exceeded; the
directional coupler was doing the majority of the work.
830
415
Figure shows an example where the coupler is

an essential part of another component.
In this case it is a very important part of a single
diode mixer.
The circuit is of a very basic mixer, but it shows
that by using a coupler the RF and LO signals have
a high degree of mutual isolation.
831
The dual-directional coupler is very useful in VSWR

measurements.
In Figure it is employed to measure the return loss which then
is converted to VSWR.
Power meter # 1 reads the forward power to the device under
test; meter # 2 measures any reflected power.
By comparing these two figures a return loss is determined
and converted to the input VSWR of the particular device
under test.
832
416
833
Summary
Directional coupler is a component which allows
two microwave circuits to be combined into one
integrated system in one direction, while being
completely isolated from each other in the
opposite direction.
Its property of being directional, its isolation, its
low input and coupled line VSWR, and its
adaptability to being 3- or 4-port or dual-
directional make the directional coupler one of
the most widely used components in microwaves.
834
417
Passive microwave
components
Detectors
Detector
The key to a reliable, trouble-free system is
simplicity.
One component whose operation is based on
simplicity is the microwave detector.
This concept of a simple and reliable
component is the very reason why the detector
is used so extensively in microwave systems.
836
418
Basic microwave detector
The basic construction of the microwave

detector is illustrated in figure below.
The heart of the detector is, of course, the
diode itself.
e.g. Schottky diodes are often used because of their
good RF characteristics, which include high
sensitivity.
837
Basic microwave detector

The diode is matched to the driving circuit (usually 50
ohms) so that maximum power transfer is obtained.
In this manner maximum efficiency of the diode is
ensured since any reflections due to mismatches are
eliminated and, thus, all of the input power reaches the
diode.
The dc return, besides acting as a ground for the diode,
also has the second function of acting as an RF choke
so that no RF is shunted to ground.
A low-pass filter is placed after the diode to eliminate
all of the high frequency ripple caused by the detection
process and allow only dc to be present at the output.
838
419
Specifications
A typical set of specifications for a microwave detector
839
Specifications
Frequency Response Frequency response
is a term used to indicate the RF performance
of the detector.
It is sometimes given in decibels per octave or
in one figure for overall performance.
Basically it is a measure of the variation in
sensitivity of the detector, expressed in dB.
840
420
Specifications
Maximum Input Power This term refers to
one of the most important considerations when
using a detector the highest level of RF power
that can be applied to its input.
It is always good practice to stay below this figure
by a reasonable margin.
The reason that there is a maximum level is that
there is a diode directly in line in the circuit;
wherever there is a semiconductor there is a
certain power level that could destroy the
detector.
841
Specifications
Polarity Polarity is either positive or
negative.
It is important to know the polarity of the
output of your detector since it will have to be
joined with some external circuit that requires
a certain polarity to operate properly.
842
421
Specifications
A typical sensitivity curve for a microwave detector
843
Specifications
Sensitivity From the previous sensitivity curve
for a microwave detector it can be seen that one
factor affecting sensitivity is the RF power
applied to the input of the detector.
Therefore, the detector sensitivity is the dc volt-
age produced at the output for a specific power
input, usually expressed in millivolts per
microwatt CW.
Sensitivity is simply how much power you need
to produce a certain voltage.
844
422
Detector applications
The microwave detector finds appications

wherever there is a need to control or convert
an RF signal to a DC voltage.
E.g. for display purposes.
845
Monitoring
One of the most common applications is
monitoring an RF line.
Power is coupled off the main line;
the detector provides a DC voltage proportional
to this power.
The monitoring device gives an instant reading of
the main line power.
846
423
Automatic Leveling Control (ALC)

A leveled power is needed, e.g., in swept measurements.
A coupler and detector work together to sample the main
line RF;
the detector converts the RF to DC to activate the leveling
circuits.
This DC voltage causes either the RF power out of the source to
be attenuated or the attenuation to be removed in order to raise
the power level.
The result of this instantaneous pattern of action and reaction of
coupler and detector is that the RF power remains level at the
system output.
Automatic Leveling Control (ALC) in microwave systems often
employs just such a process.
847
Detecting for display on a scope

It is impractical to attempt to view signals at
microwave frequencies directly on a conventional
oscilloscope.
Without a Network Analyzer figure below is the
answer.
A simple combination of an attenuator and a detector do the
job.
The attenuator is of great importance in the setup, it keeps
power both below a maximum power level and in the
linear range of the sensitivity curve of the detector, giving
much better results.
848
424
Passive microwave
components
Mixers
Mixer
A schematic representation of a mixer
850
425
Microwave mixer circuits
A basic block diagram of the three

fundamental parts of a mixer an input
coupling network, the diode circuitry, and an
output filtering network.
851
Input coupling network

The input coupling network is the circuitry
which allows the RF and LO signals to be
combined properly before being applied to the
diode portion.
This network could be a directional coupler, a
transformer, or a quadrature hybrid;
its prime feature should be that, under operating
conditions, it combines the two signals while
providing isolation between them.
852
426
Mixing process
The diode circuitry consists of the diode(s).
This part is where the actual mixing process occurs.
When the RF and LO signals are applied to the diode
circuitry, many combinations of the two are generated.
The signals which are most prominent are each of the
originals and their sum and difference.
In other words, if the RF signal is called f1, and the LO
signal called f2, the output would have f1, f2, f1 + f2, f1 -
f2, and every other combination of fundamentals and
harmonics, (2f1, 2f2, 2(f1 + f2), etc.)
853
Output filtering
These combinations of frequencies are the reason why the final
block the output filtering network is needed in a mixer.
Usually, the upper or lower sideband (that is, f1 + f2 or f1-f2 ) is the
desired output;
everything else must be eliminated to produce the desired and
necessary clean signal.
The filtering usually is accomplished by a low-pass network which
might be as simple as a shunt capacitor or as complex as a cascade
of low-pass filters; it could be a series of trap circuits which would
eliminate only specific unwanted signals.
For certain applications, the filtering could be performed by a band-
pass or high-pass filter.
The type of filtering and device used depends, of course, upon the
individual situation.
854
427
Basic types of mixers
Three basic configurations

single diode,
balanced, and
image rejection.
855
Single diode mixer

The advantage of the single diode mixer is its
simplicity of construction.
The disadvantages,
very narrow band device
No noise eliminating ability, which is a primary
concern of the sophisticated low-noise systems
however, far outweigh the advantages.
856
428
Balanced mixer
Here the input network is termed a balanced coupling
network, either a 90 or 180 hybrid.
Whichever type is used, the efficiency of the coupling
network is vastly greater than that of any single diode
mixer.
The increased efficiency of the hybrids means that virtually
all of the power reaches the diodes, and there is very little
loss due to mismatches.
So a conversion loss through the mixer is reasonably low.
Also, this type of coupling network usually operates over an
octave band, another great improvement.
857
Balanced mixer
A mixing apparatus or configuration which
contains four matched diodes called a quad.
A quad makes up a balanced configuration
with two matched sections in each side, having
the effect of lowering conversion loss and
noise.
858
429
The advantages of balanced

mixers
Balanced mixers have very good noise
characteristics and low conversion losses;
They operate over a wide band of frequencies
(both LO and RF);
They have an acceptable input VSWR on the
RF and LO ports.
859
Specifications
A typical data sheet for a balanced mixer
860
430
Specifications
Conversion Loss Conversion loss is the

level of the IF output as compared to the level
of the RF input.
Usually expressed in dB, it can be considered
as a measure of the efficiency of the mixer;
it indicates how efficiently the two input
frequencies are combined to form the IF output.
861
Specifications
Isolation It may be any one of three
combinations of measurements referred to by this
term RF/LO, RF/IF, or LO/IF.
In each case, isolation is a measure of the
undesired frequency at an auxiliary port.
An example would be as follows the RF/LO
isolation figure, in dB, is the amount of RF power
measured at the LO port compared to that at the
RF port.
The other two isolation figures are found the same
way.
862
431
Specifications
LO Power LO power is the amount,

usually expressed in dBm, necessary to
provide proper operation of the mixer.
This power level is used to bias the diodes in a
non-linear region so that mixing action
occurs.
863
Specifications
Noise Figure Every network, whether a mixer or other
component, has a certain amount of internal noise
associated with its operation.
A comparison of this level of noise to that of the desired
signal in the network is called the signal-to-noise ratio.
There are input and output signal-to-noise ratios for each
network under consideration.
The noise figure measures the reduction of a signal-to-noise
ratio by the network under consideration (for example, the
balanced mixer).
It is a term expressed in dB which compares the input and
output signal-to-noise ratios.
864
432
Noise Figure
Expressed mathematically it is:
865
Specifications
VSWR VSWR is an important term which
indicates the match, or return loss, of each port
under actual operating conditions.
If measurements are made under abnormal
circumstances, the figures obtained are useless.
Only when the diodes in the mixer are being
driven properly are true readings obtained.
866
433
Mixer applications
The mixer finds many applications in microwave systems.
In a receiver front end the mixer is used to convert the high
frequency input signal to a lower frequency in order to obtain the
transmitted information.
It is much easier to operate on this lower frequency than on the input
signal as it appears at the antenna.
The same concept is used in microwave network analyzers.
The input GHz signals are converted down even to the kHz range for
easy operation and greater accuracy.
In applications of this type, it is very important to have low noise
mixers with a high degree of isolation and minimum conversion loss.
867
Mixer applications
Other utilizations are as an up-converter in a
transmitter (to increase the frequency for
transmission), as a balanced modulator (which
incorporates a 180 hybrid and reversed diodes
in a balanced mixer) and in many instances
where signal processing must be
accomplished.
868
434
Image-rejection mixer
The image-rejection mixer gives the desired sideband (f1 + f2 or f1 - f2) with
a minimum of output filtering.
In a basic mixer the local oscillator may be above or below the RF signal in
frequency.
Here, let us suppose an LO below the RF.
When the mixer is operating there is an lF signal equal to the difference
between the two frequencies.
This operation is a proper one.
However, there is also a signal separated from the RF by the same IF
amount but it is above the RF signal and is called the image frequency
generated by the mixer.
Obviously there are occasions where this signal would create problems.
Thus, rejecting is needed.
The image- rejection mixer provides the solution.
869
Figure shows a block diagram of an image-
rejection mixer.
870
435
The input coupling network consists of a combination of a 90
hybrid for the RF, a 0 power divider for the LO, and the input
coupling networks of the two very important balanced mixers.
The diode circuitry is that contained in the mixers; it is
identical to that previously discussed two diodes in each
mixer.
The output filtering consists of the 90 IF hybrid, a quadrature
hybrid designed to operate at the IF frequency.
One output port (No. 1) produces the upper image frequency;
the other (No. 2), creates the lower.
871
Pros and cons

The quadrature characteristic (90 phase
relationship) of the hybrid results in only one
frequency being present at each port when the
opposite port is terminated, thus providing
rejection of the opposite image frequency.
When a clean IF signal is wanted, this ability is
of great importance.
The disadvantage of the image-rejection mixer
is its complexity.
872
436
Passive microwave
components
Quadrature hybrid
Quadrature hybrid
Definition 1: A quadrature hybrid is a 4-port, 3
dB coupler capable of dividing an input signal
into two mutually isolated quadrature phased
outputs while maintaining isolation of the
fourth port from the input.
Definition 2: A quadrature hybrid is a
directional coupler whose two outputs are
equal in amplitude and separated from one
another by a constant 90 phase.
874
437
Quadrature hybrid symbols
That of having output signals in quadrature,

separated 90 from one another, is one of the
most important property of the quadrature
hybrid component. In Figure ports 2 and 3
would be in quadrature with an input at either
port 1 or 4.
875
Typical data sheet for a quadrature

hybrid
876
438
Specifications
Amplitude Balance This figure is a
comparison of the power levels of the two
output ports on the hybrid ideally there
should be no difference in amplitude between
the two outputs, but in the real world there
naturally is some variation.
It is necessary for this figure to be as low as
possible so that both circuits driven have the
same level applied to them.
877
Specifications
Insertion Loss Figure illustrates the insertion loss of
a quadrature hybrid basically the difference between
the maximum height of the coupled curve (1-2) and the
straight-through curve (1-3).
This internal loss of the device causes its output
response to deviate from an ideal 3 dB figure across the
desired band.
878
439
Specifications
Isolation This term tells you how good a
conjugate pair is formed by the terminals.
The isolation between terminals is very
important to consider if the hybrid is going to
be used to drive two active devices, such as
transistors or diodes.
If the figure is not high enough, any imbalance
in one device affects both of them and may
eventually cause one or both to be destroyed.
A network has a conjugate pair if an input applied to one of the branches has no effect on the
879
other; that is, two branches form a conjugate pair if they are isolated from one another.
Specifications
Phase Balance The quadrature hybrid is
designed to have a 90 separation between the
output terminals.
But, because of variations within the hybrid,
the phase of the output varies with frequency.
The phase balance figure on a data sheet is a
measure of how well the two output terminals
track over the band of operation.
880
440
Applications of quadrature
hybrids
Since they provide isolation between the RF
and LO while applying equal signals to each of
the diodes, quadrature hybrids are an integral
part of a balanced mixer.
881
hybrids
They are ideal for a matched detector, also
because of the equal level applied to the
detector diodes.
882
441
RF and DC combination
Quadrature hybrids are also appropriate as a

means of DC biasing an RF circuit they
supply a continuous DC path while providing
isolation for RF.
883
hybrids
A common application is combining of amplifiers.
The ability of the hybrid to provide isolated signals of
equal amplitude to a device is of great value.
The quadrature effect (90 separation) allows the split
of signals at the input and their recombination at the
output.
Combining amplifiers
884
442
Applications of quadrature hybrids

Other areas in which quadrature hybrids are used as 2-way
power dividers, time delay circuits, and modulators, and as
variations on the simple PIN switch.
Obviously, the quadrature hybrid is a component with many
applications.
When a device is needed to provide two output ports of equal
amplitude and which are isolated from one another, the hybrid
satisfies that requirement.
Its quadrature characteristics are also of great value when
division and recombination are to be accomplished with active
devices.
The PIN diode is a semiconductor device that consists of three regions:

an intrinsic region is layered between an n-doped and a p-doped region. The PIN diode
885
can control large RF and microwave signals with comparably small DC signals.
Signal flow graphs and

scattering parameters
443
Multiport networks
A general circuit can be represented by a multi-port network, where the
ports are defined as access terminals at which we can define voltages and
currents.
Note: Equal and opposite currents are
assumed on the two wires of a port.
Examples:
One-port network I1 I1
V1 V1
Two-port network
I2
I1 I2 I1
V1 V2 V1
V2
887
887
Signal flow graphs

The signal flow graph is a very useful
technique for the analysis of microwave
networks in terms of transmitted and reflected
waves.
Main phases of the signal flow graph
technique use:
the construction of the flow graph itself,
the reduction, or solution, of the flow graph.
888
444
Features and the construction of

the flow graph
The primary components of a signal flow
graph are nodes and branches:
Nodes: Each port, i, of a microwave network has
two nodes, ai and bi.
Node ai is identified with a wave entering port i, while
node bi, is identified with a wave reflected from port i.
Branches: A branch is a directed path between an
a-node and a b-node, representing signal flow
from node a to node b.
Every branch has an associated S parameter or
reflection or transmission coefficient.
889
Signal flow graph of a two-port

network
It is useful to consider the flow graph of an arbitrary two-port
network with incident and reflected waves at each port, and the
corresponding signal flow graph representation.
Definition of incident and reflected waves
Signal flow graph
890
445
Graphical illustration of the

network behavior
The flow graph gives an intuitive graphical illustration of the
network behavior.
For example, a wave of amplitude a1 incident at port 1 is split, with part
going through S11 and out port 1 as a reflected wave and part transmitted
through S21 to node b2.
At node b2, the wave goes out port 2; if a load with nonzero reflection
coefficient is connected at port 2 this wave will be at 1east partly reflected
and reenter the two-port network at node a2. Part of the wave can be
reflected back out port 2 via S22, and part can be transmitted out port 1
through S12.
Source Load
end end
891
Examples
The signal flow graph representations of a
one-port network and a voltage source
892
446
Flow graph algebra
Once a microwave network has been

represented in signal flow graph form, it is not
very difficult to solve the ratio of any
combination of wave amplitudes.
This can be done using basic decomposition
rules or applying Masons rule.
893
What are S-parameters?

S-parameters or scattering parameters are
reflection and transmission coefficients.
Transmission coefficients are commonly called
gains or attenuations; reflection coefficients
are directly related to VSWRs and
impedances.
S-parameters are vector quantities they
give magnitude and phase information.
894
447
Parameter models
Conceptually, s parameters are like h or

z parameters because they describe the
inputs and outputs of a black box.
The inputs and outputs are in terms of power
for s parameters;
for h, y and z parameters, they are voltages
and currents.
895
I/O-signals
Using the convention that a is a signal into a
port and b is a signal out, the figure below
helps to explain s parameters.
Test device
896
448
I/O-signals
Here, a and b are the square roots of

power;
(a1)2 is the power incident at port 1 and
(b2)2 is the power leaving port 2.
The diagram shows the relationship between
the s parameters and the as and bs.
Test device
897
Reflection and transmission

coefficients
For example, a signal, a1, is partially reflected at port
1; the rest of the signal is transmitted through the
device and out of port 2.
The fraction of a1 that is reflected at port 1 is s11; the
fraction of a1 that is transmitted is s21.
Similarly, the fraction of a2 that is reflected at port 2
is s22, and the fraction s12 is transmitted.
Test device
898
449
I/O-relations
The signal, b1, leaving port 1 is the sum of the

fraction of a1 that is reflected at port 1 and the
fraction of a2 that is transmitted from port 2.
Thus, the outputs can be related to the inputs
by the equations:
b1 = s11a1 + s12a2
b2 = s21a1 + s22a2
899
R-T coefficients
From the previous equations
b1 = s11a1 + s12a2
b2 = s21a1 + s22a2
when a2=0 s11=b1/a1 and s21=b2/a1
when a1=0 s12=b1/a2 and s22=b2/a2
900
450
Measuring R-T coefficients

The setup below shows how s11 and s21 are measured.
Port 1 is driven and a2 is made zero by terminating

the 50 transmission line coming out of port 2 in its
characteristic 50 impedance.
This termination ensures that none of the transmitted
signal, b2, is reflected toward the test device.
901
Measuring transmission
coefficients
Similarly, the setup for measuring s12 and s22 is:
Compare this to the previous setup for measuring s11

and s21 .
902
451
Device under test

The S-parameters can be measured by
embedding the two-port network (the device-
under-test) in a transmission line whose ends
are connected to a network analyzer.
An experimental setup:
903
Two-port network under test

Figure shows more details of the connection.
The generator and load impedances are configured by
the network analyzer.
The connections can be reversed, with the generator
connected to port 2 and the load to port 1.
904
452
Two-port network under test

The two line segments of lengths 11,12 are assumed to have characteristic
impedance equal to the reference impedance Z0.
Then, the wave variables a1, b1 and a2, b2 are recognized as normalized
versions of forward and backward traveling waves.
The network analyzer measures the waves a1, b1 and a2, b2 at the
generator and load ends of the line segments, as shown in the figure.
From these, the waves at the inputs of the two-port can be determined.
905
Comment
S-parameters are vector quantities they
give magnitude and phase information.
Most measurements of microwave components
have been measured historically only in terms
of magnitude.
Obviously because it was too difficult to
obtain both phase and magnitude information.
906
453
Parameters
To represent multi-port networks we use:
- Z (impedance) parameters Not easily

- Y (admittance) parameters measurable at
- h (hybrid) parameters high
- ABCD parameters frequency
Measurable at high
- S (scattering) parameters
frequency
907
Summary
2-port and corresponding S-formalism
908
454
Summary
N-port and corresponding S-matrix
909
Summary
At high frequencies, Z, Y, h & ABCD parameters are
difficult (if not impossible) to measure.
V and I are not uniquely defined.
Even if defined, V and I are very difficult
to measure.
Required open and short-circuit conditions are
often difficult to achieve.
Scattering (S) parameters are often the best
representation for multi-port networks at high
frequencies.
Instead of open and short-circuit conditions only matched
terminal is needed.
910
455
Summary
b1 0 S11a1 0 S12a2 0
b2 0 S21a1 0 S22a2 0
Scattering parameters
b1 0 Output is
S11 matched input reflection coefficient
a1 0 a 0
2 with output matched
b1 0 Input is reverse transmission coefficient
S12 matched
a2 0 a 0 with input matched
1
b2 0 Output is forward transmission coefficient

S 21 with output matched
a1 0 a 0 matched
2
output reflection coefficient

b 0 Input is
with input matched
S 22 2 matched
a2 0 a 0
1 911
Summary
912
456
Introduction to transmission
line theory
Wave-guiding structures
A wave-guiding structure is one that carries a signal (or
power) from one point to another.
There are three common types:

Transmission lines
Fiber-optic guides
RF waveguides
Note: An alternative to waveguiding structures is wireless transmission in more or less free-

space using antennas. 914
457
Transmission line properties
Transmission line
Has two conductors running parallel
Can propagate a signal at any frequency (in theory)
Becomes lossy at high frequency
Can handle low or moderate amounts of power
Does not have signal distortion, unless there is loss
More or less immune to interference
Does not have Ez or Hz components of the fields (TEMz)
915
Note: Compare properties

RF waveguide Is a single hollow metal pipe
Can propagate a signal only at high frequency: >
c
The width must be at least one-half of a wavelength
Has signal distortion, even in the lossless case
Immune to interference
Can handle large amounts of power
Has low loss (compared with a transmission line)
Has either Ez or Hz component of the fields (TMz or
TEz)
Optical fiber wave guide Uses a dielectric rod
Can propagate a signal at any frequency (in theory)
Can be made very low loss
Has minimal signal distortion
Very immune to interference
Not suitable for high power
Has both Ez and Hz components of the fields 916
458
Example: Transmission line

Twisted pair
Coaxial cable
The two wires of the transmission line are twisted to reduce interference and radiation from
discontinuities. 917
Example: Transmission lines

Transmission lines on printed-circuit boards
w
r
h
r h w
Microstrip Stripline
w w w
r h r h
Coplanar strips Coplanar waveguide (CPW)
Using transmission lines to synthesize loads or filters is very useful in microwave

engineering. E.g., A microwave filter constructed from microstrip. 918
459
Transmission line theory

Lumped circuit elements: resistors, capacitors,
inductors
neglect time delays (phase)
Distributed circuit elements: transmission lines
account for propagation and

time delays (phase change)
We need transmission-line theory whenever the

length of a line is significant compared with a
wavelength.
919
Lumped-element model of a
transmission line
Transmission line = the length of the line > /10 (or /30,
definition issue)
A very short piece (dz << ) of a transmission line can be
modeled with a lumped-element equivalent circuit.
920
460
Transmission line
Two conductors
Four per-unit-length parameters:
C = capacitance/length [F/m]
L = inductance/length [H/m]
R = resistance/length [/m] z
G = conductance/length [ S/m]
921
Transmission line model

i z, t
+++++++
----------
v z, t
i(z,t) Rz Lz i(z+z,t)
+ +
v(z,t) Gz Cz v(z+z,t)
- -
z
922
461
Transmission line model

i(z,t) Rz Lz i(z+z,t)
+ +
v(z,t) Gz Cz v(z+z,t)
- -
z
i ( z , t )
v( z, t ) v( z z , t ) i ( z, t ) Rz Lz
t
v( z z, t )
i ( z, t ) i ( z z , t ) v( z z, t ) G z C z
t
923
Transmission line equations

Thus,
v( z z, t ) v( z, t ) i ( z , t )
Ri ( z, t ) L
z t
i ( z z , t ) i ( z , t ) v( z z , t )
Gv( z z , t ) C
z t
Now let z 0:
v i
Ri L
z t Telegraphers
Equations
i v
Gv C
z t 924
462
Derivation of the wave equation

To combine these, take the derivative of the first
one with respect to z:
v
2
i i
R L
z 2
z z t Switch the
order of the
i i derivatives.
R L
z t z
v
R Gv C
t
v v
2
L G C
t t 2
925
Wave equation
v
2
v v v 2
R Gv C L G C
z 2
t t t 2
Thus, we have:
v
2
v v
2
RG v ( RC LG ) LC 0
z 2
t t
2
The same equation also holds for i v.

926
463
Time-harmonic version
Time-harmonic waves:
v 2
v v
2
RG v ( RC LG ) LC 0
z 2
t t
2
dV
2
RG V ( RC LG ) jV LC ( )V 0 2
dz 2
927
Time-harmonic wave equation

dV
2
RG V j ( RC LG )V LC V 2
dz 2
Note that the coefficient of V can express as a product
RG j ( RC LG ) LC ( R j L)(G j C )
2
Z R j L = series impedance/length
Y G jC = parallel admittance/length
dV2
Then we can write: ( ZY )V

dz 2
928
464
Solution
ZY dV 2
( )V
2
Let Then 2
dz 2
Solution: V ( z ) Ae Be z z
is called the "propagation constant.
( R j L)(G jC )
1/2
Convention:
z z e j / 2
principal square root

j attenuatio n constant

0, 0 phase constant 929
Phasor time function

Forward travelling wave (a wave traveling in the positive z direction):
V ( z ) V0 e z V0 e z e j z

v ( z, t ) Re V0 e z e j z e jt
Re V 0

e j e z e j z e
jt
The wave repeats when:
V0 e z cos t z 2
g
t 0
g
Thus:
V e
2
z
930
465
Attenuating wave
2 2
The local maximum values when:

g
t 0
g
V 0

e z
931
Phase velocity
Track the velocity of a fixed point on the wave (a point of constant
phase), e.g., the crest.
vp (phase velocity)
v ( z, t ) V0 e z cos(t z )
932
466
Phase velocity
Set: t z constant
dz
Take d/dt: 0
dt
dz

dt
In expanded form:

Hence
v v
Im ( R j L)(G jC )
p 1/ 2
p
933
Characteristic impedance Z0
I+ (z)
+
V+(z)
- z
A wave is traveling in the positive z direction.
V ( z)
Z0
I (z)
V ( z ) V0 e z V0
so Z0
I ( z ) I 0 e z I0
Note: Z0 is a number, not a function of z. 934
467
Use Telegraphers Equation:
v i
Ri L
z t
dV
so RI j LI Note:
dz V ( z ) V0 e z
ZI I ( z ) I 0 e z
Thus V0 e z ZI 0 e z
935

1/2
V Z Z
From this we have: Z 0 0
2
ZY
I0 Y
Using
Z R j L
Y G jC
1/2
We have R j L
Z0
G j C
936
468
Forward and reverse wave

V z V0 e z V0 e z

V0 e j e z e jz V0 e j e z e jz
wave in +z
Note: direction wave in -z
direction
v z , t Re V z e jt
V0 e z cos t z
V0 e z cos t z
937
Reverse wave
I - (z)
+
V -(z)
- z
A wave is traveling in the negative z direction.
V ( z) V (z)
Z0 so Z0
I ( z) I (z)
Note: The reference directions for voltage and current are the same as for the
forward wave. 938
469
Superposition of forward and backward

waves
I (z)
+
V (z)
- z
A superposition of forward and

General case: backward traveling waves:
V ( z ) V0 e z V0 e z
1
I ( z) V0 e z V0 e z
Z0
Note: The reference directions for voltage and current are the same for forward
and backward waves. 939
Recap
940
470
Recap
V z V0 e z V0 e z Guided wavelength g

I z
V z V z 2
e
m
0 0
e g
Z0 Z0
1

j R j L G jC 2
Phase velocity vp
1
R j L
2
Z0 vp [m/s]
G jC
941
Losless transmission line

R 0, G 0
j ( R j L)(G j C )
1/ 2
j LC
so 0
vp
LC
1/ 2
R j L L 1
Z0 Z0 vp
G jC C LC
Note: Now Z0 and vp are independent of frequency. 942
471
Lossless transmission line

1
vp
LC
If the medium between the two conductors is homogeneous
(uniform) and is characterized by (, ), then it can be shown
that
LC
The speed of an electromagnetic 1

cd
wave in a dielectric medium is
Thus v p cd
The phase velocity does not depend on the frequency, and it is
always the speed of light in this kind of material. 943
Terminated transmission line

Terminating impedance (load)
V z V e 0
z
V e0
z
Amplitude of voltage wave

propagating in positive z
direction at z = 0.
propagating in negative z
direction at z = 0.
Where do we assign z = 0?
z
The usual choice is at the load.
Note: The length l measures distance from the load l=-z 944
472
Change of reference plane

V z V0 e z V0 e z
What if we know amplitudes

V+ and V- at z=-l

Can we use z = - l as
a reference plane?
V0 V 0 V e V V 0 e
V0 V 0 V e
Hence
V z V e V e
z z
945
Change of reference plane


Compare:
V z V 0 e z V 0 e z
V z V e V e
z ( ) z ( )
Note: This is simply a change of reference plane, from z = 0 to z = -l. 946
473
What is V(-l )?
V z V0 e z V0 e z
What is V(-l )?
Wave propagating Wave propagating

forwards (load) backwards
The current at z = - l is then
l = distance away from load

947
Load reflection coefficient (L)
Total voltage at distance

l from the load

propagating towards load at Amplitude of voltage
L = Load reflection coefficient
the load position (z = 0). wave propagating away
from load at the load
position (z = 0).
Similarly,
Note: l = Reflection coefficient at z = - 948

l
474
Input impedance Z(-l )
Z0 ,
Input impedance seen looking towards

load at z = -l .
949
Relation between ZL and L

At the load (l = 0):
Z L Z0
L
Z L Z0
Recall:
Thus, Z L Z0 2
1 e
Z L Z0
Z Z 0
Z Z 2
1 L 0
e
L Z Z 0
950
475
Local impedance along terminated

transmission line
Simplifying, we have
Z L Z0 2
1 e
Z Z0 Z Z L Z0 Z L Z0 e
2
e l
Z Z0 L
0
2 l
Z L Z0 Z L Z0 e
Z Z 2
1 L 0
e e
Z L Z0
Z Z 0 e Z L Z 0 e e x e x e x e x
Z0 L
cosh x and sinh x

Z L Z0 e Z L Z 0 e
2 2
Z cosh Z0 sinh cosh l

Z0 L
Z0 cosh Z L sinh cosh l
Thus, we have
Z Z0 tanh
Z Z0 L
Z0 Z L tanh
951
Terminated lossless transmission line

j j
V V0 e j 1 L e2 j Impedance is periodic with

period g/2 because tan( )
V0 j repeats when
I e 1 Le2 j
Z0
1 Le 2 j
2
Z Z 0 2 j
1 Le
g
Z jZ0 tan g / 2
Z Z0 L
Z jZ tan
0 L
e x e x ex ex
cosh x and sinh x
Note: 2
e jx e jx
2
e jx e jx
tanh tanh j j tan
cosh jx cosx and sinh jx j sin x 952
2 2
476
Lossless transmission line

From now on in our transmission line discussion we will assume that
the transmission line is lossless.
Z0 ,
953
Matched load
Z0 ,
1 Matched load: (ZL=Z0)
No reflection from the load
For any l 954
477
Short-Circuit (S-C) load

2 Short-circuit load: (ZL = 0)
Z0 ,

Note: 2
g Always imaginary!
/ g
S-C can become an O-C with a

g/4 transmission line
955
Example
Find the voltage at any point on the line.
Z0
956
478
Example
Note:
At l = d :
Zin j d 1
V0 VTH e j 2 d
Zin ZTH 1 Le
Thus
Zin j d 1 Le j 2
V VTH e j 2 d
Z m ZTH 1 Le 957
Example
1 L e j 2 d
Zin Z d Z0 j 2 d
1 Le
958
479
Example
After algebraic manipulation we have
where
Therefore, we have the following alternative form for the result:
Z0 j d 1 L e j 2
V VTH e j 2d
Z0 ZTH 1 S Le
959
Example
Z0
Z 0 j d 1 L e j 2
V VTH e j 2 d
Z0 ZTH 1 S Le
Note: Voltage wave that would exist if there were no reflections from the load (a
960
matched load or a semi-infinite transmission line ).
480
Time-averaged power flow
Z0 ,
At a distance l from the load:
Note: V V0 e 1 Le 2
V0
I e 1 L e2
Z0
j
*
If Z0 real (low-loss transmission line) L e2 *L e2
Le2 L e2
*
pure imaginary
961
Time-averaged power flow
Z0 ,
Low-loss line
Lossless line ( = 0): P

962
481
Quarter-wave (g/4) transformer
Z jZ0T tan Z0 Z0T ZL

Zin Z0T L
Z0T jZ L tan Zin
g 2 g Matching:
in 0 Z in Z 0
4 g 4 2
Z 02T
Z0
ZL
This requires ZL to be real.
Thus
Hence
2
Z
Z in Z 0T Z 0 Z L
0T 1/2
ZL
963
Voltage Standing Wave Ratio (VSWR)
Z0 ,
1+ L
V (z)
1 V0
Vmax V0 1 L
1- L
Vmin V0 1 L z
z / 2
z0
Vmax 1 L
Voltage Standing Wave Ratio VSWR VSWR
Vmin 1 L 964
482
Example: Coaxial cable

Unit length
a
Coaxial cable b r ,
Parameters per unit length (C, L, G, R)
For a TEMz mode, the shape of the fields is independent of frequency, and hence
parameter calculation can be performed using electro- and magnetostatics.
2 0 r 2
C [F/m] G [S/m]
b b b
ln L 0 r ln [H/m] ln
a 2 a a
965
2 0 r
C [F/m]
b
ln
a LC 0 0 r r
0 r b
L ln [H/m]
2 a
This result actually holds for any transmission line.
966
483

L
For a lossless cable: Z0
C
2 0 r 0 r b
C [ F/m] L ln [H/m]
b 2 a
ln
a
r 1 b
Z 0 0 ln [ ]
r 2 a
0
0 376.7303 []
0
967
2
C [F/m] 0 r
b
ln
a
G C
2
G [S/m]
b
ln
a
This result actually holds for any transmission line.
968
484

G C

More generally:
G
tan The loss tangent would arise
C from conductivity effects.
G The loss tangent actually arises from both

tan conductivity loss and polarization loss
C (molecular friction loss).
Note: It is the loss tangent that is usually (approximately) constant for a

material, over a wide range of frequencies. 969
Note
General expression for loss tangent:

e j Effective permittivity that accounts for conductivity

' j ' ' j

e ' j e ' '
Loss due to molecular friction Loss due to conductivity
The loss tangent tan

' '
'' characterizes the attenuation of
tan e the electric field in a lossy
e' '
material.
970
485

R (resistance / unit length): R Ra Rb
1
Ra Rsa
Rs = surface resistance of metal 2 a
1
b , rb Rb Rsb
2 b
a , ra 1 1
Rsa Rsb
a a a b b
2 2
b a b
0 ra a 0 rb b
Is the skin depth, the distance over which the signal amplitude decreases to 1/e of
its surface value. 971

L
(1) Z 0lossless characteristic impedance of line (neglecting loss)
C
(2) LC 0 0 r r
G 1 1
(3) tan (4) R Rsa Rsb
C 2 a 2 b
Equations (1) and (2) can be used to find L and C if we know the material
properties and the characteristic impedance of the lossless line.
Equation (3) can be used to find G if we know the material loss tangent.
Equation (4) can be used to find R.

972
486

All four per-unit-length parameters can be found
from material parameters, Z0lossless and R.
L Z 0lossless
C / Z 0lossless
G C tan
RR
973
Limitations of transmission line theory

At high frequency, discontinuity effects can become important.
transmitted
incident
Bend
reflected
The simple transmission line model does not account for the bend.
ZTH
+- Z0 ZL
974
487
Idealizations of transmission line theory
At high frequency, radiation effects can become important.

We want energy to travel from the generator to the load without radiation.
ZTH
+- Z0 ZL
When will radiation occur?
975
Idealizations of transmission line

theory
The coaxial cable is a perfectly shielded
system there is never any radiation at any
frequency, or under any circumstances.
r a
z b
The fields are confined to the region

between the two conductors.
976
488

theory
The twin lead is an open type of transmission
line the fields extend out to infinity.
+ -
The extended fields may cause
interference with nearby objects.
(This may be improved by using
twisted pair.)
977
Note: Having fields that extend to infinity is not the same thing as having radiation.

theory
The infinite twin lead will not radiate by itself,
regardless of how far apart the lines are.
backward
forward
No attenuation on an infinite lossless line.
The forward and backward waves represent an exact solution

to Maxwells equations on the infinite line, at any frequency.
978
489

A discontinuity on the twin lead will cause radiation to occur.
Incident wave
pipe
Reflected wave
Note: Radiation
effects increase Incident wave
as the frequency
increases.
h bend
Reflected wave 979

To reduce radiation effects of the twin lead at discontinuities:
1) Reduce the separation distance h (keep h << ).

2) Twist the lines (twisted pair).
CAT 5 cable
(twisted pair)
980
490
Introduction to transmission
line theory
Recap and supplement
Transmission lines
982
491
Transmission line circuit
983
Boundary condition at load
This gives us the backward going

wave amplitude in terms of the
forward going wave amplitude.
984
492
Load reflections
985
Transmission line as an impedance

transformator
986
493
Transmission line as an impedance

transformator
987
Equivalent circuit
988
494
Special case: Short-circuit load
Note: Impedance seen at the source end is inductive. The transmission line appears
989
like an inductor.
Special case: Short-circuit load
990
495
Special case: Open-circuit load
Note: Impedance seen at the source end is capacitive. The transmission line appears
991
like a capacitor.
Special case: Open-circuit load
992
496
Matched load
For matched loads:

The load reflection coefficient L is zero.
There is no reflected wave generated at the load end ( i.e. V_ = 0 ).
The impedance seen at the source end is Z0 irrespective of the length of
the transmission line.
993
Periodic impedance transformation
Note: Impedance is periodic with period equal to half-wavelength. 994
497
Voltage Standing Wave Ratio (VSWR)
The interference of forward and backward going waves leads

standing-wave behavior.
Voltage Standing Wave Ratio = VSWR =

995
Equivalent circuit
Thus, we have produced the equivalent impedance

Z(z=-l) of the transmission line + load as seen by the
source.
996
498
Example
Reduce the transmission line circuit
to the Thevenin equivalent at the load terminals:
Note: Now we want to produce the Thevenin equivalent of the source +

transmission line as seen by the load. 997
Example
1 To find Zth short the voltage source and find the impedance
looking in from the load terminals
2 To find Vth remove the load and find the voltage at the load
terminals
998
499
Cascaded infinite transmission lines

Let us study the situation where we have two unmatched transmission
lines connected together.
Now
Boundary Conditions:
(1) At z=0 the voltage on both the transmission lines must be the same.
(2) At z=0 the current on both the transmission lines must be the same.
999
Example: Power splitting in microwave circuits
Input wave: Transmitted wave:

Reflected wave:
Goal: Express V+2 and V-1 in terms of V+1.
1 Cast the circuit in the following equivalent form and find V-1
2 Voltage V+2 is the same as the voltage across the impedance Zo2
in the equivalent circuit
1000
500
Transmission line junctions and

discontinuities
1001
Transmission line junctions and

discontinuities
1002
501
Multi-sectional transmission line

Question: How does one solve a problem like this?
1003
Example
Question: How does one solve a problem like this?
1 Replace the last line with a lumped equivalent impedance

(corresponding to an infinite line).
2 Replace the middle line with the impedance Z(z=- )
1004
502
Matching transmission lines

Is it possible to use a transmission line to perfectly match two dissimilar
transmission lines so that there is no reflection?
What is the appropriate impedance Zo2? What is the appropriate

length ?
Use a Quarter-wave transformer:
A quarter-wavelength long transmission

line inverts the normalized impedance1005
Matching transmission lines
1006
503
Smith chart
Smith chart
The Smith chart is a graphical aid that is very useful when
solving transmission line problems.
The key to its understanding is to realize that it is essentially a
polar plot of the voltage reflection coefficient ().
Let the reflection coefficient be expressed in polar form as
e j
Then the magnitude | | is plotted as a radius (| | 1) from the

center of the chart and the angle (-180 180 ) is
measured from the right-hand side of the horizontal diameter.
Any passively realizable (| | 1) reflection coefficient can
then be plotted as a unique point on the Smith chart.
1008
504
1009
Normalized impedance
The real utility of the Smith chart, however, lies in the fact that
it can be used to convert from reflection coefficients to
normalized impedances (or admittances), and vice versa using
the impedance (or admittance) circles printed on the chart.
When dealing with impedances on a Smith chart, normalized
quantities are generally used, which will be denoted by
lowercase letters (z).
The normalization constant is usually the characteristic
impedance of the line (Z0).
Thus, z= Z/Z0 represents the normalized version of the
impedance Z.
1010
505
Terminated transmission line
Total voltage at distance

l from the load

propagating towards load, Amplitude of voltage L Load reflection coefficient
at the load position (z = 0). wave propagating away
from load, at the load
position (z = 0).
Similarly,
Note: l reflection coefficient at z = - 1011

l
Generalized reflection coefficient
Impedance
calculation:
Generalized reflection coefficient: 1012
506
Generalized reflection coefficient
For
Re Z L 0
L 1
Proof:
Lossless transmission line ( = 0) L

RL jX L Z 0
RL jX L Z0
R Z 0 jX L
L
RL Z0 jX L
R Z 0 X L2
2
2
L L
RL Z0 X L2
2
1013
Complex plane
Im Decreasing l (toward load)
R j I
j 2
Le l
L
j L 2
L e L Increasing l (toward
L generator)
Re
L 2
Lossless line
1014
507
Impedance (Z) chart

1
Z Z0 ,
1
Z 1
Zn
Z0 1
Define:
Zn Rn jX n ; R j I
Substitute into above expression for Zn(-l ):
1 R j I
Rn jX n
1 j
R I
Next, multiply both sides by the RHS denominator term and equate real and
imaginary parts. Then solve the resulting equations for R and I in terms of Rn
and Xn. This gives two equations. 1015
Impedance (Z) chart

1) Equation 1:
2
Rn 2 1 Equation for a circle
R I in the plane:
1 Rn 1 Rn
I
1016
508
Impedance (Z) chart

2 2
1 1
R 1 I
2
2) Equation 2:
Xn Xn
I
Equation for a circle in the plane:
0 Xn 1 Xn 1
1 Xn
Xn 0
R
X n 1
0 X n 1
X n 1
1017
Impedance (Z) chart

Smith Chart (Z-Chart) Clear-cut version
Xn = 1
Rn = 1
Xn = -1
1018
509
Impedance (Z) chart
Xn = 1 plane
Rn = 1
Xn = -1
1019
Impedance transformation along

transmission line
Define a position dependent reflection

coefficient as: (z) is a complex number
of magnitude never
greater than unity.
Normalized resistance
Normalized reactance
Thus
1020
510
Complex plane (-plane)
1021
-plane with normalized resistance curves
1022
511
-plane with normalized reactance curves
1023
Smith chart
The Smith chart enables

one to read values of both
the normalized resistance
Rn(z) and the normalized
reactance Xn(z) values on
the -plane.
1024
512
Which chart to use?

When adding elements in series use Z-chart.
When adding elements in parallel use Y-chart.
1025
Example: How to use the Smith chart

for impedances
We wish to determine the input impedance of a transmission line into which have
been inserted series components, and which is terminated into a mismatched load
with complex impedance.
A computation would require several successive applications:
Zk ZL= load impedance

s = distance from load ZL
One for every section of line.
513
Example
In the Smith chart, the sequence of operations is
carried out graphically: every section of line produces
a rotation around the center of the chart, the rotation
angle being related to the length of the section,
divided by the wavelength.
To insert a series impedance, one simply adds its
value (normalized to the line impedance) to the one at
the line's input, obtained on the Smith chart.
When adding a reactance, one moves on a circle R =
constant; when adding a resistance, on a circle X =
constant.
1027
Example
Considering the circuit below the
procedure followed is:
1. The load terminating the line (at
right) has an impedance ZL = (2.15 -
j3)Zc.
The corresponding point A in the Smith chart
chart is located at the coordinates 2.15
- j3.
2. The line section, long of 0.087,
produces a rotation from A to B by an
angle 0.087 x 720 = 62.5 around the
chart's center, moving towards the
generator (decreasing values of z),
i.e., clockwise.
The impedance at point B read on the
chart is (0.3 - j1)Zc.
One wavelength, , corresponds to
two full turns, i.e., a 720 angle.
Zc is the characteristic impedance of the line. 1028
514
Example
3. One then adds an impedance of j0.8
Zc, bringing to the point C at (0.3 -
j0.2)Zc.
The move from B to C follows a circle Smith chart
R = constant.
4. The section of line between C and D
produces a second rotation around the
chart's center, by an angle 0.174 x 720
= 125.3.
The input impedance of this section of
line at point D is (0.65 + j1)Zc.
5. To this last value is added the
impedance of the series capacitor -j
0.6Zc, yielding point E at (0.65 +
j0.4)Zc.
The displacement is done along a circle
R = constant (as between B and C).
1029
Example
6. The addition of a series resistance
between F and G leads to a move
along a circle X = constant, between
coordinates R = 1.6 Zc and R = 3 Zc.
7. Finally, the line section between G
and H produces a last rotation around
the center of the chart, by an angle 0.1
x 720 = 72 , yielding the input
impedance to the whole assembly as
(0.9 j1.15)Zc

1030
515
Example
1031
Example: Calculation impedances

using Maple
1032
516
Example
1033
Example
1034
517
Example
1035
Example
ZL = (0.9 j1.13) Zc
1036
518
Smith chart for admittances

When connecting components in shunt on the line (which is most often preferred at
microwaves), one wishes to use local admittances, rather than impedances.
The relationship of the local admittance, Yn, normalized to the characteristic
admittance Yc of the line, with the reflection factor, is given by:
It is the same relationship as the one for the impedance, in which is simply
replaced by .
In the Smith chart, the value of the normalized admittance Yn/Yc is obtained by
taking the image of the point Zn/Zc across the center of the chart.
This actually provides a simple graphical method to determine the inverse of a
complex number.
1037
Example: How to use the Smith

chart for admittance
The input admittance of a transmission line,
into which a capacitor, an inductor and a
resistor have been shunt-connected between
sections of line, is to be determined in the
following circuit.
1038
519
Example
The procedure followed is exactly the same as the one outlined previously for
series-connected elements in the Smith chart for impedances.
The very same sequence A-H is followed, this time for admittances in the Smith
chart.
The input admittance obtained is YH = (0.9 j1.13) Yc.
YH = (0.9 j1.13) Yc
1039
Example: Simultaneous use of the Smith

chart for shunt and series components
We may, of course, utilize the Smith chart to connect both series impedances and
shunt admittances, taking advantage of the central imaging property indicated in
A T-circuit, having one shunt susceptance between two series reactances, is

connected to a load ZL
ZL = (2.5 j0.5)Zc
1040
520
Example
1. The load impedance is here ZL = (2.5 j0.5)Zc so that the starting point A is
located at the coordinates 2.5 - j0.5 in the Smith chart for impedances.
2. A series capacitance is then added.
Remaining on the impedance chart, one moves along an R = constant circle, by a
distance corresponding to the reactance j2Zc; i.e., from - j0.5 to - j2.5 , reaching
point B.
ZL = (2.5 j0.)Zc
1041
1041
Example
3. The next component to be connected is a shunt susceptance.
This means that the input admittance must be determined.
This is obtained by jumping over the center of the chart, from point B to its image
B' , this time in the admittance chart.
4. The addition of a shunt susceptance -j0.6Yc is obtained by moving along a circle
of constant conductance G = constant, reaching point C' (still in the admittance
chart).
ZL = (2.5 j0.)Zc
1042
1042
521
Example
5. The next component is connected in series, so that one has to return to the
impedance chart.
This is done by going to point C, symmetrically located across the center of the
chart.
6. The second series reactance -j2Zc is then added, in the same way as the first one
(moving along a circle R = constant).
ZL = (2.5 j0.)Zc
1043
1043
Example
The point D reached is, for the particular set of values selected, the center of the
chart.
The T-circuit, containing only lossless reactive components, has actually matched
the load ZL to the transmission line of characteristic impedance Zc (reflectionless
match).
ZL = (2.5 j0.)Zc
1044
522
Example: Calculation using Maple
ZL = (2.5 j0.5)Zc
1045
Example: Calculation using Maple

This T-circuit has matched the load ZL to
the transmission line of characteristic
impedance Zc (reflectionless match).
ZD = (1.0 + 0.0j)
ZL = (2.5 j0.5)Zc
1046
523
Impedance matching
Impedance matching
Impedance matching is an important practical problem in
microwave engineering.
In practice, impedance matching is a part of the larger design
process for a microwave component or system.
The basic idea of impedance matching is illustrated in figure,
which shows that an impedance matching network is placed

between a load impedance and a transmission line.
1048
524
Matching network
The matching network is ideally lossless, to avoid
unnecessary loss of power, and is usually designed so
that the impedance seen looking into the matching
network is Z0.
Then reflections are eliminated on the transmission
line to the left of the matching network, although
there will be multiple reflections between the
matching network and the load.
1049
Benefits of impedance matching

Impedance matching or tuning is important for the
following reasons:
Maximum power is delivered when the load is matched to
the line (assuming the generator is matched), and power
loss in the feed line is minimized.
Impedance matching of receiver components (antenna,
low-noise amplifier, etc.) improves the signal-to-noise ratio
of the system.
Impedance matching in a power distribution network (such
as an antenna array feed network) will reduce amplitude
and phase errors.
1050
525
Matching load for maximum power

In many RF circuits it is extremely desirable to be able to transfer
maximum possible time-average power to a load impedance:
What is the time-average power delivered to the load?
Maximum time-average power delivered to the load:
1051
Matching load in transmission lines
1052
526
Example: Matching load in

transmission line
1 Find L
2 Find such that the impedance Z(z=-) has a real part of 50
1053

transmission line
Find the desired using Smith chart
1054
527

transmission line
But there is a problem !
Well get rid of this unwanted reactance using stub tuners.

Stub tuners are short stubs of transmission lines that are
used to cancel out unwanted reactances in microwave circuits. 1055
Choice of matching network

As long as the load impedance, ZL, has some nonzero
real part, a matching network can always be found.
Many choices are available, however, so factors that
may be important in the selection of a particular
matching network include the following:
ComplexityNearly always in case of engineering
solutions, the simplest design that satisfies the required
specifications is the most preferable.
A simpler matching network is usually cheaper, more reliable, and
less lossy than a more complex design.
1056
528
Choice of matching network

BandwidthAny type of matching network can ideally
give a perfect match (zero reflection) at a single frequency.
In many applications, however, it is desirable to match a load over a
band of frequencies.
There are several ways of doing this with, of course, a corresponding
increase in complexity.
ImplementationDepending on the type of transmission
line or waveguide being used, one type of matching network
may be preferable compared to another.
For example, tuning stubs are much easier to implement in
waveguide than multi-section quarter-wave transformers.
AdjustabilityIn some applications the matching network
may require adjustment to match a variable load impedance.
Some types of matching networks are better than others in this
regard.
1057
Basic impedance matching

methods
Reactive lumped elements
(L-section networks)
Inductors and capacitors
Distributed elements
Tuning stub
Quarter-wave transformer
Resistive matching
(attenuator)
Suits only for the applications
which do not suffer from the
limited power and additional
heat => generally not
acceptable
1058
529
Matching with lumped elements

Probably the simplest type of matching network is the L
section, which uses two reactive elements to match an
arbitrary load impedance to a transmission line.
There are two possible configurations for this network, as
shown in figure below.
In either of the configurations the reactive elements may be

either inductors or capacitors, depending on the load
impedance.
Thus, there are eight distinct possibilities for the matching
circuit for various load impedances.
1059
L-matching possibilities
Eight possible configurations of the discrete two-

component matching networks.
1060
530
Matching with lumped elements

If the frequency is low enough and/or the circuit size
is small enough, actual lumped-element capacitors
and inductors can be used.
This may be feasible for frequencies up to about 1
GHz or so, although modern microwave integrated
circuits may be small enough so that lumped
elements can be used at higher frequencies as well.
There is a large range of frequencies and circuit sizes
where lumped elements may not be realizable.
This is a limitation of the L section matching
technique.
1061
Matching with distributed

elements
Usually, distributed elements are suitable
above 1 GHz.
Basic methods are tuning stub and quarter-
wavelength transformer.
Distributed-component matching circuits can
be integrated on the surface of a printed circuit
board.
1062
531
Single-stub tuning
A matching technique that uses a single open-circuited or
short-circuited length of transmission line (a stub),
connected either in parallel or in series with the transmission
feed line at a certain distance from the load, as shown in figure
below.
1063
Stub tuning parameters

A stub tuning circuit is convenient from a microwave
fabrication aspect, since lumped elements are not
required.
The shunt tuning stub is especially easy to fabricate in
microstrip or stripline form.
In single-stub tuning, the two adjustable parameters
are the distance (d) from the load to the stub position,
and the value of susceptance or reactance provided
by the shunt or series stub.
1064
532
Matching conditions
For the shunt-stub case, the basic idea is to select d so
that the admittance Y, seen looking into the line at
distance d from the load is of the form Y0 + jB.
Then the stub susceptance is chosen as -jB, resulting
in a matched condition.
For the series stub case, the distance d is selected so
that the impedance Z, seen looking into the line at a
distance d from the load is of the form Z0 + jX.
Then the stub reactance is chosen as -jX, resulting in
a matched condition.
1065
Open-circuited stub vs. short-

circuited stub
The proper length of open or shorted transmission line can
provide any desired value of reactance or susceptance.
For a given susceptance or reactance, the difference in lengths
of an open- or short-circuited stub is /4.
For transmission line media such as microstrip or stripline,
open-circuited stubs are easier to fabricate since a via hole
through the substrate to the ground plane is not needed.
For lines like coax or waveguide short-circuited stubs are
usually preferred, because the cross- sectional area of such an
open-circuited line may be large enough (electrically) to
radiate, in which case the stub is no longer purely reactive.
1066
533
Stub tuners
Stub tuners are short stubs of transmission lines that are used to cancel
out unwanted reactances in RF circuits.
1067
Matching load using series stub tuner

From the previous example
k=2/
1068
534
Smith chart: Open-circuit stub tuner
1069
Normalized admittances on Smith chart
1070
535
Matching load using parallel stub

tuner
It is easier to work with admittances than impedances when parallel
stubs are used since admittances in parallel can simply add.
1071

tuner
1072
536

tuner
We have up to now:
k=2/ 1073

tuner
Open-Circuit stub tuner on Smith chart
1074
537
Double-stub tuning
The single-stub tuners are able to match any load impedance
(as long as it has a nonzero real part) to a transmission line, but
suffer from the disadvantage of requiring a variable length of
line between the load and the stub.
This may not be a problem for a fixed matching circuit, but
would probably pose some difficulty if an adjustable tuner
was desired.
In this case, the double-stub tuner, which uses two tuning
stubs in fixed positions, can be used.
Such tuners are often fabricated in coaxial line, with adjustable
stubs connected in parallel to the main coaxial line.
The double-stub tuner cannot match all load impedances.
1075
Double-stub tuner
The double-stub tuner circuit the load may be an arbitrary distance from the
first stub.
The stubs shown in figure are shunt stubs which are usually easier to
implement in practice than are series stubs: the latter could be used just as
well, in principle.
In either case, the stubs can be open-circuited or short-circuited.
1076
538
Quarterwave transformer
The quarter-wave transformer is a simple circuit for
matching a real load impedance to a transmission
line.
An additional feature of the quarter-wave transformer
is that it can be extended to multi-section designs for
broader bandwidth.
If only a narrowband impedance match is required, a
single-section transformer may suffice.
1077
Quarterwave transformer
One drawback of the quarter-wave transformer is that
it can only match a real load impedance.
However, a complex load impedance can always be
transformed to a real impedance by using an
appropriate length of transmission line between the
load and the transformer, or by using an appropriate
series or shunt reactive stub.
These techniques will usually alter the frequency
dependence of the equivalent load, which often has
the effect of reducing the bandwidth of the match.
1078
539
0/4-transformer
The single-section quarter wave matching transformer circuit
is shown in figure .
The characteristic impedance of the matching section is
Z1 Z 0 Z L
At the design frequency f0, the electrical length of the

matching section is 0/4, but at other frequencies the length is
different, so a perfect match is no longer obtained.
1079
Example
Matching with a quarter-wave transformer:
An 8 load is matched to a 50 line through a
20 transformer.
1080
540
Quarter-wave transformer
k=2/
1081
/4-transformer on Smith chart

For example:
1082
541
Waveguide theory
Reflection and
transmission of EM waves
at interfaces
Waves at an interface
Let us study a plane wave given by:
incident upon an interface between medium i and medium t
When the incident wave strikes the interface, it generates a

transmitted wave and a reflected wave.
We need to find the amplitudes of these reflected and transmitted
waves.
1084
542
Incident, transmitted, and reflected

waves E-fields
Note: In the above equations, Er and Et are the unknowns that we need to find
in terms of the incident field amplitude Ei. 1085
Boundary conditions at interface E-

fields
Boundary conditions:
(1) At z = 0 the E-field parallel to the interface must be continuous
This gives:
1086
543
Boundary conditions at interface H-

fields
(2) At z = 0 the H-field parallel to the interface must be continuous (no

surface currents)-
The wave impedance

of the medium =(/)
This gives: The other equation was:
1087

coefficients
These are three different ways

of writing the same results using:
1) Impedances
2) Wavevectors
3) Refractive indices
1088
544
Reflection at the perfect metal surface
For perfect metals:
Reflection and transmission coefficients for free-space and perfect metal

interface:
1089

standing waves
Standing waves
The interference between

incident and reflected waves
gives rise to standing waves
1090
545

currents
The magnetic field is parallel to

the surface and peaks at the metal
surface.
There must be surface currents
(recall the boundary conditions
for the magnetic field). 1091

coefficients for non-perfect metal
For non-perfect metals:
Reflection and
transmission coefficients
for non-perfect metal or
good conductor are
complex numbers. 1092
546
Transmitted wave in non-perfect metal
The wave penetrates a few skin depths into the non-perfect metal.
Current flows within a layer a few skin

depths thick inside the non-perfect metal
or good conductor.
1093
Wave reflection and VSWR
E max Ei 1 E min Ei 1
VSWR is found by taking

Voltage Standing
VSWR
E max

1 the ratio of the magnitude
of the maximum field value
Wave Ration:
E min 1 (wherever that occurs) to
the minimum field value
(wherever that occurs). 1094
547
Wave reflection and VSWR

No reflection (|| = 0)
VSWR
E max

1 1
E min 1
Small reflection (|| = 0.25) VSWR=1.66
Large reflection (|| = 0.75) VSWR=7
Complete reflection (|| = 1) VSWR=
1095
Waves in free space and in

transmission lines
Free space Transmission line
1096
548
Comparison: Waves at transmission lines

and interfaces
1097
Three-layer structure
How do we calculate the reflection coefficient for the structure?
Note: We can use the same method solving the equivalent transmission line
problem!
1098
549
1099
1100
550
Dielectric Anti-Reflection (AR)

coatings
1101
Dielectric Anti-Reflection (AR) coatings
1102
551
Dielectric Anti-Reflection (AR) coatings

Question: How do /4 long matching layers work?
1103
Dielectric High-Reflection (HR) coatings

What if we want to increase the reflectivity?
A periodic stack of high index-low index layers can be used as an HR
coating!
1104
552
Dielectric High-Reflection (HR) coatings
1105
Dielectric High-Reflection (HR)

coatings
1106
553
Dielectric High-Reflection (HR)

coatings
Magnitude of can be made arbitrarily close to unity by:
1) Choosing a large number N of pairs of high index-low index layers.

2) Choosing the difference between 1 and 2 to be large.
1107
Waveguide theory
Guided waves in parallel

plate metal waveguides
554
Parallel plate metal waveguide

Let us study a parallel plate waveguide (shown below).
We have studied such structures in the context of transmission lines.
Thus, we know that they can guide TEM waves (Transverse Electric
and Magnetic). in which both the electric and magnetic fields point in
direction perpendicular to the propagation direction.
But these structures can guide more than just the TEM waves that
we have considered so far.
1109
Complex wave equations

Consider a parallel plate waveguide:
1110
555
TE (guided) modes
1111
TE modes
1112
556
Dispersion relation for TE mode
1113
Cut-off frequency for TE mode
1114
557
TE guided modes Magnetic field
1115
TE guided modes
A parallel plate waveguide:
1116
558
TM guided modes
1117
TM guided modes

1118
559
TM guided modes Electric field
1119
TM guided modes Dispersion

relation
1120
560
TM guided modes Cut-off frequency
1121
TM guided modes - Summary
Note: The TM 0 mode is just the TEM mode that we were dealing with
1122
transmission lines.
561
Waveguide theory
Rectangular metal
waveguides
Rectangular metal waveguides

Rectangular metal waveguides are usually made of
copper and the best are gold plated.
They are commonly used to guide electromagnetic
power when dealing with high power levels (radars,
satellite and space communications, wireless/mobile
base stations, etc.)
Integrated versions are used in sub-millimeter
wavelength ultrahigh speed electronics (e.g. Gunn
oscillators, superconducting THz electronics, nonlinear
Schottky diode mixers) operating at frequencies between
300 GHz to 1000 GHz.
1124
562
Rectangular metal waveguide

Like a parallel plate metal waveguide that is closed by metal walls on
the remaining two sides.
1125
TE guided modes
1126
563
TE guided modes
1127
TE guided modes
1128
564
TEmn modes
Note: The TE00 mode does not exist (i.e. it corresponds to field being
trivially zero everywhere) 1129
Cut-off frequencies of TEmn modes
Note: If the frequency is less than the cut-off frequency then kz

becomes entirely imaginary and the mode does not propagate but decays
1130
exponentially with distance.
565
Dispersion relations of TEmn modes
1131
TM guided modes
1132
566
TM guided modes
1133
TM guided modes
1134
567
TM guided modes
Note: The TM00 , TMm0 TM0n modes do not exist (i.e. they correspond to
field being trivially zero everywhere). 1135
Cut-off frequency of TMmn modes
Note: If the frequency is less than the cut-off frequency then kz

becomes entirely imaginary and the mode does not propagate but decays
exponentially with distance. 1136
568
Dispersion relation of TM modes
1137
Waveguide to coaxial cable

adapters
Sometimes it is necessary to transfer power between a waveguide and
a coaxial cable.
Loop solution Pin solution
1138
569
Waveguide theory
Dielectric waveguides
Dielectric waveguide
TE-wave undergoing total internal reflection:
1140
570
Dielectric waveguide Total internal

reflection
One can have a guided wave that is bouncing between two
dielectric interfaces due to total internal reflection and moving
in the z-direction.
1141
Dielectric slab waveguide
1142
571
Comparison between dielectric and metal

waveguides
1143
Optical communications: Optical

fibers
1144
572
What is cut-off in optical fiber case ?
So what does cut-off mean? It means that the

wave is no longer being guided through total
internal reflection since i < c.
1145
Cut-off frequency
1146
573
Radio wave propagation in

more or less free space
Propagation mechanisms
Introduction
The majority of applications of microwaves are related
to radar and communication systems.
Microwave communication systems handle a large
fraction of the worlds international and other long-haul
telephone, data, and television transmissions.
Most of the current wireless telecommunications
systems, such as terrestrial radio links, satellite
communications , wireless local area computer
networks (WLAN), cellular systems, and global
positioning satellite (GPS) systems, operate in the UHF
EHF range, and thus rely on microwave technology.
1148
574
International organizations
The use of radio frequencies is controlled by

the international organizations:
ITU (International Telecommunications Union)
and ETSI (European Telecommunication
Standards Institute), for example
Allocation the limited resources of frequency spectrum
and geostationary orbits.
Formulation of rules and procedures to avoid mutual
disturbance of services.
1149
Frequency allocation
1150
575
Maxwells equations
The initial understanding of radio wave propagation goes back to the
work of James Clerk Maxwell, who in 1864 formulated the theory
of electromagnetic propagation which predicted the existence of
radio waves.
Electromagnetic waves propagate through environments where they
are reflected, scattered, and diffracted by walls, terrain, buildings,
and other objects.
The ultimate details of this propagation can be obtained by solving
Maxwells equations with boundary conditions that express the
physical characteristics of these obstructing objects.
This requires the calculation of the Radar Cross Section (RCS) of
large and complex structures.
Since these calculations are difficult, and many times the necessary
parameters are not available, approximations have been developed
to characterize signal propagation without resorting to Maxwells
equations.
1151
Propagation model categories

The simplest model for signal propagation: free space path
loss.
A signal propagating between two points with no attenuation
(absorption) or reflection follows the free space propagation
law.
Ray tracing propagation models are used to approximate wave
propagation according to Maxwells equations, and are
accurate models when the number of multipath components is
small and the physical environment is known.
Ray tracing models depend heavily on the geometry and
dielectric properties of the region through which the signal
propagates.
Empirical models use parameters based on measurements for
both indoor and outdoor channels.
When the number of multipath components is large, or the
geometry and dielectric properties of the propagation
environment are unknown, statistical models must be used. 1152
576
Propagation models
The most common approximations use ray-tracing techniques.
These techniques approximate the propagation of electromagnetic waves
by representing the wavefronts as rays: the model determines the reflection
and refraction effects on the wavefront but ignores the more complex
scattering phenomenon predicted by Maxwells coupled differential
equations.
The simplest ray-tracing model is the two-ray model, which quite
accurately describes signal propagation when there is one direct path
between the transmitter and receiver and one reflected path.
The reflected path typically bounces off the ground, and the two-ray model
is a good approximation for propagation along highways, rural roads, and
over water.
More complex General Ray Tracing (GRT) models are needed to describe
additional reflected, scattered, or diffracted components
Many propagation environments are not accurately enough described by
ray tracing models.
In these cases it is common to develop analytical models based on
empirical measurements.
These are called empirical models.
1153
Statistical models
Often because of complexity and variability of the radio channel it
doesn't pay to develop a deterministic channel model.
For these cases statistical models are used.
The attenuation caused by signal path obstructions such as buildings
or other objects is typically characterized statistically.
Statistical models are also used to characterize the constructive and
destructive interference for a large number of multipath components.
Statistical models are most accurate in environments with fairly
regular geometries and uniform dielectric properties.
Indoor environments tend to be less regular than outdoor
environments, since the geometric and dielectric characteristics
change dramatically depending on whether the indoor environment
is an open factory, office, flat or metal machine factory.
For these environments computer-aided modeling tools are available
to predict signal propagation characteristics.
1154
577
Microwave propagation
In free-space, electromagnetic waves propagate in straight
lines without absorption or other adverse effects.
Free-space, however, is an idealization that is only
approximated when microwave energy propagates through
the atmosphere or in the presence of the earth.
In practice the performance of a communication, radar, or
radiometry system may be seriously affected by propagation
effects such as reflection, refraction, attenuation,
diffraction, or scattering.
It is important to realize that more or less free-space
propagation effects cannot be quantified in any exact or
rigorous sense, but can only be described in terms of their
statistics.
1155
Radio propagation mechanisms

Propagation in free space is straightforward.
Received power is proportional to 1/d in free space.
(d = distance
Receiving power additionally influenced by between sender
fading (frequency dependent) and receiver)
shadowing
reflection at large obstacles
refraction depending on the density of a medium
scattering at small obstacles
diffraction at edges
fading shadowing reflection refraction scattering diffraction 1156
578
Electromagnetic spectrum and

atmospheric opacity
1157
Low frequency
1158
579
Microwave frequency bands
FSW = Free Space Wave, SW = Scattered Wave
1159
ELF and ULF

Extremely Low Frequency (ELF), f =30-300
Hz, =10000-1000km.
Ultra Low Frequency (ULF), f =0.3-3 kHz,
=1000-100km.
Waves do not penetrate ionosphere.
Antennas would be extremely large (yet << ).
Very low data rate.
Submarine communications, radio
communications to mines.
1160
580
VLF
Very Low Frequency: f =3-30kHz, =100-10

km
Waves do not penetrate ionosphere guided
waves between the earth and the ionosphere.
Antenna height << .
Telegraph to sea vessels, time standards.
1161
LF
Low Frequency (LF), f =30-300 kHz, = 10-1
km.
Waves do not penetrate ionosphere guided
wave between the earth and the ionosphere.
Surface wave propagates beyond the horizon.
Antenna height <<.
Broadcasting service, radio communications to
boats, radio navigation (Decca, Loran).
Interference between the ionosphere wave and
the ground wave fading.
1162
581
Propagation mechanism: Surface

(ground) wave
Surface wave or ground wave for LF and MF broadcasting (f < 0.5MHz)
1163
Ground wave propagation over a

curved surface
Sommerfelds theory or its approximations for the ground wave
propagation are valid for a flat surface and thus, can be used if the
following conditions for the distance and the antenna height are
valid
a is the radius of the earth

Example: Let f = 300 MHz dmax = 34 km and hmax =180 m.
If the flat ground assumptions are violated, the curvature of the

Earth should be taken into account in the modeling.
The higher the frequency the higher the losses.
Thus, the best frequency regime for the ground wave propagation is
VLF-MF bands (3 kHz-3 MHz).
1164
582
Note: Conductor vs. insulator

Radio wave see conducting, non-magnetic
medium (r=1)
as a good conductor if
as a good insulator if
1165
Example
Dry (bad) ground: r 4, 0.001 S/m
good insulator if f >> 4.5 MHz (VHF and higher
frequencies)
good conductor if f << 4.5 MHz (LF and lower
frequencies)
Sea water: r 80, 5 S/m
good insulator if f >> 1.1 GHz (EHF and higher
frequencies)
good conductor if f << 1.1 GHz (HF and lower
frequencies)
Note: 0.001 S/m=1 mmho/m 1166
583
Note: Skin depth vs. attenuation

coefficient
Skin depth () is a distance s0= which the wave [y(s)] decreases
to 1/e of its initial (surface) value [y(0)]
y(s=)=(1/e)y(0)0.37y(0)
where y(s)=y(0)exp(-s), the attenuation coefficient (attenuation

per unit length)
If set s= then
y(s=)= y(0)exp(-)= y(0)exp(-1) =1
The inverse of the skin depth is =1/ is the attenuation

coefficient.
1167
MF
Medium Frequency (MF), f = 0.3-3 MHz, = 1-
0.1 km.
Ionosphere waves attenuate in daytime (D-layer
absorption).
Ionospheric refraction at night from E- and F-
layers (the range of ionospheric wave at night
1000 km)
Surface wave propagates beyond the horizon but
attenuates (range 100km).
Antenna heights < .
1168
584
HF
High Frequency (HF), f =3-30MHz, =100-

10m
Waves can penetrate the ionosphere, E- and F-
layers can cause refraction.
Antennas: monopoles, dipoles and antenna
arrays.
Very long communications can be achieved.
1169
Propagation mechanism: Ionosphere

(sky) wave
Medium (MF) and short wave (HF) communication over long distances
1170
585
VHF
Very High Frequency (VHF), f = 30-300 MHz = 10-1
m.
Usually no refraction of waves from the ionosphere.
Direct wave is the main propagation mechanism but
scattering from troposphere and ionosphere, and
troposphere ducting (waveguide type propagation) is
possible.
Hills attenuates waves, reflections from surfaces,
multipath propagation.
Helix and yagi antennas usable.
100 km range for FM broadcasting service.
Radio beacons, mobile communication in sea, air and land
applications.
1171
Ultra High Frequency (UHF)

f =0.3 -3 GHz =100-10 cm.
Scattering from troposphere can provide even 300-600 km
communication links.
Direct wave is the main propagation mechanism and tropospheric
ducting (waveguide type propagation) is possible.
Hills shadow propagating waves.
Antennas are usually Yagis and parabolic reflectors (dishes).
TV broadcasts, radio navigation (GPS), air surveillance radar,
mobile communications etc.
Line-of sight communications (LOS)
Rain attenuation is negligible.
1172
586
Super High Frequency (SHF)

f = 3-30 GHz = 10-1 cm.
Direct wave
Hydrometeors (rain and fog) attenuate waves
(>10GHz)
Horn antennas and parabolic reflectors.
Radio links, satellite communications, radars,
remote sensing, mobile communications,
WLAN, WPAN, etc.
Line - of sight communications
1173
Extremely High Frequency (EHF)
f =30-300GHz =10-1mm.
Direct wave limited by strong attenuation to
gases, rain and mist.
Atmosphere gases absorb waves (windows).
Small parabolic reflectors, lenses.
Short line-of sight communication links,
remote sensing from satellite
Scientific and experimental use.
1174
587
1175
Propagation mechanisms
Free space and reflected wave
TV and radio broadcast, navigation, wireless communication

etc. f > 30 MHz
1176
588
Troposphere scattering
Communication over horizon.
Using UHF and higher frequencies.
Military applications.
1177
Radio propagation mechanisms
1178
589
Line of sight (LOS) transmission link
1179
Friis formula for a radio link

Assume two antennas whose mutual distance is r and gains
along the line of sight G1 and G2
Antenna 1 transmits power P1. What is the received power, P2?
The power density, S1, at the antenna 2:
Antenna 2 receives the power:
1180
590
Friis formula for a radio link

Since power density in terms of the electric field
is
The wave impedance (a
vacuum value is 120377)
Friis formula can be given in terms of radiated

field by the transmitter
Here it is assumed that the antennas are in free

space and in the far field.
1181
Note: Far-field concept

The near and far field are defined in terms of the Fraunhofer distance.
The Fraunhofer distance is the value of:
d f D

D 2

where D is the largest physical dimension of the radiator (or the diameter of
the antenna) and is the wavelength of the radio wave.
This distance provides the limit between the Fresnel and Fraunhofer region.
Additionally, a far-field region distance must satisfy the condition
where df is the far-field distance.

The far-field distance is the distance
from the transmitting antenna to the
beginning of the Fraunhofer region, or far field.
1182
591
Far-field Radiation zone Free

space
In the far-field, the shape of the antenna pattern is
independent of distance from the source.
The far-field is also frequently referred to as the
"radiation zone", or "free-space".
A more precise definition is given by the propagation
properties.
The radiation zone is important because far-field falls off in
amplitude by 1/r (E1/r, r>>).
This means that in the far-field the total energy per unit
area at a distance r is proportional to 1/r2.
The area of the sphere is proportional to r2, so the total
energy passing through the sphere is constant.
1183
Environment factor
The effect of real conditions for the propagated
field can be modeled by environment factor,
F, which is the ratio of free-space and true
fields, E1 and E1true, respectively:
E1true =FE1
Hence, free-space Friis formula can be
generalized to real conditions with the
environment factor
1184
592
Design inequality
Friis formula leads to a design inequality for a radio link
where P2,min is the minimum value for the received power.

Minimize the right hand side of the inequality by choosing first the
frequency f=c/ F and P2,min which is restricted by noise.
Antenna gains, G1 and G2, and transmitted power P1 state antenna
technology and costs.
The environment factor, F, depends heavily on frequency and
propagation mechanism.
The exact value is indeterminable but its average and statistics are
interesting average field and its variation.
1185
Propagation in more or less free-

space
Propagation in free space:
An antenna radiation power P1 gives rise to electric field
Propagation near ground:
1186
593
Ray theory
In the far field the Poynting vector has only a radial
component then propagation can be modeled as flux
tubes in which power is conserved
Thin flux tubes are rays that are
straight lines in a homogeneous medium
curved lines in a nonhomogeneous medium
Note: The Poynting vector S=ExH represents the energy flux (in W/m2) of an electromagnetic
1187
field.
Ray theory
The fundamental
assumption of the ray
theory is that obstacles
shadow only those rays
that are impinging on it
without affecting the
others.
The validity of the ray
theory is the better the
greater is the frequency.
1188
594
Atmospheric propagation
While the electromagnetic properties of air are
very close to those of a vacuum (r=1, r =1), they
are not rigorously identical.
When considering a long trajectory (path) within
the atmosphere, the variations due to the pressure
p, the temperature T, and the moisture level
have to be considered.
Electromagnetic rays do not travel in a straight
line within the atmosphere, but follow slightly
curved trajectories.
1189
Inhomogeneous atmosphere
The troposphere is very slightly inhomogeneous
caused by the gravitation which makes the gas
mixture called air thinner as height increases.
Free electrons in the ionospheric layers change the
effective permittivity as the function of height.
Inhomogeneous media bend radio waves or radio
rays.
1190
595
Atmosphere
1191
Standard atmosphere
The relative permittivity of air at microwaves is given by an
experimental expression:
where p is the barometric pressure in millibar, T the

temperature in Kelvin, the partial pressure of water vapor in
millibar ( or hPa) and n the refractive index.
A large number of measurements led to the establishment of an
average profile for r as a function of height h, called the
standard atmosphere.
Note: Waves in the visible spectrum are not affected by moisture: their refractive index is given
by the equation, letting =0. 1192
596
Atmospheric effects
The relative permittivity of the atmosphere is
close to unity.
It generally decreases (approaches unity as
altitude increases, since pressure and humidity
decrease with height faster than does temperature.
This change in permittivity with altitude causes
radio waves to bend toward the earth.
Such refraction of radio waves can sometimes be
useful, since it may extend the range of radar and
communication systems beyond the limit imposed
by the presence of the earths horizon.
Note: In a radar system, refraction effects can lead to errors when determining the elevation of
1193
a target close to the horizon.
Refractivity of the standard

troposphere
The refraction index of the troposphere
depends exponentially of the height:
N is the modified refraction index, refractivity

Ns = 315 in the refractivity at ground level under
normal atmospheric conditions (temperature,
atmospheric pressure, the partial pressure of water
vapor)
hs=7.35km
1194
597
Refractivity
At low altitudes up to h = 1 km the linearized model is valid
for the refractivity
Since the velocity of a radio wave is inversely proportional to

the refractive index (n) the wavefronts tend to bend towards
increasing direction of n.
Note: The refractivity function is spherical, but a planar model can be used as an approximation
1195
for low altitudes.
Standard atmosphere
1196
598
Radio wave propagation through

inhomogeneous medium
Consider a radio ray traveling in a planar medium whose refraction index
depends only height coordinate n = n(h) (a model for the troposphere)
Let us study the system of two homogeneous layers, having respectively
the refraction indices n, n+dn
dn

dh
d
Note: The equation for a bending radio ray can be derived assuming that the refraction index
1197
changes stepwise over layer boundaries.
Curvature radius of the radio ray

Snells law of refraction:
n (h) sin [ (h)] = constant
From this the radius of curvature of the ray R (h) can

be solved:
1198
599
Bending direction
Radio ray bends towards electrically denser

medium, i.e., towards the direction of higher
refraction index n, , i.e., where n is higher.
1199
Linear approximation for n(h)

Consider a linearly changing n(h) for which
constant, if n1 and /2
This means circular ray path in the troposphere for

small elevation angles!
Note: Here is assumed that radio ray propagates nearly as a direction of the Earths surface. So
we can assume n 1 and approximately zero elevation angle ( 0), which means /2. 1200
600
Alternative trajectories
1201
Effective radius of the earth

If a radio wave propagates in the troposphere nearly
parallel to the surface of the ground i.e. /2,
the radius of curvature of the path is then approximately
the bending path is almost circular in standard

tropospheric conditions which means

where the real Earth radius a = 6370 km.
1202
601
Affine transformation
The analysis of radio wave propagation is difficult
since both the wave and the surface are bending.
Using an affine transformation one of the paths
can be made straight.
After an affine transformation, only one curved line is
left.
This transformation must be done so that the
distance between the lines do not change.
So, the height of radio ray from the Earths
surface is the same as before the transformation.
1203
Affine transformation
Assume the path height h(d)<< d << a or R
Cosine law:
Omitting the small terms:
1204
602
Effective radius
The height of the radio path h(d),
remains the same if values a ae and R Re such that
1205
Effective radius
Now we can straighten the radio path by setting Re=

In standard tropospheric conditions R 4a corresponding the
effective radius of the Earth ae = Ka = (4/3)a = 8500km
This transformation replaces the refractive index of the
troposphere and its derivate to ne and dne/dh=0
How does the refractive index change in the affine
transformation?
1206
603
Example
If an antenna is at a height h, above the earth, simple
geometry gives the line-of- sight distance to the horizon
as
d 2ah
where a is the radius of the earth.
From the previous study we see that the effect of
refraction on range can be accounted for by using an
effective earth radius ae=Ka where k> 1.
A value commonly used is K = 4/3, but this is only an
average value.
Actually, bending depends on weather conditions.
1207
Geometric and radio horizon

If an antenna is at a height h, above the earth, simple geometry
gives the line-of- sight distance to the horizon.
If we assume idealized free-space type straight line propagation, we
get the distance d to the geometric horizon
d a h2 a 2 a 2 2 ah h 2 a 2
2ah h 2 2ah , when h a
In standard troposphere, of course, the refraction bends the

radio wave.
So, the distance dst to horizon in the standard troposphere is
4
d st 2 ae h 2 Kah d K d 1.15d
3
1208
604
Radio and optical horizon (smooth

earth) in the standard atmosphere
1209
Note
Prerequisites to exploit effective Earth radius
dn/dh = constant ( n(h)=constant*h+Cint)
propagation almost parallel to the surface (radio
wave path is circular)
For microwave links, the elevation angle is very small.
Operation area is small enough (radius < 100 km
from transmitter)
1210
605
Profile chart
Antenna heights and terrain heights representations are defined
with respect to the ground surface (sea level)
the design of radio mast height is difficult
The height function is a parabola!
1211
Profile chart
Knife edge diffraction causes attenuation (shadowing
effect)! Increase antenna height (h2) to avoid
shadowing.

1212
606
Profile chart usage

The middle point of the radio link
is set to the center.
The highest points from the
terrain profile between Rx and Tx
antennas should be marked to the
chart.
The radius of the Fresnel zone
should be calculated at the highest
points.
The line-of-sight path is drawn to
the chart such that the highest
points are below the Fresnel
ellipsoid.
The needed radio mast heights are
read from the chart.
1213
Note
Radio ray propagation in the troposphere is
relatively easy to analyze because its path is
practically circular besides in very exceptional
temperature conditions and in the case of too
steep elevation angle.
The spherical inhomogeneity of troposphere can be
approximated by a planar linear model for the
refractive index.
In the ionosphere the ray path must be analyzed
with more rigorous method using nonlinear
spherical permittivity model.
1214
607
Real atmosphere
The standard atmospheric profile is only a
statistical value, representing an average taken over a
large sample of experimental data.
In fact, the term K depends upon the latitude, varying
between 1.2 and 1.5.
When the atmosphere fluctuates, n may even locally
increase with altitude: waves are then trapped within a
channel (anomalous refraction, mirages).
Shadow regions are located next to such channels,
where it is apparently impossible to emit or to receive.
Such anomalies often take place over sea shore.
1215
Standard atmosphere
1216
608
Ducting in the troposphere

Weather conditions can produce a temperature
inversion, where the temperature increases with
altitude.
This condition can sometimes lead to ducting (also
called trapping, or anomalous propagation), where a
radio wave can propagate long distances parallel to the
earths surface, via the duct created by the layer of air
along the temperature inversion.
The situation is very similar to propagation in a
dielectric waveguide.
Such ducts can range in height from 15 150 m, and
may be near the earths surface, or higher in altitude.
1217
Tropospheric channel
Temperature inversion causes channels, or ducts, of cool air to
form between layers of warm air, which can cause radio waves
to travel far beyond the normal line-of-sight distances.
A tropospheric channel (duct) near Earths surface
1218
609
Radiowave propagation in
tropospheric channel
In suitable weather conditions (temperature,
pressure,...) in the troposphere the curvature radius of
a radio wave is smaller than that of the Earth.
Physically this means that a radio wave can propagate
long distances within the tropospheric channel.
In order to analyze the guided propagation in the
troposphere it is convenient to straighten the surface of
the ground which alters the profile of the refractive
index.
The effective radius of the Earth, ae, and that of the
radio wave, Re, are connected with their values in
reality:
1219
Modified refractivity
Because the radius of curvature of the radio wave can be
approximated using the derivative of the refractive index it
follows that
The surface of the ground becomes planar if ae.

This implies that modified refractive index (ne(h) = m(h))
corresponds the refractive index profile in the flat ground
geometry
Correspondingly, the modified refractivity is defined as
1220
610
Super refraction
dN/dn <-157 km-1 leads to super refraction
and the emergence of a tropospheric channel.
A ray launched at an angle 0 from the ground

can be traced by the planar ray equation up to the
turning point at the height H.
1221
Alternative trajectories
Trajectory 7 super refraction
1222
611
Guided wave by refractive index

profile
1223
Tropospheric ducting
The propagation in a tropospheric duct is similar to propagation in a
dielectric waveguide.
a small attenuation
a cut-off frequency ( a minimum propagating frequency)
depending on ducts height and variation of the refractive index
inside the duct.
The wavelength corresponding to the cut-off frequency is
approximately
where h is the height of the duct and n is the variation of n inside

the duct.
1224
612
Example
Calculate max min when h=30m and

n=4*10-6.
max 2.5hn 2 2.5 30 4 10 6

1
0.15 m f min 2 GHz
1225
Absorption in microwave frequencies
Attenuation is an atmospheric effect, caused

primarily by the absorption of microwave energy
by water vapor and molecular oxygen.
Maximum absorption occurs when the
frequency coincides with one of the molecular
resonances of water or oxygen, thus atmospheric
attenuation has distinct peaks at these frequencies.
At frequencies below 10 GHz the atmosphere has
very little effect on the strength of a signal.
1226
613
Absorption in microwave frequencies

At EHF, resonance peaks occur due to water vapor
resonances, while resonances of molecular oxygen
cause peaks at 60 and 120 GHz.
Thus there are minimum loss windows in the
millimeter wave band near 35, 94, and 135 GHz where
radar and communication systems can operate with
minimum loss.
Hydrometeors such as rain, snow, or fog will increase
the attenuation, especially at higher frequencies.
The effect of atmospheric attenuation can be included
in system design when using the Friis transmission
equation or the radar equation.
1227
Atmospheric absorption
1228
614
Remark
In some instances the system frequency may be chosen at a
point of maximum atmospheric attenuation.
Remote sensing of the atmosphere (temperature, water
vapor, rain rate) is often done with radiometers operating
near 20 or 55 GHz, to maximize the sensing of atmospheric
conditions.
Another interesting example is spacecraft-to-spacecraft
communication at 60 GHz.
This millimeter wave frequency has the advantages of a
large bandwidth and small antennas with high gains and,
since the atmosphere is very lossy at this frequency the
possibilities of interference, jamming, and eavesdropping
from earth are greatly reduced.
1229
Effects of weather
At frequencies above 10 GHz, the atmosphere produces
absorption and diffraction of the signal, caused by oxygen,
water vapor, rain, clouds and snow.
These phenomena are quite complex and measured data
present a wide spread of values.
The practical results are an additional attenuation of the
signal, a rotation of its polarization plane, as well as noise.
As communications bands become increasingly saturated, it
becomes imperative to also operate more and more at
frequencies above 10 GHz.
Statistical studies of meteorology are made to predict the
reliability of a transmission.
1230
615
Plasma
A plasma is a gas consisting of ionized particles.
The ionosphere consists of spherical layers of atmosphere
with particles which have been ionized by solar radiation,
and thus forms a plasma region.
Plasmas are also produced by lightning, meteor showers,
and nuclear explosions.
A very dense plasma is formed around the spacecraft as it
reenters the atmosphere, due to the high temperatures
produced by friction.
A plasma is characterized by the number of ions per unit
volume; depending on this density and the frequency, a
wave might be reflected, absorbed, or transmitted by the
plasma medium.
1231
Plasma effects
If the effect of the Earths magnetic field is neglected and a
fixed frequency propagation is studied, the simple plasmas
electrical properties can be modeled by an effective
permittivity which can be defined for a uniform plasma region
as
1232
616
Plasma effects
By studying the solution of Maxwells equations for
plane wave propagation in a plasma medium it can be
shown that wave propagation through a plasma is
only possible for > p.
Lower frequency waves will be totally reflected.
If a magnetic field is present, the plasma becomes
anisotropic, and the analysis is more complicated.
Note: The presence of a plasma prevents the propagation of signals at frequencies lower than
cut-off fp. A similar effect occurs with a spacecraft entering the atmosphere. The high velocity of
the spacecraft causes a very dense plasma to form around the vehicle. The electron density is
high enough so that the plasma frequency is very high, thus inhibiting all outside
communication with the spacecraft until its velocity has decreased.
Besides this blackout effect, the plasma layer may also cause a large impedance mismatch
between the antenna and its feed line. 1233
Ionosphere
The high layers of the atmosphere are constantly
bombarded by ionizing solar radiation, so that spherical
layers of ionized particles (plasma) surround the earth:
they are called the ionosphere .
While the ionosphere is by no means a simple plasma,
as it fluctuates and is placed within the earths magnetic
field.
The ionosphere is made up of layers D, E, F1 and F2.
The upper layer F2 varies considerably, in density and
height, depending further on seasons and solar cycles.
1234
617
Ionosphere
The cutoff frequency is located, in the average, around 8
MHz.
The presence of this sparse plasma prevents the propagation
of signals trough this plasma layer at frequencies lower than
fp.
Thus, signals at frequencies less than 8 MHz (e.g. short-
wave radio) can reflect off the ionosphere to travel distances
well beyond the horizon.
Higher frequency signals, however, will pass through the
ionosphere.
Microwave signals, at much higher frequencies, cross the
ionosphere without reflection and almost without
dispersion.
1235
Layer structure of the ionosphere
1236
618
Bending of a radio wave in the

ionosphere
The ionosphere is, in general, anisotropic, inhomogeneous
and dispersive medium.
If the effect of the Earths magnetic field is neglected and a
fixed frequency propagation is studied, the ionospheres
electrical properties can be modeled by
1237
Example
Calculate plasma frequency fp,max using the
maximum electron density Nmax1012 m-3 .
1238
619
Ionosphere
The electron density of the ionosphere depends on the altitude,
time of the day and the sunspot number.
1239
Electron densities and plasma

frequencies
Average electron density in the ionosphere and corresponding
plasma frequency versus height.
1240
620
Electron density N(h) vs. n(h,)

The effect of the electron density on the refractive index and the path of the
radio wave
Because n < 1 the radio wave bends always away from the ionosphere.
If <p refractive index is imaginary and the wave attenuates.
1241
A planarly stratified model for the

ionosphere
This approximation can used if the wave stays
in the ionosphere only a short range
1242
621
A planarly stratified model for the

ionosphere
Snells law can be applied for the
radio ray
At the turning point = /2

the reflection height H=n-1( )
1243
Secant law
Geometric interpretation: All waves which have the same

product cos0 are reflected at the same height H.
1244
622
Spherically stratified model for the

ionosphere
Snells law is valid in planarly inhomogeneous
medium.
Bouguers law replaces Snells law in the spherical
stratified inhomogeneity model for the ionosphere.
Ray paths in a spherically stratified (lossless) media can
be described by Bouguers law
which defines the path of a ray.
The secant law is also invalid without generalization.
1245
Radio paths
1246
623
Radio link via ionosphere

The distance of the transmitter and the receiver is
d.
Which frequency should be chosen?
The range of low frequencies is short and high frequencies
penetrate the ionosphere
What is the correct elevation angle ?
What is the height of the turning point?
If either n(h) or N(h) is known, the problem can
be solved iteratively using the ray equation, but
it can be solved more easily using an ionogram!
1247
Ionogram
An ionosonde is an instrument to obtain a view of the
state of the ionosphere.
It transmits vertically upwards (elevation angle /2) a
frequency-sweeped RF-pulse (1-25 MHz) to the
ionosphere and records the reflected signal as a
function of travel time and produces a plot called an
ionogram.
Higher frequencies penetrate deeper into ionosphere
than lower frequencies.
The two-way travel time as a function of frequency
provides information about the height of the reflection
and consequently about the electron density.
1248
624
Height of the turning point
The ionogram gives, however, the virtual

height of the turning point H() = c0T()/2
where T() is the travel time of the signal
spectrum part at .
1249
Ionospheric link design

A radio link via the ionosphere can be designed if the ionogram i.e.
H (p(H)) is known halfway between the transmitter and receiver.
In the case of a flat Earth model the secant law is valid for bending
rays
Note that p(H) is the plasma frequency at the true bending height
but H is the virtual reflection height.
H is not known but the halfway ionogram yields H.
The connection between H and p(H) can be solved from the above
equation
and then graphically connected to the ionogram.

1250
625
Maximum Usable Frequency (MUF)
Maximum Usable Frequency is the highest frequency

reflected by the ionosphere.
Frequencies lower than MUF have at least two reflection
heights.
Note: Anisotropy generates more solutions (o- and x-waves reflect at different heights). 1251
Maximum Usable Frequency (MUF)

The higher the frequency of a radio wave, the lower the rate of refraction by an
ionized layer.
Therefore, for a given angle of incidence and time of day, there is a maximum
frequency that can be used for communications between two given locations.
This frequency is known as the maximum usable frequency (MUF).
Waves at frequencies above the MUF are refracted so slowly that they return to
Earth beyond the desired location, or pass on through the ionosphere and are lost.
Note, however, that use of an established MUF certainly does not guarantee
successful communications between a transmitting site and a receiving site.
Variations in the ionosphere may occur at any time and consequently raise or lower
the predetermined MUF.
This is particularly true for radio waves being refracted by the highly variable F 2
layer.
The MUF is highest around noon when ultraviolet light waves from the sun are the
most intense.
It then drops rather sharply as recombination begins to take place.
1252
626
Lowest Usable Frequency (LUF)

As there is a maximum operating frequency that can be used
for communications between two points, there is also a
minimum operating frequency, which is known as the lowest
usable frequency (LUF).
As the frequency of a radio wave is lowered, the rate of
refraction increases.
So a wave whose frequency is below the established LUF is
refracted back to Earth at a shorter distance than desired.
Refraction
below the LUF.
1253
LUF
The transmission path that results from the rate of refraction is not
the only factor that determines the LUF.
As a frequency is lowered, absorption of the radio wave increases.
A wave whose frequency is too low is absorbed to such an extent
that it is too weak for reception.
Likewise, atmospheric noise is greater at lower frequencies: thus, a
low-frequency radio wave may have an unacceptable signal-to-
noise ratio.
So, for a given angle of incidence, the LUF for successful
communications between two locations depends on the refraction
properties of the ionosphere, absorption considerations, and the
amount of atmospheric noise present.
1254
627
Note
Optimum working frequency is the most practical
operating frequency and the one that can be relied
on to have the fewest problems.
E.g.
Precipitation attenuation can be caused by rain, fog,
snow, and hail, and can affect overall communications
considerably.
Temperature inversion causes channels, or ducts, of
cool air to form between layers of warm air, which can
cause radio waves to travel far beyond the normal line-
of-sight distances.
1255
Effects of ionozation layer density

on radio waves
1256
628
Ground wave coverage, skip distance

and sky wave coverage
1257
Huygens source
Instead of determining the radiation directly from an antenna,
equivalent source (Huygens source) can be applied.
Huygens source can be defined at any surface S surrounding the
original antenna
Huygens source consists electric and magnetic surface currents
which depend on the (true) fields at the surface
The benefit of Huygens source is that it can be used

approximatively.
1258
629
Approximation of Huygens source
Assume an isotropic point source at point A

and a receiver at point B.
The far-field amplitude of the electric field at a
plane between the transmitter and receiver can
be written as
2
j r p

e
E p
r p
1259
Huygens principle
Diffraction is based on Huygens principle of wave
propagation: Points on a wave front can be considered
sources for additional wavelets.
Huygens principle states
all the points (obstructions) on a wave-front become point
sources for production of secondary wavelets which combine
(vector sum) to form a new wave-front (in the direction of
propagation), and this new wave-front is called diffracted
wave-front
1260
630
Fresnel zone
According to Huygens principle the electric field at the
receiver can be written as
2 2 2
j r p j s p j r p s p

e e e
E p dS p dS p
r p s p r p s p
The value of the integral is collected mainly over the area near
the line-of-sight (LOS) path.
This area is called Fresnel zone.
1261
The first Fresnel zone

The definition for the first Fresnel zone:
2
r d 2 r1 r2 d1 d 2 r d
2
Using series expansion: 1 x n 1 nx nn 1 x 2 , nx 1

2!
((b<<d1, d2))
1262
631
The nth order Fresnel zone

The definition for the nth order Fresnel zone:
2
r d 2 r1 r2 d1 d 2 n r d n
2
1263
Fresnel ellipsoid
Paths from transmitter to receiver which have the
exactly same electric lengths describe the Fresnel
ellipsoid
The thickness of Fresnel ellipsoid:
Thickness of the Fresnel ellipsoid is the

geometric mean of wavelength and
distance.
The radius of nth Fresnel zone is
Note: r1+r2-(d1+d2)=r1+r2-d=/2 r1+r2=d+/2 1264
632
1st Fresnel zone

The main part of energy of a radio wave
propagates through the 1st Fresnel zone.
Hence, a free Fresnel ellipsoid indicates free space
conditions (environment factor F1).
Radio link design rule: Make free this Fresnel
ellipsoid.
Obstacles in the Fresnel ellipsoid cause attenuation
1265
Note
Another symbol for the radius of the 1st Fresnel zone: rF
The two antennas are located at the focal points of the ellipsoid
The link span is D=a+b.
The short half-axis is
1266
633
Knife-edge inside the Fresnel

ellipsoid
The distance between the Fresnel ellipsoid axel
and the knife-edge is h.
h<0 LOS or the line-of-sight path is free.
h>0 NLOS or the line-of-sight path is blocked
off by an obstacle (knife-edge).
An obstacle
1267
Knife-edge diffraction model

It is essential to estimate the signal attenuation caused
by diffraction of radio waves over hills and buildings
in predicting the field strength in the given service
area.
In practice, prediction for diffraction loss is a
process of theoretical approximation modified by
necessary empirical corrections.
The simplest case: shadowing is caused by a single
object such as a hill or mountain.
1268
634
Knife-edge diffraction gain

Fresnel-Kirchhoff diffraction parameter
The electric field strength Ed,
where E0 is the free space field strength

The diffraction gain:
G(dB)=20log(F) 1269
Knife-edge diffraction gain as a function

of Fresnel diffraction parameter v
G(dB)=20log(F)
1270
635
Attenuation caused by knife-edge

inside the 1st Fresnel ellipsoid
A shadowing obstacle inside the Fresnel ellipsoid cause a diffraction field, which
can be approximated by a type of Fresnel integral. The normalized field is
where
E0 is the corresponding free-space field

h is the distance between the Fresnel ellipsoid
axel and the knife-edge
Note: h<0 LOS or the line-of-sight path is free.

h>0 NLOS or the line-of-sight path is blocked off by an obstacle (knife-edge). 1271
Propagation losses
Obstacle loss also called diffraction loss or diffraction attenuation.
One method of calculation is based on knife edge approximation.
Having an obstacle free 60% of the Fresnel zone gives 0 dB loss
First Fresnel Zone
16dB 20dB
0 dB 0 dB 6dB
1272
636
Knife edge and earth surface losses
1273
Effect of the knife-edge diffraction
Attenuation (dB) caused by knife-edge obstacle inside the Fresnel ellipsoid
dB
h
H 2
rF 1274
637
Example
f = 1 GHz 2bmax = 55 m (=The thickness of

Fresnel ellipsoid)
Radio masts are needed
In reality the radio ray bends in the
troposphere which must be taken into account.
1275
Example
The link span (D) is 40 km. Calculate the maximum value
(rFmax) of the 1st Fresnel zone when a) f=250 MHz and b) f=4
MHz. ab
rF , ab D
ab
D
Obviously, rF rF max , when a b
2
a2 a D D
rF max
2a 2 2 2 2
D c f 1.2 40103
a) rF max 110 m
2 2
D 0.075 40103
b)rF max 27 m
2 2
Note: Radio masts are needed. 1276
638
Table: Thickness of Fresnel ellipsoid
Link span D=10 km The thickness of

Fresnel ellipsoid
1277
Multiple knife-edge diffraction

Attenuation due to multiple knife-edge diffraction
can be estimated by a recursive algorithm
(Myyrylainen (1960), Deygout (1966))
Note: This method can only use to calculate attenuations caused by distinguishable obstacles.1278
639
Multiple knife-edge diffraction

Procedure:
1. Select the most attenuating obstruction and forget the others.
2. Calculate attenuation coefficient F1 caused by this obstruction.
3. Select the top of the previous obstruction as an antenna and
calculate the same way attenuating coefficients F2 and F3 for these
two paths (Txobs.1 ,obs1.Rx) including only the worst
obstruction in each path section.
Iterating this way a sequence of attenuation coefficients Fi is
obtained.
Normally , it is enough to take into account two or three
obstructions, because additional coefficients are about 1.
1279
Note: If attenuation coefficients Fi are expressed as dB form then FdB=F1dB+ F2dB +.
Ground effects
The most obvious effect of the presence of the ground on microwave
propagation is reflection from the earth surface (land or sea).
A radar target or receiver antenna may be illuminated by both a
direct wave from the transmitter and an indirect wave, a wave
reflected from the ground.
The reflected wave is generally smaller in amplitude than the direct
wave, because of the larger distance it travels, the fact that it usually
radiates from the sidelobe region of the transmit antenna, and
because the ground is not a perfect reflector.
Nevertheless, the received signal at the target or receiver will be the
vector sum of the two wave components.
The two waves cover different distances: they have then different
phases when reaching the receiver.
Depending on the relative phases of the two waves, the sum may be
greater or less than the direct wave alone.
1280
640
Plane wave reflection from

boundaries
If the ground area inside the Fresnel zone illuminated by the
incoming wave is smooth and plain, the reflection can be
analyzed as originating from ideal flat surface
1281
Rough surface
Reflection from a rough surface is composed

by a coherent and a scattered part.
As the roughness of the surface increases the
coherent part diminishes
1282
641
Reflection from boundaries

Specular reflection corresponds to reflection from a smooth
surface.
Diffuse reflection is emitted to all directions
As the roughness of the surface increases, the specular
component decreases and scattered energy increases.
1283
Fading and scintillation

Line of sight (LOS) links also may present significant variations in
time, mostly due to interference.
The received signal is produced by the interference of the direct path
and the reflected path.
Depending upon the relative phases of the two signals, the field may
be enhanced or reduced; maxima and minima are, respectively,
with E0 amplitude of the direct field

|| modulus of the ground reflection factor.
The atmosphere is inhomogeneous and fluctuating in time, so that

the relative phases of the two signals and their amplitudes keep
changing.
1284
642

Because the distances involved are usually very large in terms
of the electrical wavelength, even a small variation in the
permittivity of the atmosphere can cause fading (long term
fluctuations) or scintillation (short term fluctuations) in the
signal strength.
These effects can also be caused by reflections from
inhomogeneities of the atmosphere.
The direct path may be divided into several different paths by
these atmospheric inhomogeneities.
In communication systems such fading can sometimes be reduced by making use of the fact that
the fading of two communication channels having different frequencies, polarizations, physical
locations , or in different times is essentially independent. If so, a communication link can
reduce fading by combining the outputs of two (or more) such channels; this is called a diversity
system. 1285
Diversity
Rayleigh fading and log-normal shadowing induce a very large
power penalty on the performance of modulation over
wireless channels.
One of the most powerful techniques to mitigate the effects of
fading is to use diversity-combining of independently fading
signal paths.
Diversity-combining uses the fact that independent signal
paths have a low probability of getting deep fades
simultaneously.
The idea behind diversity is to send the same data over
independent fading paths.
These independent paths are combined in some way such that
the fading of the resultant signal is reduced.
1286
643
Need of diversity
In a typical cellular radio environment, the communication
between the cell site and mobile is not by a direct radio path
but via many paths.
The direct path between the transmitter and the receiver is
obstructed by buildings and other objects.
Hence the signal that arrives at the receiver is either by
reflection from the flat sides of buildings or by diffraction
around man made or natural obstructions.
When various incoming radiowaves arrive at the receiver
antenna, they combine constructively or destructively, which
leads to a rapid variation in signal strength.
The signal fluctuations are known as multipath fading.
1287
Probability that R-distributed field is less than

abscissa value using 1-10 independent fading channels
1288
644

The most common source of fading remains,
however, the interference between the direct
path and the reflections.
It can be avoided to a large extent by keeping
the first Fresnel zone free from any obstacles.
The effect of changes in path length increases
with frequency.
For this reason, fading is more frequent and
pronounced at high frequencies.
1289
Ground diffraction
In the same way as it does in optics, a sharp edge, for
instance a mountain ridge in radio waves, produces
diffraction field behind the ridge.
Signals transmitted in this manner are strongly
attenuated, but may nevertheless be used, as they are
steady and independent from the atmospheric
conditions.
A transmission using diffraction over a ridge requires a
large emitted power.
Sometimes, an obstacle such as a mountain can be
taken advantage of when setting up a long range link.
1290
645
Diffraction over the horizon

Propagation over the horizon takes place when the straight line
from emitter to receiver intersects the terrestrial globe.
The curvature of the radio rays produced by the atmospheric
inhomogeneity leads to the definition of a radio horizon
different from the geometric one (effect of atmospheric
refraction).
Diffraction of the signal by the terrestrial globe can take place,
and thus the signal is transmitted beyond the radio horizon.
A transmission can thus take place at a distance beyond this
radio horizon, but the signal decreases quite fast within the
diffraction zone.
1291
Diffraction zone
1292
Note: As soon as the radio horizon line is crossed, the signal decreases quite rapidly.
646
Pros and cons

Ground diffraction offers possibilities to microwave
communication links beyond the horizon.
Of course, when obstacles such as hills, mountains, or
buildings are in the path of propagation, they attenuate
radio signal and thus, hinder communications.
In a radar system, unwanted reflections often occur
from terrain, vegetation, trees, buildings and the surface
of the sea.
Such clutter echoes generally degrade or mask the
return of a true target, or show up as a false target, in
the context of a surveillance or tracking radar.
In mapping or remote sensing applications such clutter
returns may actually constitute the desired signal.
1293
Example
1294
647
Scattering propagation
Radio wave scattering is a common phenomenon in the
atmosphere.
It is caused by irregularities
hydrometeors, i.e., particles consisting of liquid or solid
water suspended in or falling through the atmosphere: rain,
snow, ice, hail, fog etc.
turbulences and thermals
meteor traces
Scattering attenuates radio signal and thus, hinders
communications.
Scattering offers possibilities to high-frequency
communication links beyond the horizon.
1295
Basic concepts of scattering

A single-particle scattering from a small obstacle
The intensity of this Rayleigh scattering is inversely proportional
to the fourth power of wavelength.
The blue color of the sky is due to Rayleigh scattering of light
from small molecules.
Single scattering from large objects
A difficult numerical task for general shapes
In spherical case the Mie theory is applicable, e.g., light
scattering from a rain drop
Multiple scattering
Computationally very difficult in a general case
The radiative transfer theory is a commonly used approximation.
1296
648
Radio link via a turbulent troposphere
1297
Scattering height vs. propagation

distance
1298
649
Environment factor
1299
Scattering via ionosphere

High frequency region: f = 30-70 MHz > MUF
if f<30 MHz too high losses due to absorption in the
ionosphere
if f>70 MHz scattering diminishes
Connections are obtained via the E-layer (h100 km)
Achievable distances: d = 1000-2 000 km
Directive antennas (yagi)
Transmission power 10-100 kW
Benefits compared to the ionosphere reflection:
changes of the ionosphere do not affect much
at polar regions the ionosphere reflections are very unstable
1300
650
Meteor trace scattering

The number of meteors in the atmosphere is about
109 in a day.
The height of the trace is 80-120 km (E-layer) and
its length is 10-20 km.
Life time of the trace is 0.1-100 sec.
20 dB lower attenuation compared to ionoscatter
(lower power levels).
The same VHF band as for the ionoscatter.
Usage: meteorological stations, positioning
applications.
1301
Trans-horizon propagation
As soon as the radio horizon
line is crossed, the signal
decreases quite rapidly.
However, when the receiver
keeps moving further away
from the transmitter, the
decrease becomes slower
after a certain distance, not
as would be expected from
diffraction theory .
The field is then small, but
remains usable with large
gain antennas and a
powerful transmitter.
1302
651
Troposcatter
This propagation can be explained by the
diffusion of the wave in a part of the troposphere,
in direct line of sight of both transmitter and
receiver.
This diffusion appears to be produced by
atmospheric turbulence, which gives rise to
fluctuating local inhomogeneities.
The received field is the result of many small
contributions, produced by diffusion over a very
wide region of the atmosphere.
1303
Tropospheric scatter
1304
652
Tropospheric scatter
1305
Microwave links
Two questions must be answered when installing a line
of sight microwave link:
1. Choose a path undisturbed by obstacles, taking into
account the atmospheric diffraction and the ground
effect.
The topographic profile is drawn on a height-position chart.
On the same diagram, the first Fresnel zone is drawn, and
the locations and heights of the antennas can then be
determined.
2. The Friis equation is used to select the most suitable
components for the distance considered.
1306
653
Microwave links
A microwave link has telephone and television
channels.
It is easier to install than a cable system, particularly,
over rough terrain: antennas can be brought into
position on mountain tops by helicopter, the repeater
stations are powered by batteries, which may even be
charged with solar cells.
When installing a microwave link, care must be taken
to avoid interference between consecutive sections.
This is done by assigning different carrier frequencies
for adjacent sections.
1307
Example
Utilization of a height-position chart to set up a microwave
link above an imaginary Earth of radius KR with K= 4/3.
1308
654
Communications satellites
The utilization of stationary satellites for communications
was first advocated in 1945 by A.C. Clarke.
The first satellite, Sputnik I, was placed in orbit on October
4 1957.
In 1962 the first communications satellite, Telstar was
launched; it was a fly-past satellite, on a low-earth orbit.
The first stationary satellite, Early Bird, dates to 1965 and
was followed by several generations of Intelsat satellites.
At a height of 35800 kilometers, a satellite orbits the earth
in 24 hours: it remains steady over a fixed point on the
equator.
This is what is called the stationary orbit.
1309
Communication satellites
Geostationary Satellites 1965

Medium-Earth Orbit Satellites GPS
24 GPS satellites orbiting at 18,000 km
every 6 hours.
Low-Earth Orbit Satellites (Communications)
1310
655
Communication satellites
P
24 h
6h
1.5 h
Latency = round-trip delay time

Number of satellites needed for global coverage.
1311
ITU allocates the orbits as well

as satellite bands
1312
656
Geostationary satellite orbit
1313
Benefits of the stationary orbit

The use of a stationary orbit presents very significant
advantages for communications:
1. The antenna of the ground station points towards a
fixed location in the sky, only small corrections being
required to compensate for atmospheric diffraction and
orbital variations.
2. As the distance to the satellite remains constant, there
is no Doppler shift.
As a result, the frequency does not change.
For a fly-past satellite, on the contrary, the radial motion of
the satellite with respect to the antenna produces a
frequency shift, which must be taken into consideration.
1314
657
Benefits of the stationary orbit
3. As the satellite is located very far out, it is

far above the upper part of the atmosphere, and
therefore practically not slowed down by it.
It could remain in orbit for more than a million
years.
Its operation is, however, limited in time by the
fuel required for stabilization.
1315
Drawbacks of the stationary orbit

On the other hand, the stationary orbit also presents
drawbacks:
1. The time needed for a one-way transmission from the
ground to the satellite is greater or equal to 35,800
km/c0 = 119 ms.
A back-and-forth telephone transmission (four ways) thus
takes more than 476 ms.
This is a rather long delay, which might become awkward.
2. The large distance reduces considerably the signal
level.
The receiving station must utilize a large antenna, and a
high sensitivity receiver.
1316
658
Positions of the geostationary orbit

Positions of the geostationary orbit above the
equator taken up by communications satellites.
1317
Low-Earth Orbit satellites

Motorola Iridium launched 1997
Moving cells
covered by
satellites
The Iridium: six necklaces by 11 = 66 satellites (750 km).

Satellite phones didnt have much success in competition with terrestrial
cellular systems.
1318
659
Globalstar
(a) Relaying in space (Iridium).

(b) Relaying on the ground (Globalstar 48 LEOs).
1319
Space communications
At much further distances, communications with space
expeditions and probes are always carried by microwaves.
The distance from earth to the moon being 384000
kilometers, the signal transmission to the moon suffers a
1.28 second delay.
Images from Jupiter and its moons were transmitted by the
space probes Voyager 1 and 2.
The average distance from Jupiter to the sun is 778 millions
of kilometers, or 43 light-minutes.
The radius of our solar system is approximately 0.001 light-
year.
Some space probes are just leaving the solar system.
The signal received from such a probe is exceedingly small.
1320
660
Introduction
Microwave communication
A microwave communication system utilizes the radio
frequencies >1 GHz
These radio frequencies are called microwaves
Typically, spanning in microwave communication is from
2 to 60 GHz.
According to IEEE, electromagnetic waves between 30
and 300 GHz are called millimeter waves or Extremely
High Frequencies (EHF) as their wavelengths are from 10
to 1 mm.
Frequencies between 300 and 3000 GHz are called Hyper
High Frequencies (HHF) or submillimeter waves
1322
661
Microwave communication
Small capacity systems generally employ the

frequencies less than 3 GHz while medium and
large capacity systems utilize frequencies
ranging from 3 to 15 GHz.
Frequencies > 15 GHz are essentially used for
short-haul transmission.
1323
Why radio links
No requirement for right-of-way

Less vulnerable to vandalism
Not susceptible to accidental cutting of the
link
Often more suited to crossing rough terrain
Often more practical in heavily urbanized
areas
As a backup to fiber-optic cable links
1324
662
Advantages of microwave radio
Less prone to accidental damage

Links across mountains and rivers are more
economically feasible
Two point installation and maintenance
Two point security
They are quickly deployed
1325
Competitors
Fiber-optic cable systems provide

strong competition with line-of-sight
(LOS) microwave.
Satellite communications is an extension
of line-of-sight microwave.
Drawback; The excessive delay when the
popular geostationary satellite systems are
utilized.
1326
663
Line-of-sight microwave system
Line-of-sight (LOS) microwave provides broadband

bearer connectivity over a link or series of links
upwards into the millimeter spectrum.
Each link can be up to 40 - 50 km long or more
depending on terrain topology and antenna mast
height.
The key term here is line-of-sight (LOS).
It implies that the antenna of the radio link on one
end has to be able to see the antenna on the other
end.
This may not necessarily be true, but it does give
some idea of the problem.
1327
Radio horizon
The distance to the radio horizon varies with

the index of refraction of the intervening
space.
Under normal atmospheric conditions (k=4/3),
the radio horizon is about 15 percent beyond
the optical horizon because the microwave ray
beam being bent toward the earth.
However, this generalization may be overly
optimistic under certain circumstances.
1328
664
Radio and optical horizon (smooth earth)
1329
Refraction
1330
665
Line-of-sight considerations
Microwave radio communication requires a clear line-of-sight (LOS)
condition
Radio LOS takes into account the concept of Fresnel ellipsoids and their
clearance criteria
Fresnel zones are specified employing ordinal numbers that correspond to
the number of half wavelength multiples that represent the difference in
radio wave propagation path from the direct path
Ideally, the first Fresnel zone must be clear of all obstructions.
a+b+
Radius of the 1.Fresnel zone

rF rF=[*a*b/(a+b)]
a b
1331
Example
The link length is 40 a+b+
km. Calculate the

rF
maximum value of the
a b
first Fresnel radius,
when
a) f=250 MHz
b) f=4 GHz
Note.
1332
666
Radius of the first Fresnel zone
R=17.32(x(d-x)/fd)1/2
where d = distance between antennas (in km)
R= first Fresnel zone radius in meters
f= frequency in GHz
R y
x d=x+y
1333
Typically the first Fresnel zone is used to determine
obstruction loss
The direct path between the transmitter and the receiver
needs a clearance above ground of at least 60% of the
radius of the first Fresnel zone to achieve nearly free space
propagation conditions
Clearance is described to ensure sufficient antenna heights
so that, in the worst case of refraction (for which Earth-
radius factor k is minimum) the receiver antenna is not
placed in the diffraction region
1334
667
Clearance criteria to be satisfied under normal

propagation conditions:
Clearance of 60% or greater at the minimum k
suggested for the certain path
Clearance of 100% or greater at k=4/3
In case of space diversity, the antenna can have
a 60% clearance at k=4/3
1335
Microwave link design is a systematic

process that includes
Loss/attenuation calculations
Fading and fade margins calculations
Frequency planning and interference
calculations
Quality and availability calculations
1336
668
Microwave link design process

The whole process is iterative and may go through many
redesign phases before the required quality and availability
are achieved
Interference
analysis Frequency
Planning
Propagation losses Rain

attenuation
Branching Fading
Link Budget
losses Predictions Diffraction-
refraction
Other Losses
Quality losses
and
Multipath
Availability
propagation
Calculations
1337
Hop calculations
Q&A stands here for Quality and Availability 1338
669
Loss / Attenuation calculations
The loss/attenuation calculations are composed of

three main contributions
Propagation losses
(Due to Earths atmosphere and terrain)
Branching losses
(comes from the hardware used to deliver the
transmitter/receiver output to/from the antenna)
1339
Loss / Attenuation calculations

Miscellaneous (other) losses
unpredictable and sporadic in character like fog, moving
objects crossing the path, poor equipment installation and
less than perfect antenna alignment etc
Detailed contribution of these losses is not
calculated but they are considered in the planning
process as an additional loss
1340
670
Propagation losses
Free-space loss - when the transmitter and receiver have a
clear, unobstructed line-of-sight
Lfsl=92.45+20log(f)+20log(d) [dB]
where f = frequency (GHz)

d = LOS range between antennas (km)
One estimation of vegetation attenuation (provision should

be taken for 5 years of vegetation growth)
L=0.2f 0.3R0.6(dB) , where
f=frequency (MHz) and
R=depth of vegetation in meters
1341
Diffraction
1342
671
Propagation losses
Obstacle loss also called diffraction loss or diffraction attenuation.
One method of calculation is based on knife edge approximation.
Having an obstacle free 60% of the Fresnel zone gives 0 dB loss
First Fresnel Zone
16dB 20dB
0 dB 0 dB 6dB
1343
Propagation losses
Gas absorption
Primarily due to the water vapor and oxygen in
the atmosphere.
The absorption peaks are located around 23 GHz
for water molecules and 50 to 70 GHz for oxygen
molecules.
The specific attenuation (dB/km) is strongly
dependent on frequency, temperature and the
absolute or relative humidity of the atmosphere.
1344
672
Gas attenuation vs. frequency
Total specific
gas attenuation 23GHz
1.0 (dB/km)
T=40oC
RH=80%
0.4 T=30o
RH=50%
Frequency (GHz)
0 25 50
1345
Gas absorption
1346
673
Propagation losses
Attenuation due to precipitation
Rain attenuation increases exponentially with rain intensity
The percentage of time for which a given rain intensity is attained or
exceeded is available for 15 different rain zones covering the entire
earths surface
The specific attenuation of rain is dependent on many parameters such
as the form and size of distribution of the raindrops, polarization, rain
intensity and frequency
Horizontal polarization gives more rain attenuation than vertical
polarization
Rain attenuation increases with frequency and becomes a major
contributor in the frequency bands above 10 GHz
The contribution due to rain attenuation is not included in the link
budget and is used only in the calculation of rain fading
1347
1348
674
Rain attenuation
Two types of attenuating mechanisms:
absorption and scattering caused by the rain drops.
Horizontally polarized waves are attenuated more

than vertically polarized waves.
1349
Rain attenuation
1350
675
Ground reflection
Reflection on the Earths surface may give rise to multipath
propagation
The direct ray at the receiver may interfered with by the ground-
reflected ray and the reflection loss can be significant
Since the refraction properties of the atmosphere are constantly
changing the reflection loss varies.
The loss due to reflection on the ground is dependent on the reflection
coefficient of the ground
The reflection coefficient is dependent on the frequency, grazing angle
(angle between the ray beam and the horizontal plane), polarization and
ground properties
1351
Ground reflection
The grazing angle of radio-relay paths is very small
usually less than 1
It is recommended to avoid ground reflection by shielding
the path against the indirect ray
The contribution resulting from reflection loss is not
automatically included in the link budget.
When reflection cannot be avoided, the fade margin may be
adjusted by including this contribution as additional loss
in the link budget
1352
676
Multipath
1353
Path analysis
(Link budget)
677
Introduction
The path analysis (or link budget) is carried out to
dimension the link.
This means here to establish operating parameters
such as transmitter power output, parabolic antenna
aperture (diameter), and receiver noise figure, among
others.
The type of modulation and modulation rate (number
of transitions per second) are also important
parameters.
Next table shows basic parameters in two columns.
The first we call normal and would be the most
economic; the second column is titled special,
giving improved performance parameters, but at an
increased price.
1355
LOS microwave basic equipment

parameters
1356
678
Space distribution of power
1357

Design objective:
Full clearance of the 1. Fresnel zone.
1358
679
The radiated power is distributed in a zone

surrounding the direct line-of-sight.
1359
Example
1360
680
Obstacle loss
1361
Knife edge and earth surface

losses
1362
681
True and equivalent earth radius
1363
Earth bulge calculation
1364
682
Free 1.Fresnel zone
The 1st Fresnel shall be free from obstacles when

k=4/3.
On paths over water surfaces or desert areas, it is
recommended to have the 1st Fresnel zone free from
obstacles when k=1.
1365
Link budget
The link budget is a calculation involving the gain and
loss factors associated with the antennas, transmitters,
transmission lines and propagation environment, to
determine the maximum distance at which a
transmitter and receiver can successfully operate
Receiver sensitivity threshold is the signal level at
which the radio runs for a given bit-error rate at a
specified bit rate
System gain depends on the design of the radio
1366
683
Link budget
The gains from the antenna at each end are added to the system
gain
larger antennas provide a higher gain
The free space loss of the radio signal is subtracted.
the longer the link the higher the loss
These calculations give the fade margin
In most cases since the same duplex radio setup is applied to
both stations the calculation of the received signal level is
independent of direction
1367
Approach
First, we calculate the free-space loss between the transmit
antenna and receive antenna.
This is a function of distance and frequency (i.e., the
microwave transmitter operational frequency).
We then calculate the EIRP (effective isotropically radiated
power) at the transmit antenna.
The EIRP is the sum of the transmitter power output, minus
transmission line losses plus the antenna gain, all in decibel
units.
When we add the EIRP to the free-space loss (in dB), the
result is the isotropic receive level (IRL).
When we add the receive antenna gain to the IRL and subtract
the receive transmission line losses, we get the receive signal
level (RSL).
This relationship of path gains and losses is shown in the
following figure.
1368
684
LOS microwave link gains and losses.

Transmitter output is 1 watt or 0 dBW.
1369
Path loss
If the isotropic radiator (an antenna that radiates uniformly in
all directions) is fed by a transmitted power Pt, it radiates
power density Pt/4d2 (W/m2) at a distance d, and if a radiator
has a gain Gt, the power flow is enhanced by the factor Gt.
So power density at a distance d is S0= GtPt/4d2 .
Finally, the power intercepted by an antenna of effective cross
section A (related to the gain by Gr = 4A/2) is
Pr=AS0=PtGtGr( /4d)2.
The term 1/(/4d)2 is known as the free-space loss and
represents the steady decrease of power flow (in W/m2) as the
wave propagates.
From this we can derive the commonly used formula of free-
space path loss, which reduces to
where L is the free-space attenuation between isotropic

antennas in dB, F is the frequency in GHz, and D is the path
distance in km.
1370
685
Example
Consider the problem from a different aspect.
It requires 22 dB to launch a wave to just 1 wavelength () distant from
an antenna.
Thus for an antenna emitting +10 dBW, we could expect the signal one
wavelength away to be 22 dB down, or -12 dBW.
Whenever we double the distance, we incur an additional 6 dB of loss.
Hence at 2 from the + 10-dBW radiator, we would find -18 dBW; at
4, -24 dBW; 8, -30 dBW; and so on.
Now suppose that we have an emitter where FGHz = 1 GHz. What is the path
loss at 1 km?
From rough calculations, the 6-dB relationship is worthwhile

and also gives insight in that if we have a 50-km path and
shorten or lengthen it by a km, our signal level will be affected
little.
1371
Calculation of EIRP
Effective isotropically
radiated power is calculated
by adding decibel units:
the transmitter power output (in
dBm or dBW),
the transmission line losses in
dB (a negative value because it
is a loss), and
the antenna gain in dBi.
dBi = decibels referenced to
an isotropic (antenna).
EIRP = Trans. output (dBW) - Trans. line loss (dB) + Ant. gain (dB) 1372
686
Example
If a microwave transmitter has a 1-watt (0-

dBW) power output, the waveguide loss is 3
dB, and the antenna gain is 34 dBi, what is the
EIRP in dBW?
EIRP = Trans. output (dBW) - Trans. line loss (dB) + Ant. gain (dB)
1373
Calculation of Isotropic Receive Level

(IRL)
The IRL is the RF power level reaching the receive
antenna.
It would be the power we would measure at the base
of an isotropic receive antenna.
The calculation is shown below
IRLdBW = EIRPdBW Path lossdB
below 10 GHz : Path loss = Free-space loss (FSL) 1374
687
Calculation of Receive Signal Level (RSL)
The receive signal level (RSL) is the power

level entering the first active stage of the
receiver:
Note: Power levels can be in dBm as well.
1375
Example
Suppose the isotropic receive level (IRL) is -

121 dBW, the receive antenna gain is 31 dB,
and the line losses are 5.6 dB. What is the
RSL?
1376
688
Link budget
Receive Signal Level (RSL)
RSL = Pout Lctx + Gatx Lcrx + Gatx FSL
Pout = output power of the transmitter (dBm)

Lctx, Lcrx = Loss (cable,connectors, branching unit) between
transmitter/receiver and antenna(dB)
Gatx = gain of transmitter/receiver antenna (dBi)
FSL = free space loss (dB)
Link feasibility formula

RSL Rx (receiver sensitivity threshold)
1377
Calculation of receiver noise level

The thermal noise level of a receiver is a
function of the receiver noise figure and its
bandwidth.
For analog radio systems, receiver thermal
noise level is calculated using the bandwidth
of the intermediate frequency (IF).
For digital systems, the noise level of interest
is in only 1 Hz of bandwidth using the notation
N0, the noise level in a 1-Hz bandwidth.
The noise that a device self-generates is given
by its noise figure (dB) or a noise temperature
value.
1378
689
Calculation of receiver noise level

Any device above absolute zero generates thermal noise.
We know the thermal noise power level in a 1-Hz bandwidth of a perfect
receiver operating at 1 K is
where Pn is the noise power level.

From numerical value we will recognize this as Boltzmanns constant
expressed in dBW.
We can calculate the thermal noise level of a perfect receiver operating at
room temperature (T=290 K) using the following formula:
Pn=kTB, k=1.3806504 1023 J/K 1379
Noise Figure (NF)
Noise figure simply tells us how much noise has been

added to a signal while passing through a device in
question.
Noise figure (dB) is the difference in signal-to-noise
ratio between the input to the device and the output of
that same device.
We can convert noise figure to noise temperature in
kelvins with the following formula:
where Te is the effective noise temperature of a

device.
1380
690
Example
Suppose the noise figure of a device is 3 dB.
What is the noise temperature?
Lets round 1.995 to 2; thus
1381
Thermal noise power
The thermal noise power level of a device

operating at room temperature is
where BW is the bandwidth of the device in

Hz.
1382
691
Example
A microwave receiver has a noise figure of 8 dB and its
bandwidth is 10 MHz. What is the thermal noise level
(sometimes called the thermal noise threshold)?
If the receiver in the above example was operating in a digital

regime, wed want to calculate N0 (noise spectral density)
1383
Calculation of Eb/N0 in digital radio

systems
Instead of familiar signal-to-noise ratio (S/N, SNR) in digital
systems we often use Eb/N0, meaning energy per bit per noise
spectral density ratio.
We can relate Eb/N0 to bit error rate (BER) given the
modulation type in question.
Suppose the RSL was 1 watt and we were receiving 1000 bits
per second.
How much energy is imparted to 1 bit? We simply divided 1
watt by 1000 bits per second. So it is 1 mWs.
In radio work we do the division logarithmically because we
work with decibels.
Eb can be stated as follows:
Note: The receive signal level (RSL) 1384
692
Example
A certain radio system receives 1.544 Mbps

and the RSL is -108 dBW. What is the energy
per bit (Eb)?
1385
Eb/N0
We can now develop a formula for Eb/N0:
Simplifying, we obtain
Note.
1386
693
Some notes on Eb/N0

Eb/N0, for a given BER, will be different for
different types of modulation (e.g., FSK, PSK,
QAM, etc.).
When working with Eb, we divide RSL by the
bit rate, not the symbol rate nor the baud rate.
There is a theoretical Eb/N0 and a practical
Eb/N0.
The practical Eb/N0 is always a greater value
than the theoretical, greater by the modulation
implementation loss in dB.
1387
Digital modulation of LOS microwave

radios
Digital systems, typically standard PCM, are
notoriously wasteful of bandwidth compared to their
analog counterparts.
For example, the analog voice channel is nominally 4
kHz, whereas the digital voice channel, assuming one
bit per hertz of bandwidth is 64 kHz.
This is a 16-to-1 difference in required bandwidth.
Thus various national regulatory authorities require
that digital systems must be bandwidth conservative.
One term that is used in this connection is bit
packing.
This means packing more bits into a hertz of
bandwidth.
1388
694
How does bit packing work?

In the binary domain we can estimate bandwidth to
approximately equate to 1 bit/Hz.
For example, if we were transmitting at 2.048 Mbps wed need 2.048
MHz of bandwidth.
Quadrature phase shift (QPSK) keying is one example of
higher levels of modulation.
In this case we achieve a theoretical packing of 2 bits/Hz.
Again, if we are transmitting 2.048 Mbps with QPSK we would need
2.048 MHz/2 or 1.024 MHz.
QPSK is one of a family of modulation schemes that are based
on phase-shift keying (PSK).
1389
Block diagram of a QPSK

modulator
With binary PSK (BPSK) we might assign a binary 1 to the 0 position
(i.e.. no phase retardation) and a binary 0 to the 180 phase retardation
point.
For QPSK, the phase-circle is broken up into 90 segments.
In this case, for every transition we transmit 2 bits at a time.
QPSK modulator really only consists of two BPSK modulators where one
is out of phase with the other by 90.
1390
695
8-PSK
In 8-ary PSK modulation case the phase circle

is broken up into 45 phase segments.
Now for every transition, 3 bits at a time are
transmitted.
The bit packing in this case is 3 bits per Hz.
1391
QAM
We can construct a hybrid waveform covering both amplitude
modulation as well as phase modulation.
This family of waveforms is called quadrature amplitude
modulation (QAM).
For example, 16-QAM has 16 different state possibilities: eight are
derived for 8-ary PSK and two are derived from the two amplitude
levels.
Here for each state transition, 4 bits are transmitted at once.
The bit packing in this case is 4 bits/Hz.
Some digital LOS microwave systems use 256-QAM and 512-
QAM, theoretically achieving 8 bits/Hz and 9 bits/Hz of bit
packing.
1392
696
Example
Draw the constellation diagram for the

following:
a. ASK, with peak amplitude values of 1 and 3
b. BPSK, with a peak amplitude value of 2
c. QPSK, with a peak amplitude value of 3
d. 8-QAM with two different peak amplitude
values, 1 and 3, and four different phases.
1393
Solution
a. We have two signal elements with peak amplitudes 1 and 3. The phase of both
signal elements are the same, which we assume to be 0 degrees.
b. We have two signal elements with the same peak amplitude of 2. However, there
must be 180 degrees difference between the two phases. We assume one phase to be
0 and the other 180 degrees.
c. We have four signal elements with the same peak amplitude of 3. However, there
must be 90 degrees difference between each phase. We assume the first phase to be
at 45, the second at 135, the third at 225, and the fourth at 315 degrees. Note that
this is one out of many configurations. The phases can be at 0, 90, 180, and 270. As
long as the differences are 90 degrees, the solution is satisfactory.
d. We have four phases, which we select to be the same as the previous case. For
each phase, however, we have two amplitudes, 1 and 3 as shown in the figure. Note
that this is one out of many configurations. The phases can be at 0, 90, 180, and
270. As long as the differences are 90 degrees, the solution is satisfactory.
1394
697
Solution
1395
Theoretical/Practical bandwidth
The difference between theoretical bit packing and
the practical deals with filter design.
For QAM-type waveforms, depending on design,
practical bit packing may vary from 1.25 to 1.5 the
baud-rate-bandwidth.
The extra bandwidth required provides a filter with
spectral space to roll-off.
In other words, a filters skirts are not perfectly
vertical.
1396
698
Example
Suppose we are using a 48-Mbps bit stream to input to our
transmitter which was using 16-QAM modulation.
Its baud rate, which measures transitions per second, would be
48/4 megabauds.
If we allowed 1 baud/Hz, then 12 MHz bandwidth would be
required.
If we used a roll-off factor of 1.5, then the practical bandwidth
required would be 18 MHz.
Carry this two steps further to 64-QAM.
Here the theoretical bit packing is 6 bits/Hz and for the 48-
Mbps bit stream, a practical 12-MHz bandwidth would be
required.
baud (unit symbol "Bd") is synonymous to symbols per second. It is the unit of
symbol rate, also known as baud rate or modulation rate; the number of
distinct symbol changes (signaling events) made to the transmission medium per
second in a digitally modulated signal or a line code 1397
BER vs. Eb/N0 for various PSK and QAM
As an amount of signal elements (symbols) M=2n increases (e.g., M = 64), for a

given error rate, Eb/N0 increases 1398
699
Example
A digital link operates in the 7-GHz
band with a link 37 km long.
The bit rate is 1.544 Mbps and the
modulation is 64-QAM.
The specified BER for the link is 10-7
and the modulation implementation
loss is 2 dB.
The receiver noise figure is 8 dB.
The antennas have 35-dB gain at each
end, and transmission line losses are
1.8 dB at each end.
What link margin can be expected
assuming the transmitter has a 1-watt
output?
Solution
First turn to next figure and derive the
required Eb/N0.
This is about 19.5 dB; add to this the
modulation implementation loss of 2
dB and the result is that the required
value for Eb/N0 is 21.5 dB.
1399
Example (cont.)
The next step is to calculate a candidate RSL
(Receive Signal Level) value.
We know that Eb must be 21.5 dB above N0.
We can calculate N0 because we have the receiver
noise figure (8 dB)
1400
700
Example (cont.)
Thus RSL, in this case, is 10 log(1.544106) greater

than Eb, because
So
Example (cont.)
We will hold this minimum RSL value for future

reference (we use this reference to calculate the
prevailing link margin), and now we turn to the
transmit side of the link.
Assume the transmitter has a 1-watt output or 0 dBW.
Calculate EIRP (effective isotropically radiated power)
in dBW
EIRP = Trans. output (dBW) - Trans. line loss (dB) + Ant. gain (dB)
1402
701
Example (cont.)
Calculate the free-space loss (path loss):
Calculate the IRL (isotropic receive level) :

IRLdBW = EIRPdBW Path lossdB
1403
Example (cont.)
Calculate RSL (Receive Signal Level) :
Calculate the link margin: RSL-RSLreference
1404
702
Antenna gain
Often we are faced with the problem of What

antenna gain will provide the margin or
provide the gain necessary to meet
performance objectives?
1405
Parabolic antenna gain

At a given frequency the gain of a parabolic antenna is a
function of its effective area and may be expressed by the
formula
where G is the gain relative to an isotropic antenna, A is the

area of antenna aperture, is the aperture efficiency, and is
the wavelength at the operating frequency.
Commercially available parabolic antennas with a
conventional horn feed at their focus usually display a 55%
efficiency or somewhat better. With such an efficiency, gain
(G, in decibels) is then
where D is measured in meters and F in gigahertz.

1406
703
Calculation of antenna gain (=0.55)
c

f
FGHz 109
2

G 10 log 4 D / 2 0.55
2

3 108

0.55 100
2
G 20 log D 20 log FGHz 10 log
9
G 20 log D 20 log FGHz 17.8
1407
Example
What size antenna would be required in the preceding

example? Let G = 35dB and F=7GHz.
D1m
1408
704
Antenna types
Parabolic dish antennas, with waveguide
(horn) feeds, are probably the most economic
antennas for radio links operating from 3 GHz
upward.
Below 3 GHz, coaxial feeds are used, and
often the antennas are Yagis.
Coaxial cable transmission lines deliver the
RF energy from/to transmitter/receiver to the
antenna in this range.
Above 3 GHz, coaxial cable becomes too
lossy and waveguide is more practical.
Other types of antennas may also be used,
such as the horn, and spiral.
Besides cost and gain, other features are
front-to-back ratio, side lobes, and efficiency.
1409
Power budget
LFSL 92.4 20 log FGHz 20 log Dkm

Prx Ptx 2Gant 2 L feed LFSL

Margin Prx Pthreshold
LFSL: free space loss

Dkm: distance
FGHz: frequency
Prx: received power [dBm]
Ptx : transmitted power [ dBm]
Pthreshold: receiver threshold [dBm]
M : margin
Gant: antenna gain
Lfeed: feeder cable loss 1410
705
Link budget
The fade margin is calculated with respect to the
receiver threshold level for a given bit-error rate
(BER).
The link is available if received signal is above
receiver threshold level, otherwise the link goes down
Link feasibility formula
RSL Rx = receiver (sensitivity) threshold
Radio path link budget

waveguide
Transmitter 1 Transmitter 2
Splitter Splitter
Receiver 1 Receiver 2
Antenna Gain
Propagation
Output Branching
Losses
Power (Tx) Losses Branching

Losses
Received
Power (Rx)
Fade Margin
Receiver threshold Value
1412
706
Fading
On very short radio paths below about 10 GHz, the signal level
on the distant end receiving antenna, assuming full LOS
conditions, can be predicted (calculated) accurately (about
1dB inaccuracy).
If the transmitter continues to give the same output, the receive
signal level will remain constant.
As the path is extended, the measured RSL will vary around a
median.
The signal level may remain at that median for minutes or
hours, and then suddenly drop and then return to the median
again.
In other periods and/or on other links, this level variation can
be continuous for periods of time.
Drops in level can be as much as 30 dB or more.
This phenomenon is called fading.
The system and link design must take fading into account
when sizing or dimensioning the system/link.
1413
Fades
As the RSL drops in level, so does the Eb/N0.
As the Eb/N0 decreases, there is a deterioration in
error performance; the BER degrades.
Fades vary in depth, duration, and frequency (i.e.,
number of fade events per unit of time).
We cannot eliminate the fades, but we can mitigate
their effects.
The primary tool we have is to overbuild each link
by increasing the margin.
1414
707
Availability
Link margin is the number of dB we have as a surplus in the link design.
We could design an LOS microwave link so we just achieve the RSL at the
distant receiver to satisfy the Eb/N0 (and BER) requirements using free-
space loss as the only factor in link attenuation (besides transmission line
loss).
Unfortunately we will only meet our specified requirements about 50% of
the time.
So we must add margin to compensate for the fading.
We have to determine what percentage of the time the link meets BER
performance requirements.
We call this time availability.
If a link meets its performance requirements 99% of the time, then it does
not meet performance requirements 1 % of the time.
We call this latter factor unavailability.
1415
Scales of fading
Signal Power (dBm)
Large scale fading component
Small scale fading

component
1416
708
Link margin = Fade margin

To improve time availability, we must increase the link margin, often called
the fade margin.
How many additional dB are necessary?
There are several approaches to the calculation of a required fade
margin.
One of the simplest and most straightforward approaches is to assume that
the fading follows a Rayleigh distribution, often considered worst-case
fading.
If we base our premise on a Rayleigh distribution, then the following fade
margins can be used:
1417
Probability calculations
1418
709
Probability calculations
1419
Example
1420
710
Example
Calculate the following table a) with given time availability and b) with
given fade margin
1421
Time availability
1422
711
Probability that R-distributed field is less

than abscissa value
1423
Per-hop time availability

More often than not, LOS microwave systems consist of multiple hops.
Here our primary interest is the time availability at the far-end receiver in
the system after the signal has progressed across all of the hops.
From this time availability value we will want to assign an availability
value for each hop or link.
Suppose a system has nine hops and the system time availability specified
is 99.95%, and we want to calculate the time availability per hop or link.
The first step is to calculate the system time unavailability.
This is simply 1.0000 0.9995 = 0.0005.
We now divide this value by 9 (i.e., there are nine hops or links):
Now we convert this value to time availability:
1424
712
Multipath
The most common cause of
fading is multipath conditions.
As the term implies, signal energy
follows multiple paths from the
transmit antenna to the receive
antenna.
Two additional paths, besides the
main ray beam, are shown next
figure.
Most of the time the delayed
signal energy (from the
reflected/refracted paths) will be
out of phase with the principal
ray beam which causes fading.
1425
Multipath
1426
713
Multipath propagation:
Reflection (odd Fresnel zone numbers) and
Scattering
Reflected signal strength is dependent on:

The total reflection coefficient (reflection and/or scattering)
1427
The total phase shift 1427
Multipath Propagation:
Reflection (even Fresnel zone numbers) and
Scattering
Reflected signal strength is dependent on:

The total reflection coefficient (reflection and/or scattering)
1428
The total phase shift 1428
714
Dispersion
In digital systems, there is the additional

impairment of dispersion caused by multipath.
Of course, the delayed energy arrives later,
spilling into the next bit or binary symbol
position, increasing the probability that that
bit decision will be in error.
1429
Overbuilding
Probably the most economic way to overbuild a link is to increase the
antenna aperture.
Every time we increase the aperture doubling the diameter of the parabolic
dish, we increase the gain by 6 dB
It is recommend that apertures for LOS microwave antennas not exceed 4

m.
Not only does the cost of the antenna get notably greater as aperture
increases over 3 m, but the equivalent sail area of the dish starts to have an
impact on system design.
Wind pressure on large dishes increases tower twist and sway, resulting in
movement out of the capture area of the ray beam at the receive
antenna.
This forces to stiffen the tower, which could dramatically increase system
cost.
Also, as antenna aperture increases, gain increases and beamwidth
decreases.
1430
715
Overbuilding
Other measures we can take to overbuild a link are:
Insert a low-noise amplifier (LNA) in front of the receiver - mixer.
Improvement: 6 - 12 dB.
Use an HPA (high-power amplifier).
Usually a traveling-wave tube (TWT) amplifier; 10 watts output.
Improvement: 10 dB.
Implement FEC (forward error correction).
Improvement: 15 dB.
Implement some form of diversity.
Space diversity is preferable in many countries.
Can be a fairly expensive measure.
Improvement: 5 - 20 dB or more.
1431
Fading conditions
It should be appreciated that fading varies with path length,
frequency, climate, and terrain.
The rougher the terrain, the more reflections are broken up.
Flat terrain, and especially paths over water, tends to increase
the incidence of fading.
For example, in dry, windy, mountainous areas the multipath
fading phenomenon may be nonexistent.
In hot, humid coastal regions very strong fading may be
expected.
1432
716
Fading types
Fading is defined as the variation of the strength
of a received radio carrier signal due to
atmospheric changes and/or ground and water
reflections in the propagation path.
Four fading types are considered while planning
links.
They are all dependent on path length and are
estimated as the probability of exceeding a given
(calculated) fade margin
1433
Fading margin
1434
1434
717
Fading margin
Safety margin.
Should be large enough
to guarantee that quality
and availability
objectives are met
during fading
conditions.
Typical value 40 dB
1435
Fading
1436
718
Fading types
Multipath fading
Flat fading
Frequency-selective fading
Rain fading
Refraction-diffraction fading
1437
Multipath fading
Multipath fading is the dominant fading mechanism for
frequencies lower than 10 GHz.
A reflected wave causes a multipath, i.e. when a reflected
wave reaches the receiver as the direct wave that travels
in a straight line from the transmitter
If the two waves reach the receiver out of phase they
weaken the overall signal.
As a thumb rule, multipath fading, for radio links having
bandwidths less than 40MHz and path lengths less than
30km is described as flat instead of frequency selective
1438
719
Flat and frequency selective

fading
Flat fading
A fade where all frequencies in the channel bandwidth are
equally affected.
If necessary, flat fade margin of a link can be improved by
using larger antennas, a higher-power microwave
transmitter, lowerloss feed line and splitting a longer
path into two shorter hops
On water paths at frequencies above 3 GHz, it is
advantageous to choose vertical polarization
Frequency-selective fading
There are amplitude and delay distortions across the
channel bandwidth
It affects high capacity radio links
1439
Multipath fading
1440
720
Rain fading
Rain attenuates the signal caused by the scattering and
absorption of electromagnetic waves by rain drops
It is significant for long paths (>10km)
It starts increasing at about 10 GHz and for frequencies above
15 GHz, rain fading is the dominant fading mechanism
Rain outage increases dramatically with frequency and then
with path length
Microwave path lengths must be reduced in areas where rain
outages are severe
The available rainfall data is usually in the form of a statistical
description of the amount of rain that falls at a given
measurement point over a period of time
1441
Reducing the effects of rain
Multipath fading is at its minimum during periods of

heavy rainfall, so entire path fade margin is available
to combat the rain attenuation
Route diversity with paths separated by more than a
few km can be used successfully
Vertical polarization is far less susceptible to rainfall
attenuation (40 to 60%) than horizontal polarization.
1442
721
Fading margin
Fading margin > Rain fading + Multipath fading

Rain fading
Dominant cause of fading for f> 10 GHz
Multipath fading
Dominant cause of fading for f< 10 GHz
Common fading margin: 25 - 40 dB
1443
Refraction diffraction fading

Also known as k-type fading
For low k values, the fictive Earths surface becomes curved and
terrain irregularities, man-made structures and other objects may
intercept the Fresnel zone.
For high k values, the fictive Earths surface gets close to a plane
surface and better LOS (lower antenna height) is obtained
The probability of refraction-diffraction fading is therefore indirectly
connected to obstruction attenuation for a given value of Earth
radius factor
Since the Earth-radius factor is not constant, the probability of
refraction-diffraction fading calculation is based on cumulative
distributions of the Earth-radius factor
1444
722
Interference fade margin

To accurately predict the performance of a digital radio path,
the effect of interference must be considered.
Interference in microwave systems is caused by the presence
of an undesired signal in a receiver.
When this undesired signal exceeds certain limiting values, the
quality of the desired received signal is affected.
To maintain reliable service, the ratio of the desired received
signal to the (undesired) interfering signal should always be
larger than the threshold value.
1445
Interference fade margin

In normal unfaded conditions the digital signal can
tolerate high levels of interference but in deep fades
it is critical to control interference.
Adjacent-channel interference fade margin (AIFM)
(in decibels) accounts for receiver threshold
degradation due to interference from adjacent channel
transmitters
Interference fade margin (IFM) is the depth of fade to
the point at which RF interference degrades the BER
to 10-3 .
1446
723
Far-end interference
1447
Near-end interference
1448
724
Adjacent channel interference
1449
Output power
1450
725
Output power
1451
Antennas
1452
726
Threshold degradation
1453
Threshold degradation
1454
727
Link budget
1455
1455
Microwave link multipath outage models
A major concern for microwave system users is how

often and for how long a system might be out of
service.
An outage in a digital microwave link occurs with a
loss of digital signal frame sync for more than 10 sec.
Digital signal frame loss typically occurs when the
BER increases beyond 10-3.
1456
728
Outage and availability

Outage (Unavailability) (%) = (SES/t) x 100
where
t = time period (expressed in seconds)
SES = severely errored second (1-s intervals in which the BER
exeeds 10-3 )
Availability is expressed as a percentage as :

A = 100 - Outage (Unavailability)(%)
A digital link is unavailable for service after a ten

consecutive BER>10-3 SES outage periods.
1457
Quality and Availability (Q&A)

The main purpose of the quality and availability
calculations is to set up reasonable quality and availability
objectives for the microwave path.
The ITU-T recommendations G.801, G.821 and G.826
define error performance and availability objectives.
The objectives of digital links are divided into separate
grades: high, medium and local grade.
The following grades are usually used in wireless
networks
Medium grade for the access network
High grade for the backbone network
1458
729
Improving the microwave system

Diversity improvement
Space Diversity
Polarity Diversity
Frequency Diversity
Route Diversity
Hybrid Diversity
Media Diversity
Hardware Redundancy
Hot standby
Repeaters
1459
Diversity operation
Diversity reception means the simultaneous reception of the
same radio signal over two or more paths.
Each path is handled by a separate receiver chain and then
combined in the radio equipment so that effects of fading are
mitigated.
The separate diversity paths can be based on space, frequency,
and/or time diversity.
The simplest form of diversity is space diversity.
The two diversity paths in space diversity are derived at the
receiver end from two separate receivers with a combined
output.
Each receiver is connected to its own antenna, separated
vertically on the same tower.
The separation distance should be at least 70 wavelengths and
preferably 100 wavelengths.
Probably, fading will not occur on both paths simultaneously.
1460
730
Diversity antenna systems
Combining
Combined signal
fed to receiver Signal 2
Signal 1
C o m b in e d s ig n a l
S ig n al 1
S ig n al 2
Signal Strength
T im e
1461
Concept of diversity
Fade
Transmission
media 1
Information Receiver
Transmission
Tmedia 2
Peak
1462
731
Diversity
Primarily used to reduce the effects of multipath
fading.
1463
Diversity
1464
732
A space-diversity configuration
1465
Antenna separation
1466
733
Correlation between branches

The branch correlation coefficient (r) represents the degree of
similarity between the signals from two different receiving
branches.
The correlation coefficient ranges from 0 to 1.
r=1 means the signals from two different branches behave exactly
the same. In this case, the signals are coherent.
r=0 means the signals from two different branches behave
completely different. In this case, the signals are uncorrelated.
To achieve the best performance, a diversity antenna system is
required to provide uncorrelated signals.
For r=1, the diversity antenna becomes ineffective in combating the
multipath fading.
A diversity antenna system can perform satisfactorily provided that
r<0.7.
1467
Probability that R-distributed field is less than abscissa

value with different number of diversity channels
Number of diversity channels 1468
734
Frequency diversity
Frequency diversity is more complex and more costly than
space diversity.
It has advantages as well as disadvantages.
Frequency diversity requires two transmitters at the near end
of the link.
The transmitters are modulated simultaneously by the same
signal but transmit on different frequencies.
Frequency separation must be at least 2%, but 5% is
preferable.
The two diversity paths are derived in the frequency domain.
When a fade occurs on one frequency, it will probably not
occur on the other frequency.
1469
A frequency-diversity configuration
1470
735
Frequency diversity
The more one frequency is separated from the other, the less chance
fades will occur simultaneously on each path.
Frequency diversity is expensive, but there is increased assurance of
path reliability.
It provides full and simple equipment redundancy and has the
great operational advantage of two complete end-to-end electrical
paths.
In this case, failure of one transmitter or one receiver will not
interrupt service, and a transmitter and/or a receiver can be taken out
of service for maintenance.
The primary disadvantage of frequency diversity is that it doubles
the amount of frequency spectrum required in this day when
spectrum is at a premium.
In many cases it is prohibited by national licensing authorities.
For example, the U.S. Federal Communications Commission (FCC)
does not permit frequency diversity for industrial users.
1471
Hot-standby
The full equipment redundancy aspect is very attractive to the
system designer.
Another approach to achieve diversity improvement in propagation
plus reliability improvement by fully redundant equipment is to
resort to the hot-standby technique.
On the receive end of the path, a space-diversity configuration is
used.
On the transmit end a second transmitter is installed as in in the
previous figure, but the second transmitter is on hot standby.
This means that the second transmitter is on but its signal is not
radiated by the antenna.
On a one-for-one basis the second transmitter is on the same
frequency as the first transmitter.
On failure of transmitter 1, transmitter 2 is switched on
automatically.
1472
736
One-for-N hot standby

One-for-N hot standby is utilized on large radio link systems employing
several radio carriers, where the cost for duplicate equipment for each
channel may be prohibitive.
On such multi-RF channel arrangements it is customary to assign some sort
of priority arrangement.
Often the priority channel enjoys the advantage of frequency diversity,
whereas the other RF channels do not.
On failure of one of the other channels, the diversity improvement is lost on
the priority channel, with the diversity pair switched to carry the traffic on
the failed pair.
In another arrangement the standby channel carries low-priority traffic and
does not operate in a frequency diversity arrangement, while providing
protection for perhaps two or three other channels carrying the higher-
priority traffic.
On failure of a high-priority channel, the lower-priority channel drops its
traffic, replacing one of the RF channels carrying the more important traffic
flow.
Once this occurs, the remaining channels operate without standby
equipment protection.
1473
Diversity improvement
Propagation reliability improvement can be exemplified as
follows.
If a 50 km path required a 51 dB fade margin to achieve a
99.999% reliability on 7 GHz without diversity, with space
diversity on the same path, only a 33-dB fade margin would be
required for the same propagation reliability, namely,
99.999%.
In the non-diversity condition, assuming Rayleigh fading, a
30-dB fade margin would display 99.9% path reliability.
But under the same circumstances with frequency diversity,
with only a 1% frequency separation, propagation reliability
on the same path would be improved to 99.995%.
1474
737
LOS microwave repeaters
Digital LOS microwave repeaters completely

demodulate the incoming signal to baseband (i.e., to
the raw electrical signal of 1s and 0s.).
This full demodulation causes regeneration of the
1s and 0s.
The outgoing signal is squared up and retimed.
The regenerated baseband signal is remodulated,
upconverted and retransmitted on a different
frequency.
1475
Repeaters
Repeater receives signal

Repeater amplifies
the signal and
rebroadcasts the
signal
Poor Coverage area
Repeater units are designed to receive signals from a donor site,

amplify and rebroadcast the donor sites signals into poor coverage
areas or to extend the coverage range of the donor site.
1476
738
Path/Site survey
This can turn out to be the most important step in the
design of an LOS microwave link (or hop).
Much of the survey is to verify findings and
conclusions of the path profile.
Of course each site must be visited to determine the
location of the tower and the radio equipment shelter.
Site/path survey personnel must personally inspect
the sites in question, walking/driving the path or
flying in a helicopter, or a combination of these.
The positions must be accurately enough (<1m)
cross-checked in three dimensions
1477
Frequency planning
The objective of frequency planning is to
assign frequencies to a network using as few
frequencies as possible and in a manner such
that the quality and availability of the radio
link path is minimally affected by
interference.
1478
739
Frequency planning and frequency

assignment
To derive optimum performance from an LOS microwave system, the
design engineer must set out a frequency-usage plan that may or may not
have to be approved by the national regulatory organization.
The problem to choose frequencies has many aspects.
In practice, the useful RF spectrum is limited to about 150 GHz.
The upper limit is technology-restricted.
To some extent it is also propagation-restricted.
There is the problem of congestion
Those frequencies above 10 GHz could be called rainfall-restricted,
because above 10 GHz excess attenuation due to rainfall can become an
important design factor.
Around urban and built-up areas, frequency assignments below 10 GHz are
hard to obtain from national regulatory authorities.
However, if we plan properly for excess rainfall attenuation, nearly equal
performance is available at those higher frequencies.
1479
Some LOS microwave frequency bands
1480
740
Channel plan
1481
Frequency selection
Scenario 1: (Perfect)
Frequency band and sub-band decided.
Early ordering of equipment possible.
Scenario 2: (Critical)
Frequency band unknown.
Major changes of assumed bands may alter the
network design.
Application to frequency authority must be made.
1482
741
Frequency planning
Determining factors for the influence of interference:
The carrier to interference ratio (C/I) at the input of the
interfered receiver
The receiver selectivity i.e. receiver bandwidth and the
frequency separation between carrier and interfering signal
The spectrum width of the interfering signal
The influence of interference can be reduced by
increasing C/I and/or the frequency separation.
1483
Frequency planning/Interference
reducing methods
Frequency (channel) separation
Transmitter attenuation (at interfering transmitter)
Cross-polar antenna discrimination (V/H)
Co-polar antenna discrimination (directivity)
Larger antennas (smaller beam width and higher
antenna gain)
High performance antennas (side- and back lobe
attenuation)
Reduced spectrum width
1484
742
Frequency planning
The following aspects are the basic considerations
involved in the assignment of radio frequencies
Determining a frequency band that is suitable for the
specific link (path length, site location, terrain topography
and atmospheric effects)
Prevention of mutual interference such as interference
among radio frequency channels in the actual path,
interference to and from other radio paths, interference to
and from satellite communication systems
Correct selection of a frequency band allows the required
transmission capacity while efficiently utilizing the
available radio frequency spectrum
1485
Frequency channel arrangements
The available frequency band is subdivided into

two halves, a lower and an upper duplex half.
The duplex spacing is always sufficiently large so
that the radio equipment can operate interference
free under duplex operation.
The width of each channel depends on the capacity
of the radio link and the type of modulation used
1486
743
Frequency planning tools
Frequency planning of a few paths can be

carried out manually but, for larger networks (=
for complex microwave design), it is profitable
to employ some software transmission design
tool.
Tools include standards, different diversity
schemes, diffraction and reflection (multipath)
analysis, rain effects, interference analysis etc.
1487
Path/Site survey
Line of sight must be precisely studied
reiteration for each obstacle in the LOS microwave
path
at least 0.6 of the first Fresnel zone must be added
on top of obstacles including 20 m for trees and a
few meters more for growth if in a vegetation area
(to avoid foliage loss penalties)
often it is advisable to add a few meters of safety
factor on top of the first Fresnel zone clearance to
avoid any diffraction loss penalties.
1488
744
System tests
A series of tests should be carried out to verify

if the link (or system) meets the performance
requirements.
The first is the measurement of receive signal
level (RSL) at a links far-end receiver.
The second test is the bit error rate test
(BERT).
Ideally, the tests should be done over time.
This means to run the BERT continuously for
at least a few days or more to capture the
effects of fading.
1489
Basic recommendations
Use higher frequency bands for shorter hops and
lower frequency bands for longer hops
Avoid lower frequency bands in urban areas
In areas with heavy precipitation, if possible, use
frequency bands below 10 GHz.
Use protected (stand-by) systems (1+1) for all
important and/or high-capacity links
Leave enough spare capacity for future expansion of
the system
1490
745
Common planning considerations

Perform detailed path surveys on ALL microwave
hops.
Maps are used only for initial planning, as a first
approximation.
Below 10 GHz, multipath outage increases rapidly
with path length.
Multipath effect can be reduced with higher fade
margin.
If the path has excessive path outage the performance
can be improved by using one of the diversity
methods.
However, diversity is an expensive way of improving
the performance of the microwave link and it should
be used carefully and as a last resort
1491
Difficult areas for microwave

links
In areas with lots of rain, use the lowest
frequency band allowed for the project.
Microwave hops over or in the vicinity of the
large water surfaces and flat land areas can
cause severe multipath fading.
Reflections may be avoided by selecting sites
that are shielded from the reflected rays.
1492
746

Radio Engineering by Jorma Kekalainen

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Radio Engineering by Jorma Kekalainen

Încărcat de

Drepturi de autor:

Formate disponibile

Lecture notes Radio Engineering by Jorma Kekalainen

The field strength changes periodically.

Polarization of a vector field

Wavelength and frequency ranges of

Fractions and multiples

Note: Angstrom is a unit of length used in the study of spectra (spectroscopy). 16

EM field and matter

EM field and matter

Dipole polarization in material

Note. If the material medium is isotropic, both

Sourceless medium and EM waves

Gathering information about the

Electromagnetic radiation: Wave model

Wave model of EM energy James Clerk Maxwell (1831-1879)

Unified theories of electricity and magnetism (via

Electric field (E) perpendicular to magnetic field (M)

Note angles in radians (rad)

James Clerk Maxwell (1831-1879) 39

3. Faradays Law of Induction: Line integral of electric B

4. Amperes Law: For a static electric field, the line integral

Note: is divergence operator and x is curl operator. 40

Note: Sometimes it is used wavenumber, k = 1/ [m-1]. 43

Energy radiated from sun or active sensor

We will see how energy is related to frequency, f (and hence

Electromagnetic spectrum: visible

Electromagnetic spectrum: microwave

Energy from a blackbody?

Note that peak of suns energy around 0.5 m

Generalization of Stefan-Boltzmann Law

Radiation emitted from unit area of any plane surface

Peak of emitted radiation: Wiens Law

Wiens (Displacement) Law

Straight line in log-

Particle model of radiation

Einstein explained photoelectric effect by

Light has both wave- and particle-like

Particle model of radiation

Particle model of radiation

Bohr and Planck recognized discrete nature of transitions.

Particle model of radiation

Particle model of radiation

Plancks Law of blackbody radiation

Consequences of Plancks Law: Plants

Applications of Plancks Law

Total radiant energy

Applications of Plancks Law: example

Finally, F00.38 =0.193*(0.273-0.067)+0.067=0.11

Deviation from blackbody assumption

Plancks Law - shape of power spectrum for given T (Wm-2 m-1)

Invention of radio waves

Heinrich Hertz: I do not think that the wireless

Note: Radio spectrum is only a tiny fraction of electromagnetic spectrum

Microwave transmission uses frequencies above

Radio communication system

RF and IF input spectrum

Software defined radio

Software Defined Radio (SDR)

Signals and waveforms

Need for higher frequencies

Elements of a radio system

1. Message signal source

The source of the message signal may be

2. Amplifier and 3. Oscillator

4. RF amplifier and frequency