NISI

PUBLICATIONS.'

DATE DUE
UCl 1 s h;^

fEb

^7Q

ia--o-"3;

(o-n-^a r
JAl<

m7
AJft-v.'t>-

h-i^fiom

CA VtOBO

PRINTED INU.S.A.

jm

14

it'DD

^ecUnlcciL

flote

92. /o/

TRANSMISSION LOSS PREDICTIONS FOR

TROPOSPHERIC COMMUNICATION CIRCUITS

VOLUME

p. L.

RICE,

A.G.

LONGLEY, K.A.NORTON,

AND

A. P.

BARSIS

U.

S.

DEPARTMENT OF COMMERCE

NATIONAL BUREAU OF STANDARDS

THE NATIONAL BUREAU OF STANDARDS

The National Bureau

maximum

of Standards is a principal focal point in the Federal Government for assuring application of the physical and engineering sciences to the advancement of technology in

industry and commerce. Its responsibilities include development and maintenance of the national standards of measurement, and the provisions of means for making measurements consistent with those standards; determination of physical constants and properties of materials; development of methods for testing materials, mechanisms, and structures, and making such tests as may be necessary, particularly for government agencies; cooperation in the establishment of standard practices for incorporation in codes and specifications; advisory service to government agencies on scientific and technical problems; invention and development of devices to serve special needs of the Government; assistance to industry, business, and consumers in the development and acceptance of commercial standards and simplified trade practice recommendations; administration of programs in cooperation with United States business groups and standards organizations for the development of international standards of practice; and maintenance of a clearinghouse for the collection and dissemination of scientific, technical, and engineering information. The scope of the Bureau's activities is suggested in the following listing of its four Institutes and their organizational units.

Institute for Basic Standards.

plied Mathematics.

Electricity. Metrology. Heat. Radiation Physics. Mechanics. ApAtomic Physics. Physical Chemistry. Laboratory Astrophysics.* Radio Standards Laboratory: Radio Standards Physics; Radio Standards Engineering.** OflSce of Standard Ref-

erence Data.

Institute for Materials Research. Analytical Chemistry. Polymers. Metallurgy. Inorganic Materials. Reactor Radiations. Cryogenics.** OflBce of Standard Reference Materials.

Central Radio Propagation Laboratory.** Ionosphere Research and Propagation. Troposphere and Space Telecommunications. Radio Systems. Upper Atmosphere and Space Physics.
Institute for

Applied Technology.

Textiles

Industrial Equipment.

Transport Systems. ing Standards. Office of Industrial Services.

Information Technology. Office of Technical Services.

and Apparel Technology Center. Building Research. Performance Test Development. Instrumentation. Office of Weights and Measures. Office of Engineer-

* NBS Group, Joint Institute for Laboratory Astrophysics at the University of Colorado. ** Located at Boulder, Colorado.

NATIONAL BUREAU OF STANDARDS

technical 92ote fOf
Issued

May

7,

1965

TRANSMISSION LOSS PREDICTIONS FOR TROPOSPHERIC COMMUNICATION CIRCUITS

VOLUME
1

P. L. Rice, A. G. Longley, K. A. Norton, and A. P. Borsis

Central Radio Propagation Laboratory

National Bureau of Standards

Boulder, Colorado

NBS Technical Notes are

reau's

designed to supplement the Buregular publications program. They provide a means for making available scientific data that are of transient or limited interest. Technical Notes may be listed or referred to in the open literature.

For sale by the Superintendent of Documents, U. S. Government Printing Office Washington, D.C. 20402
Price:

$1.00

National

Bureau of Standards

AUG

5
'

1965

FOREWORD
A
methods were
Note will permit the reader
to

150,SSg

QOlOOJJSlS'd
short history of the development of the prediction methods in this Technical

compare them with earlier procedures,,

Rice and Vogler [1955].

Some

of these of

first reported by Norton,

Further development

forward scatter predictions and a better understanding

the

of the refractive index structure of

atmosphere led

to

changes reported in an early unpublished

NBS

report and in
of

NBS

Technical Note 15 [Rice, Longley and Norton, 1959].

The methods

Technical Note 15

served as a basis for part

of

another xinpublished

NBS

report which was incorporated in

Air Force Technical Order T.O. 31Z-10-1 in 1961.

technical note

preliminary draft
to the

of the

current
in 1962.

was submitted as a U.S. Study Group V contribution

CCIR

Technical Note 101 uses the metric system throughout.

For most computations

These

both a graphical method and formulas suitable for a digital computer are presented.

include simple and comprehensive formulas for computing diffraction over smooth earth and

over irregular terrain, as well as methods for estimating diffraction over an isolated rounded
obstacle.

New empirical graphs are

For paths

included for estimating long-term variability for sev-

eral climatic regions, based on data that have been

in a continental
in 1961.

made

available to NBS.

temperate climate, these predictions are practically

the

same as those published

The reader will find that a number

of

graphs have been

simplified and that

many

of the calculations

are more readily adaptable to computer program-

ming.

The new nnaterial on time availability and service probability in several climatic re-

gions should prove valuable for areas other than the U.S.A.

Note:

This Technical Note consists of two volumes as indicated in the Table of Contents.

TABLE OF CONTENTS
Volume
1.
1

PAGE NO.
INTRODUCTION
1-1
2.

THE CONCEPTS OF SYSTEM LOSS, TRANSMISSION LOSS, PATH ANTENNA GAIN, AND PATH ANTENNA POWER GAIN
2,
1

2-1
2-1

System Loss and Transmission Loss

Available Power from the Receiving Antenna

2.2
2.3

2-3

Antenna Directive Gain and Power Gain

Polarization Coupling Loss and Multipath Coupling Loss

2-6
2-8

2.4
2. 5

Path Loss, Basic Transmission Loss, Path Antenna Gain, and Attenuation Relative to Free Space
Propagation Loss and Field Strength

2-10

2. 6

2-13
3-1
3-1

3.

ATMOSPHERIC ABSORPTION
3,
1

Absorption by Water Vapor and Oxygen

3.2
3.3
3. 4

Sky-Noise Temperature
Attenuation by Rain Attenuation in Clouds

3-3

3-4 3-6
4-1

4.

DETERMINATION OF AN EFFECTIVE EARTH'S RADIUS

TRANSMISSION LOSS PREDICTION METHODS FOR WITHIN- THE

5.

HORIZON PATHS
5.
1

5-1

Line-of-Sight Propagation Over a Smooth or Uniformly Rougly Spherical Earth /

5. 1. 1

5-1

curve-fit to terrain
o",

.\

...

5-5
5-6

5,2

h Line-of-Sight Propagation Over Irregular and Cluttered

5.

1.2

The terrain roughness factor,

................

Terrain
6.

5-7

DETERMINATION OF ANGULAR DISTANCE FOR TRANSHORIZON PATHS

6. 1 6. 2

6-1

Plotting a Great Circle Path

Plotting a Terrain Profile and Determining the Location Radio Horizon Obstacles
of

6-1

6-3

6, 3

Calculation of Effective Antenna Heights for Transhorizon Paths

Calculation of the Angular Distance, 9
.

6-4

6.4
7,

6-5
7-1 7-1 7-3

DIFFRACTION OVER A SINGLE ISOLATED OBSTACLE

7. 1

Single Knife Edge,

No Ground Reflections
Ground Reflections

7.2

Single Knife Edge with

PAGE NO.
7.3

Isolated Rounded Obstacle,

No Ground Reflections
Ground Reflections

7-4

7.4
8.

Isolated Rounded Obstacle with

7-6

DIFFRACTION OVER SMOOTH EARTH AND OVER IRREGULAR TERRAIN

8.
1

8-1
8-1 8-i

Diffraction Attenuation Over a Smooth Earth Diffraction Over Irregular Terrain

8.Z,
1

8.Z

Diffraction over paths where

d
st
,

=d sr

8-4 8-4
8-5
9-1

8.2,2
8.3
9.

For horizontal polarization

Single-Horizon Paths, Obstacle not Isolated

FORWARD SCATTER
9.
1

The Attenuation Function,

F(ed)

9-2
9-3

9.2
9.3

The Frequency Gain Function,

H
F

The Scattering Efficiency Correction,

9-5

9.4
9.5
10.

Expected Values

of

Forward Scatter Multipath Coupling Loss

....

9-6
9-7
10-1

Connbination of Diffraction and Scatter Transmission Loss

LONG-TERM POWER FADING

10.
1

The Effective Distance,

The Functions V(50, d
e
)

d
e

10-7
e
)

10.2
10.3

and Y(p, d

10-8

Continental Temperate Climate

10-9

10.4
10.5

Maritime Temperate Climate

Other Climates
Variability for Knife-Edge Diffraction Paths
.

10-12
10-13

10.6
11. 12.

10-13
11-1

REFERENCES
LIST OF

SYMBOLS AND ABBREVIATIONS

12-1

IV

4 8 1

TABLE OF CONTENTS
Volume
2

PAGE NO.

ANNEX

I:

AVAILABLE DATA AND STANDARD CURVES

Available Data as a Function of Path Length
.

I-l I-l

I. 1

1.2

Standard Point-to-Point Transmission Curves

1-2

ANNEX
n.
n.

n:

BEAM ORIENTATION, POLARIZATION, AND MULTIPATH

COUPLING LOSS
II. .

Representation of Complex Vector Fields

IIII-

Principal and Cross-Polarization Components

Unit

II. 3

Complex Polarization Vectors

......

n-6
II-

n. 4

Power Flux Densities

Polarization Efficiency

n.5
n. 6
II. 7

11-10

Multipath Coupling Loss

............
....

11-12

Idealized Theoretical Antenna Patterns

11-15

n.8

Conclusions

n-23

ANNEX

in:

FORMULAS, COMPUTER METHODS, AND SAMPLE CALCULATIONS

Line-of-Sight
Diffraction Over a Single Isolated Obstacle

III. 1

m-1 m-2
in-15
in- 17
III- 23

III. 2

ni.3

Diffraction Over a Single Isolated Obstacle with Ground Reflections

.

ni.4
ni. 5

Parameters K and b

for

Smooth Earth Diffraction

Forward

Scatter

in-24
in-37

m. 6 m.
7

Transmission Loss with Antenna Beams Elevated or Directed Out of the Great Circle Plane

Long-Term Power Fading

in. 7.
1

in-44

Diurnal and seasonal variability in a continental temperate climate

m-45
in- 50

in. 7.2

To mix distributions

m.

Examples
III. 8. 1 III.

in-69
Line of oight predictions
Transinission loss prediction for a rounded isolated
obstacle
III- 73
.

Ill-

69

8.2

III.

8.3

Predicted transmission loss for a transhorizon path

III- 77

PAGE

NO.

ANNEX ANNEX
V,

IV:

FORWARD SCATTER
PHASE INTERFERENCE FADING AND SERVICE PROBABILITY
The Two Components
of
.

IV-

V:
1

V-1

Fading

V-3

V.2
V.3

The Nakagami-Rice Distribution

Noise -Limited Service

V-5
V-11 V-13
of

V.4
V.
5

Interference-Limited Service

The Joint Effect

Simultaneously

of

Several Sources

Interference Present

V-17
V-18

V. V.

6
7

The System Equation for Noise -Limited Service The Time Availability
of

Interference-Limited Service

V-ZO V-21

V.8
V. 9

The Estimation

of

Prediction Error

The Calculation of Service Probability Availability p

for a

Given Time

V-23

V. 10

Optimum Use

of the

Radio Frequency Spectrum

V-Z9

VI

TRANSMISSION LOSS PREDICTIONS FOR

TROPOSPHERIG COMMUNICATION CIRCUITS

P. L. Rice, A. G. Longley, K. A, Norton, and A. P. Barsis

1.

INTRODUCTION

This report presents comprehensive methods for predicting cumulative distributions

of

transmission loss for a wide range

of

radio frequencies over any type of terrain and in sevof

eral climatic regions.

Such quantitative estimates

propagation characteristics help to de-

termine how well proposed radio systems will meet requirements for satisfactory service,
free

from harmful interference.

Thus they should provide an important step toward more

spectrum.
is

ef-

ficient use of the radio frequency

The need

for

comprehensive and accurate prediction methods

loss data for a large
I. 1

clearly demonstrated

when measured transmission

tion of path length.

number

of

radio paths are shown as a funcof attenuation

In figures

to

I.

4 of annex

I,

long-term median values

relative to free space for

more

than 750 radio paths are plotted versus distance.

The ex-

tremely wide scatter

files

of these data is

due mainly to path-to-path differences in terrain pro-

and effective antenna heights.

Values recorded for a long period of time over a single

path show comparable ranges, sometimes exceeding 100 decibels.

Such tremendous path-to-

path and time variations must be carefully considered, particularly in cases of possible in-

terference between co-channel or adjacent-channel systems.

The detailed point-to-point prediction methods described here depend on propagation

path geometry, atmospheric refractivity near the surface of the earth, and specified characteristics of antenna directivity.

They have been tested against measurements

in the radio

frequency range 40 to

10,000

MHz

(megacycles per second).

Extension of the methods

to higher frequencies requires

estimates of attenuation due to absorption and scattering of

radio energy by various constituents of the atmosphere.

Predictions of long-term median reference values of transmission loss are based on

current radio propagation theory.

pirical predictions of regional,

A
to

large sample of radio data was used to develop the

em-

seasonal, and diurnal changes in long-term medians.

Esti-

mates

of

long-term fading relative

of time,

observed medians are given for several climatic regions

available.

and periods

including

some regions where few observations are

Predictions of transmission loss for paths within the radio horizon are based on geometricoptics ray theory.

For paths with a common horizon, Fresnel-Kirchoff knife-eage diffraction

theory

is

applied and extended to predict diffraction attenuation over isolated rounded obstacles.
that extend only slightly

For double horizon paths

beyond the horizon, a modification

of the

Van der Pol-Bremmer method

for

computing

field intensity in the far diffraction region is

1-1

used.

For longer paths, extending well beyond

Radio data were used

the radio horizon,

to

predictions are based on

at

forward scatter theory.

estimate the efficiency of scattering

exists as to which propagation

var-

ious heights in the atmosphere.

Where some doubt

mechan-

ism predominates, transmission

bined.

loss is calculated by two methods and the results are

com-

Annex

includes a set of "standard" curves of basic transmission loss and curves

showing attenuation below free space for earth space communications, prepared using the

methods described
I.
1

in the report.

Such curves, and the medians of data shown on figures

but clearly do not take account of

to 1.4,

may serve

for general qualitative analysis,

particular terrain profiles or climatic effects that

may

be encountered over a given path.

Annex

II

supplements the discussion

2.

of

transmission loss and directive antenna gains

of

given in section
tion,

This annex contains a discussion

antenna

beam

orientation,

polariza-

and multipath coupling loss.

Annex
for

III

contains information required for unusual paths, including exact formulas

computing line-of-sight transmission loss with ground reflections, as well as modifica-

tions of the formulas for antenna

beams which are

elevated,

or directed out of the great

for use on a digital

circle plane.

Sample calculations and analytic expressions suitable

com-

puter are also included.

Annex IV reviews tropospheric propagation theory with particular

attention to the

mechanisms

of

forward scatter from atmospheric turbulence, from layers, or from small

References
of;

randomly oriented surfaces.

to

some

of the

work

in this field are included.

Annex V presents
term power
fading",

a discussion

"phase interference fading" as contrasted to "long-

provides a method for computing the probability of obtaining adequate

service in the presence of noise and/ or interfering signals, and includes a brief
of

summary

ways

to

achieve optimum use of the radio frequency spectrum.

Figures are placed at the end of each section, and those which are not vertical
should be turned counter-clockwise.

(The ordinate labels would be upside down

if

the usual

convention were followed.)

Previous

Technical Notes in this series, numbered 95

to 103,

describe tropo1961],

spheric propagation phenomena and siting problems [Kirby, Rice, and Maloney,

certain meteorological phenomena and their influence on tropospheric propagation [Button,

1961; Button and Thayer,

1961],

synoptic radio meteorology [Bean, Horn, and Riggs,

[

1962], techniques for measuring the refractive index of the atmosphere

Mc Gavin,

1962],

determination
for

of

system parameters
links [Barsis,
et al,

Florman and Tary,

1962],

perfornnance predictions

communication
[

Norton, Rice, and Elder,

.

I96I],

and equipnnient charac-

teristics

Barghaufen,

1963]

1-2

2.

THE CONCEPTS OF SYSTEM LOSS, TRANSMISSION LOSS, PATH ANTENNA GAIN, AND PATH ANTENNA POWER GAIN
Definitions have been given in

CCIR Recommendation
L
.

341 for system loss,

L
s

trans-

mission

loss,

L, propagation loss,

basic transmission loss,

path antenna gain.

and path antenna power gain, G P PP troduces a definition of "path loss", L

,

This section restates some of the definitions, in,

illustrates the use of these

1953,

terms and concepts,

The notation used

and describes methods of measurement [Norton,

1959,

Wait 1959].

here differs slightly from that used in Recommendation 341 and in Report 112 [CCIR 1963a, b].

For

the frequency range considered in this report

system

loss,

transmission loss, and propa-

gation loss can be considered equal with negligible error in almost all cases,

because antenna

gains and antenna circuit resistances are essentially those encountered in free space.

\>
2.1

System Loss and Transmis sion Loss

radio circuit consisting of a transmitting antenna, receiving an-

The system loss

of a

tenna, and the intervening propagation

medium

is

defined as the dimensionless ratio,

Pl/p'>

where
p'

p'

is

the radio frequency

power input

to the terminals of the transmitting antenna

and

is the

resultant radio frequency signal power available at the terminals of the receiving

antenna.

The system loss

is

usually expressed in decibels:

L
s

:=

10 log (p'/p') ^ P' - P'

t

db

(2.1)

Throughout

this

report logarithms are to the base 10 unless otherwise stated.

of

The inclusion

ground and dielectric losses and antenna circuit losses

in

pro-

vides a quantity which can be directly and accurately measured.

In propagation studies,

however,

it is

convenient to deal with related quantities such as transmission loss and basic

transmission loss which can be derived only from theoretical estimates of radiated power and
available power for various hypothetical situations.
In this report,

capital letters are often used to denote the ratios,

expressed

in

db,

dbu,

or

dbw, of the corresponding quantities designated with lower-case type.

P' = 10 log p!
in
is

For instance,

in (2. 1),

dbw corresponds

to

p'

in watts.
p,

Transmission loss
total

defined as the dimensionless ratio

/p
of

where p

is the

power radiated from

the transmitting antenna in a given

band

radio frequencies, and

is

the resultant radio frequency signal

power which would be available from an equivalent

is

loss-free antenna.

The transmission loss

usually expressed in decibels:

= 10 log (p /p
t

= P^ t

L
s

et

er

db

(2.2)

et

= 10 log

et

er

=10

log

er

(2.3)

2-1

where

/i

et

and

IjH

er

as defined in the next subsection are power radiation and reception

effeciencies for the transmitting and receiving antennas, respectively.

With the frequencies

and antenna heights usually considered for tropospheric communication circuits, these efficiencies are nearly unity and the difference between L and L is negligible. With antennas a fraction of a wavelength above ground, as they usually are at lower frequencies, and
especially when horizontal polarization
is

used,

and L

are not negligible, but are

influenced substantially by the presence of the ground and other nearby portions of the an-

tenna environment.

From
Transmitter Output

transmitter output to receiver input, the following symbols are used:

Power
Input to

Total

Available Power
at

Radiated

Loss-Free

Power

Antenna

Power

Receiving Antenna

Available Power at Actual Receiving Antenna

Available Power
at

Recei'ver Input

^ir

P|_

are conceptually different. Since P and r J r ^ ^ represent the power observed at the transmitter and at the transmitting antenna, respecIt

should be noted that

L^ it

and L

tively,

L
t

includes both transmission line and

mismatch

losses.

Since

P'

represent available power at the receiving antenna and at the receiver, mismatch losses must
be accounted for separately, since L
includes only the transmission line loss between the

air
and P

antenna and the receiver.

2-2

I
2. 2

Available

Power

from, the Receiving Antenna

The above definitions of system loss and transmission loss depend upon the concept
of available

power, the power that would be delivered to the receiving antenna load
to the receiving
,

if its

impedance were conjugately matched

frequency
v

antenna impedance.

For a given radio

in hertz,

let

and

represent the impedances of the load, the

actual lossy antenna in its actual environment, and an equivalent loss-free antenna, respectively:
c

"iv^

"iv

+
iv

(2.4a)

z' = r' f ix' V V V

:

(2.4b)

Z
V

::

r
V

+ ix

(2.4c)

where

and x in (2.4) represent resistance and reactance, respectively.

p'

Let p.
,

rep-

resent the power delivered to the receiving antei.na load and write
for the available
of the

and p

respectivt

power

at the

terminals of the actual receiving antenna and at the terminals

If

equivalent loss-free receiving antenna.

v'

is

the actual open-circuit

r.m.s,

voltage at the antenna terminals, then

V ,2 r
V

Piv=77- z' + z^
'

iv

(^-^J

iv'

When
r,

the load
V

impedance conjugately matches the antenna impedance, so

iv
= -x'
V
,

that

= z'

or

iv

= r'

and x,
p'

(2.5)

shows that the power delivered p, ^ ^iv

to the load is equal to ^

the

power

av

available from the actual antenna:

v.^

P^v-iT^
V

('-'^

Note that the available power from an antenna depends only upon the characteristics of the
antenna,
its

open-circuit voltage

v'

and the resistance

r'

and

is

independent of the load

2-3

impedance.

Comparing

(2,5) and (2,6),

we define a mismatch loss factor

^l^^X^Ji
^iv

P'

r'
I

+ r

^\)

Y+ ^x'
\.

+ X, ^"Z

(2

7)

4 r' r V iv

such that the power delivered

to a

load equals
i

p'

av

/i

mv

When

the load

impedance coniu^
-^

gately matches the antenna impedance, ^ ' ^

p'
,

= value of unity, and p ^ mv ^iv For any other load impedance, somewhat less than the available power is delivered to

has

its

minimum

the load.

The power available from the equivalent loss-free antenna

V ^~
4 r
V
2

is

Pa V

(2.8)

where v

is

the open circuit voltage for the equivalent loss-free antenna.

(2.8),
it

Comparing (2,6) and

should be noted that the available power

is

p'

at the

terminals of the actual lossy receiving antenna

for a loss-free antenna at the

less than the available power

av

p'

erv

av

same location as

the actual antenna:

p
-2

r*

erw

..=-:f^=-^^^i p' i2
V
V

(2.9)

The open circuit voltage

circuit voltage

v'

for the actual lossy antenna will often be the

same as

the open

for the equivalent loss-free antenna, but

each receiving antenna circuit

must be considered

individually.
p'

Similarly, for the transmitting antenna, the ratio of the total power
the antenna at a frequency
v

delivered to
v:

is

times the

total

power p

radiated at the frequency

^ .

etv

P'

^tw

/p. tv

(2.10) ^

'

The concept

of available
is

power from

a transmitter is not a useful one, and

for the trans-

mitting antenna

best defined as the above ratio.

However,

the

magnitude

of this ratio

can

be obtained by calculation or

measurement by
i

treating the transmitting antenna as a receiving

antenna and then determining

to be the ratio of the available

received powers from the

equivalent loss-free and the actual antennas, respectively.

General discussions

of

i
1

are given by Crichlow et al

1955] and in a report prei

pared under CCIR Resolution No.

[Geneva 1963c].

The loss factor

was successfully

2-4

determined

in one

case by measuring the power p

radiated from a loss-free target trans-

mitting antenna and calculating the transmission loss between the target transmitting antenna

and the receiving antenna.

There appears

to

be no way of directly measuring either

or

without calculating

some

quantity such as the radiation resistance or the transmission

loss.

In the case of reception with a unidirectional

i.

rhombic terminated
2

in its characteristic

impedance,
the received

could theoretically be greater than

is

[Harper, 1941], since nearly half

power

dissipated in the terminating impedance and

[

some

is

dissipated in the

ground.

Measurements were made by Christiansen

of

1947] on single and multiple wire units

and arrays

rhombics.

The ratio

of

power lost
3

in the

termination to the input power varied

with frequency and was typically less than

db.

For
tors

the frequency band

to

it is

convenient to define the effective loss fac-

er

and L

as follows:
at

p
- 10 log

m
(d

P,,/dv) dv

er

db
(d
'^i

(2.11)

p;^/dv) dv

p
_ 10 log
et

m
(d

P^/dv) dv
(2.12)

j-j

(d Pt,/dv)
"i

dv

The limits

and

on the integrals (2.11) and (2.12) are chosen to include esv

sentially all of the wanted signal modulation side bands, but

is

chosen

to be sufficiently

large and

sufficiently small to exclude any appreciable

harmonic or other unwanted radia-

tion emanating

from

the

wanted signal transmitting antenna.

2-5

2. 3

Antenna Directive Gain and Power Gain

g (r)

transmitting antenna has a directive gain

(1)
it

in the direction of a unit vector f

if:

radiates a total of
its

watts through the surface of any large sphere

with the antenna at

(2)
it

center, and

radiates

g p

/{4Tr)

watts per steradian in the direction

g'(r)

r.

The same antenna has a power gain

(1) (2)

in the direction
is

if:
,

the
it

power input

to the

antenna terminals

= I

and
r.

radiates

g'p'/(4Tr)

watts per steradian in the direction

The antenna power gain

of the loss factor
i
.

g'

is

smaller than the directive gain g

simply as a result

It

et

follows that

G^(r) = G'(f

+ L^^

(2.13a)

expressed
gain G'(f)
t

in decibels
is

above the gain

of

an isotropic radiator.

Note that the antenna power

db,
et

less than the antenna directive gain

1/i
et
is

G
t

(r)

by the amount L
.

where

the

power radiation efficiency

The gain
of

independent

of the
it is

direction r

an antenna

is the

same whether

used for transmitting or receiving.

r

For a receiving antenna,

the directive gain

(r)

and power gain G'(f) are related by

G^{f) = G^(r) +

{2.13b)

The remainder

of this

report will deal with directive gains, since the power gains

mav

be determined simply by subtracting

)

or
in

The
II,

maximum
it is

value of a directive
to divide

gain G(f

is

designated simply as

G.

As noted

Annex

sometimes useful
5

the directive gain into principal and cross-polarization

components.

An

idealized antenna in free space with a half- power semi-beamwidth

expressed
x
percent

in radians,
of its

and with a circular beam cross- section,

to

may

be

assumed

to radiate

power isotropically through an area equal

Tr5

on the surface of a large sphere of

unit radius, and to radiate (100-x) percent of its

of the sphere.

power isotropically through the remainder

In this case the

power radiated
gain g

in the direction of the

is,

main beam

is

equal to
2
).

xp /(100it6

2
)

watts and the

a

maximum
g =

by definition, equal to 4Trx/(l OOtt 6

One
lOdb

may assume

beam

solid angle efficiency

x
2

= 56
.

percent for parabolic reflectors with

tapered illumination, and obtain

2.24/6

The

maximum

free space gain

in decibels

relative to an isotropic radiator is then

10 log g = 3. 50 - 20 log 6

db

(2. 14)

2-6

If

azimuthal and vertical beamwidths

25
\v

and

25

are different:
z

w 5z

(2.15)

The above analysis

is

useful in connection with

measured antenna radiation patterns.

For antennas such as horns

or parabolic reflectors which have a clearly definable

is

physical aperture, the concept of antenna aperture efficiency

free space

useful.

For example,

the

maximum

gain of a parabolic dish with a 56 percent aperture efficiency and a di-

ameter D

is the ratio of 56

percent

of its

area

to the effective

absorbing area of an iso-

tropic radiator:

= 10 log

Q.

56ttD^/4

= 20 log

+ 20 log

f -

42. 10

db

(2.16)

X^/4u

where D and
Equations
their

X.

are in meters and

is the

radio frequency in megahertz,

MHz,

(2. 14)

and

(2. 16)

are useful for determining the gains of actual antennas only when

beam

solid angle efficiencies or aperture efficiencies are known, and these can be de-

termined accurately only by measurement. With a dipole feed, for instance, and 10 < D/X. <
lowing emipirical formiula to be superior to
(2. 16):

25,

experiments have shown the

fol-

G
where D

= 23.3 log

+ 23.3 log

f -

55.

db

(2.17)

is

expressed
[

in

meters and

in

MHz.
for determining paraboloidal

Cozzens

1962] has published a

nomograph

maximum
commonly-

gain as a function of feed pattern and angular aperture.

Discussions
I960]

of a variety of
.

used antennas are given in recent books

Jasik,

1961; Thourel,

Much more
Most

is

known about

the amiplitude, phase, and polarization response of avail-

able antennas in the directions of

of the theoretical

maximum

radiation or reception than in other directions.

and developmental work has concentrated on minimizing the transof the

mission loss between antennas and on studies

standard plane wave.

response of an arbitrary antenna to a

is

An

increasing amount of attention, however,

in

being devoted to maxi-

mizing the transmission loss between antennas

this

order to reject unwanted signals.

in statistical

For

purpose

it is

important to be able to specify, sometimes

terms, the

directivity, phase, and polarization response of

an antenna

in

every direction from which

multipath components
this subject.

of

each unwanted signal

may

be expected.

Appendix

II is

devoted

to

For the frequencies

any radio frequency
v

of interest in this report, antenna radiation resistances

r
V

at

hertz are usually

assumed independent

of their

environment, or

else the immediate environment is considered part of the antenna, as in the case of an

antenna mounted on an airplane or space vehicle.

2-7

'

2.4
It is

Polarization Coupling Loss and Multipath Coupling Loss

to

sometimes necessary

minimize

the response of a receiving antenna to unof different paths.

wanted signals coming from

a single

source by way

This requires atof

tention to the amplitudes, polarizations, and relative phases of a

number

waves arriving

from different directions.

In any theoretical model,

the

phases of principal and cross-

polarization components of each wave, as well as the relative phase response of the receiving

antenna to each component, must be considered.

Complex voltages are added

at the an-

tenna terminals to nnake proper allowance for this amplitude and phase information.
In

Annex

II it is

shown how complex vectors

and

may

be used to represent

transmitting and receiving antenna radiation and reception patterns which will contain amplitude, polarization, and phase information [Kales,
\, e
,

1951] for a given free- space wavelength,

e

A
e

bar
c

is

used under the symbol for a complex vector

are real vectors which

may

+ i e where i = \j - I and c p be associated with principal and cross-polarized com= e

,

ponents of a uniform elliptically polarized plane wave.

Calculating the power transfer between two antennas in free space, complex polarization vectors
p(r)

and p (-r) are determined for each antenna as

if it

were

the transmitter

and the other were the receiver.

of the other:

Each antenna must be

in the far field or radiation field

p(r)^7/\^

J-r)

=7/g\

(2.18)

+ie cr e+ie,e=e pr p
c

(2.19)

^1^

= e^ +

e^
c

1^

1^

e^ + e^ cr pr

(2.20)

The sense

of polarization of the field

is

right-handed or left-handed depending on whether

a
,

the axial ratio of the polarization ellipse,

is

positive or negative:

= e c

/e

(2.21)

= e and linear if e =0, c c p principal polarization direction defined by the unit vector e
is

The polarization

circular

if

where

= e e

is in

the

P P

The available power p

may

be written as

= s(r)

a (-r)

|p

watts

(2.22)

2-i

s{r) = |e|

r 12
l{Zr\
)

watts /km

(2.23)

ag(-?) - g^(-?)

X^/(4Tr)]

km^

(2. 24)

where

s(r)

is the total

mean power

flux density at the receiving antenna, a (-r)

-f,

is

the ef-

fective absorbing area of the receiving antenna in the direction

and

Ip

is the
|

po-

larization efficiency for a transfer of energy in free space and at a single radio frequency.

The corresponding polarization coupling loss

is

L^p

-10 log

g".

p^P
15)

db

(2.25)

In
4*

terms

of the axial ratios

and a

defined by
e

(II.

and

(11.

17)

and the acute angle

between principal polarization vectors

and

the polarization efficiency

may

be

written as
cos
2
ij*

(a

lp.pl'=

"^
II

2
1)

+ sin

2
4^

(a

+ a

2
)

P
, ,

"

"^

(2.26)

(a^-Hl)(a^^+l)
This
is the

same as

(11.29).

Annex

explains

how these

definitions and relationships are ex-

tended to the general case where antennas are not in free space.

There

is a

rnaximum transfer

of

power between two antennas

if

the polarization el-

lipse of the receiving antenna has the

same sense,

eccentricity, and principal polarization

direction as the polarization ellipse of the incident radio wave.

The receiving antenna

is

completely

"blind" to the incident wave

if

the sense of polarization is opposite, the eccenis

tricity is the

same, and the principal polarization direction

In theory this situation

orthogonal to that of the inof

cident wave.

would result

in the

complete rejection

an unwanted

signal propagating in a direction

-f

Small values

of

g (-r)

could at the same time dis-

criminate against unwanted signals coming from other directions.

When more than one

source, there

plane wave

is incident

upon a receiving antenna from a single

may

be a "multipath coupling loss" which includes

beam
of

orientation, polari-

zation coupling, and phase

fects,

mismatch losses. A

statistical
is

average

phase incoherence ef-

such as that described in subsection 9.4,

is the

called "antenna- to- medium coupling loss."

Multipath coupling loss

next subsection.
in

same as

the 'loss in path antenna gain, "

defined in the

Precise expressions for L

gP

may

also be derived from the relationships

annex

II.

2-9

2.5

Path Loss, Basic Transmission Loss, Path Antenna Gain, and Attenuation Relative

to

Free Space
Observations of transmission loss are often normalized to values of "path loss"
subtracting the
by-

sum

of the

maximum
is

free space gains of the antennas,

from

the

transmission loss, L.

Path loss

defined as

G
t

db.
r

(2.27)
'

Path loss should not be confused with basic transmission loss.

loss,

Basic transmission
at the

is

the

system loss for a situation where the actual antennas are replaced

same

locations by hypothetical antennas which are:

(1)

Isotropic, so that
tion directions,
r.

(r) =

db and

G
L

(-f) =

db for

all

important propaga-

(2)
(3)

Loss-free, so that

Free

=0 db. er of polarization and multipath coupling loss,

L
=

db and

et

so that

L
,

cp
is

db

Corresponding

to this

same
if

situation, the path antenna gain,

defined as the

change in the transmission loss

hypothetical loss-free isotropic antennas with no multi-

path coupling loss were used at the

same

locations as the actual antennas.

is

The transmission loss between any two antennas

defined by {2.

2):

L
where P

P
t

db

dbw

is

the total

power radiated from the transmitting antenna and P

is

dbw

is

the

corresponding available power from a loss-free receiving antenna which

lent to the actual receiving antenna.

otherwise equiva-

Replace both antennas by loss-free isotropic antennas

coupling loss between

at the

same

locations, with no

them and having

the

same

radiation resistances as the actual antennas,

represent the resulting available power at the terminals of the hypothetical ab the path antenna gain isotropic receiving antenna. Then the basic transmission loss L

and

let

Gp'

power * gain and the path antenna ^ "^

G
pp
P,
t

are given by ' ^

L^

P P P

ab

^ =

G
p

db

(2. 28)

G
p

ab

= L,

L L

db

(2. 29a)

pp

P' a

ab

^ 3

L^

db
s

(2.

29b)

2-10

In free space, for instance:

^a

^t ^ ^t^^^ ^ ^r<-^^

"

^cp

+ ^ ^S

(~4^
dbw

^^^

<2. 30a)

P A
s

ab

P
t

+ 20 log

f^ 4iTr
V

"^
'

3( ((2. ^

special symbol,

is

used to denote the corresponding basic transmission loss

in free

pa c e

L
where
equals

20 log

(^J^
r
is

32.45 + 20 log

+ 20 log r

db

(2.31)

the antenna separation

expressed

in

kilometers and the free space wavelength

f

X.

0.2997925/f kilometers for a radio frequency

in

megahertz.

When low

gain antennas are used, as on aircraft, the frequency dependence in (2.31)

indicates that the service range for

UHF equipment
in direct

can be made equal to that

in the

VHF
Fixed

band only by using additional power

proportion to the square of the frequency.

point-to-point communications links usually employ high-gain antennas at each terminal, and
for a given antenna size

more

gain

is

realized at

pensating for the additional free space loss at

UHF than at VHF, thus more UHF indicated in (2.3l).

than com-

Comparing
space,

(2. 28),

(2. 29),

and

(2. 30),

it is

seen that the path antenna gain in free

.,

is

pf

G
t

(f)

(-f)
r

cp

db

(2,32)
^

'

For most wanted propagation paths,

this is well

approximated by G + G
t

the

sum

of the

maximum G
.
.

antenna gains.

For unwanted propagation paths

it is

often desirable to

minimize

This can be achieved not only by making

(r)

and

(-?)

small, but also by using

different polarizations for receiving and transmitting antennas so as to

In free space the

maximize L

transmission loss

is

The concepts

of basic

transmission loss and path antenna gain are also useful for normalizing

the results of propagation studies for paths

lent free- space transmission loss", L^,

which are not

in free space.

Defining an "equiva-

as

2-11

note that
path.

in (2. 34) is not equal to

unless this

is

true for the actual propagation

It is

often convenient to investigate the "attenuation relative to free space",

A,

or

the basic transmission loss relative to that in free space, defined here as

"^^^-^f = ^-^f
This definition, with
terrain has
(2, 34),

^^

(^=35)

makes A independent

of the path
it is

antenna gain,

G
P

Where

little effect

on line- of- sight propagation,

L..

sometimes desirable

to study

rather than the transmission loss,

Although G

varies with time,

it is

customary

to

suppress this variation

Hartman,

1963] and to estimate only the quantity

G
and L

pm

= L,

bm (50)

m (50)
of

(2,36)

where

L,

bm (50)

m (50)

are long-term median values

in

L, b

and L.

Multipath coupling loss, or the "loss

path antenna gain",

is

is

defined as the

difference between basic transmission loss and path loss, which

equal to the

sum

of the

maximum

gains of the transmitting and receiving antennas minus the path antenna gain:

L
gp

L^

G
t

G
r

G
p

db

(2 371 ^'^^'1

The loss

in

path antenna gain will therefore, in general, include components of

beam

orientathat

tion loss and polarization coupling lOSS as v/ell as any aperture-

to-medium coupling loss

may

result from scattering by the troposphere, by rough or irregular terrain, or by terrain

clutter such as vegetation, buildings, bridges, or

power

lines.

2-12

2, 6

Propagation Loss and Field Strength

This subsection defines terms that are most useful at radio frequencies lower than
those where tropospheric propagation effects are dominant.

Repeating the definitions

of

and

r'

used in subsection 2.2, and introducing the new

parameter
r

t.r
r'
t,

antenna radiation resistance,

resistance component of antenna input impedance,

= r
fr

r,
ft,

^r -

antenna radiation resistance in free space,

t

where subscripts
tively.

and

refer to the transmitting antenna and receiving antenna, respec-

Next define

L^^=
L^^=

10 log (r'/r^),

L^^ =

10 log

(rVr^)

(2.38)

10 log (r;/r^^),

L^^ = 10 log (rVr^^)

(2.39)

L^^=10

1og(r^/r^^) =

L^^-L^^

(2.40a)

\^
[Actually, (2.11) and (2.12) define

= 10 log (r^/r^^) = L^^

L^^

(2.

40b)

and L

while (2.38) defines

and

given

r'

and

r'

r-'

Propagation loss first defined by Wait [1959]

[1963a] as

is

defined by the

CCIR

L
p

L
s

L
ft

L
fr

rt

L rr

db

f2 V-^o^J-;

4n

Basic propagation loss

is

Basic propagation loss in free space

Lj^^,

is the

same as

the basic transmission loss in free space,

defined by

(2. 31).
is

The system loss L^ defined by (2.1)

L, path loss

measurable quantity, while transmission loss

attenuation relative to free space
in

L^,

basic transmission loss

L^,

A, propa-

gation loss

Lp,

and the field strength E are derived quantities, which

L^^^

general require a

theoretical calculation of
in

^^ and/or

L^^^^

^^ as well as a theoretical estimate of the loss

path antenna gain

2-13

The following paragraphs explain why

the concepts of effective radiated powel-,

E.R. P.

and an equivalent plane wave field strength are not reconnmended for reporting propagation data.

A
equal to

half -wave antenna radiating a total of

watts produces a free space field intensity

1.64p

/(4-n-r

watts /km

(2.43)

at a distance

kilometers in
1.64,

its

equatorial plane, where the directive gain

is

equal to

its

maximum
antenna.

value

or

2.15 db.

The
s

field is linearly polarized in the direction of the

In general, the field intensity

at a point

in free

space and associated with

the principal polarization for an antenna is

(r)=p

g {r)/(47rr

watts /km

(2.44)

as explained in annex

II.

In
r
.

(2.44),

tive gain in the direction

(r) is the principal polarization direcP similar relation holds for the field intensity s ( r ) associated

r =

rr

and

with the cross -polarized component of the field.

Effective radiated power is associated with a prescribed polarization for a test antenna

and

is

determined by comparing
s

as calculated using a field intensity meter or standard

the test antenna:

signal source with

as
P

measured using

E.R. P.

=P t +10

1og(s /s ) p o

P
t

+Gpt (rj
1

2.15

dbw

(2.45)

where

in free space is the direction

towards the receiving antenna and in general

to the receiver.

is the

initial direction of the

most important propagation path

This ambiguity in definition, with the difficulties which sometimes arise in attempting
to separate characteristics of

an antenna from those

of its

environment, make the effective

total radiated

radiated power

E. R. P.

an inferior parameter, compared with the The following equation, with

power

which can be more readily measured.

(2.45),

determined from

may

be used to convert reported values of E.R. P.

to estimates of the transmitter

power output ^ ^
tion efficiency

P
it

when transmission
1/i

line

and mismatch losses

L,
it

and the power radia-

et

are known:

P,
it

=P t +L et =P'+L, it t
discussed in annex

+ L,
it

dbw

(2.46)

The electromagnetic

field

II is

complex vector function

is

in space

and time, and information about amplitude, polarization, and phase

it.

required to describe
E, which
field,

real antenna responds to the total field surrounding

to the

it,

rather than to

corresponds

r.m.

s.

amplitude of the usual "equivalent"

electromagnetic

defined at a single point and for a specified polarization.

2-14

For converting reported values

the available
ful:

of

in

dbu

to

estimates of

or

estimates of

power

at the input to a receiver, the following relationships

may

be use-

P
it

E E

+ L,
it

+ L,
ft

G
t

L,

pb

20 logf
*

107.22

dbw
dbw

(2. 47) ^
'

L
ir

ir

L, +
fr

Gr
ir

- L,

gp
-

20 logf
'^

107.22

(2.

48)

P,
ir

= P'

L,

er

L,
ir

dbw

(2.49)

In

terms

of reported values of field strength

in

dbu per kilowatt of effective

radiated power, estimates of the system loss,

basic propagation loss

or basic

transmission loss

Ll

may
+

be derived from the following equations,

Ls

= 139.37

et

+ L^

fr

-G
p +

+ G^
t

Gpt (r J 1

+ 20 log

f -

ikw
,

db

(2.50)

Lpb ^
L,

= 139.37 -

rt

Gt

G (r) 1
pt

+ 20 logf

E
E

ikw
, ,

db

(2. 51)

= 139.37

+L rr +Gt

Gpt J? 1 J

+ 20 log

f -

ikw

db

(2.52)

provided that estimates are available for all of the terms in these equations. For an antenna whose radiation resistance is unaffected by the proximity
vironment,
rr tant for frequencies less than
rt

of its en-

=0

db,

L,
ft

L
et

and L^
fr

In other cases,

especially imporit

er

30
db,

MHz
L
ft

with antenna heights

=

commonly
L^
fr =

used,

is often

as-

sumed
to the

that

rt

rr

3.01

L
et

+ 3.01 db,

and

er

+ 3.01 db,

corresponding

assumption

of short vertical electric dipoles

above a perfectly-conducting infinite plane.

At low and very low frequencies, '

curves at

et

HF

and lower frequencies

may

er ft be given in

L,

and L,
fr

may
of

be very large.
or
I_

Propagation
it is

terms

L
p

pb

so that

not

necessary

to specify

and
et

er

Naturally,

it is

better to

measure

L L
s
,

directly than to calculate

it

using

(2.50).

It

may

be seen that the careful definition of

L
p

L,

or

is

simpler and more direct

than the definition of

K, L

>

A,

or

E.

The equivalent free -space

radiated ^ power
L^ = L.^ =
it ft

field strength

E P
t

in
=

dbu for one kilowatt

= 30

of effective

is

obtained by ^ substituting ^

P,
it
^

E. R. P.

dbw,

(r ,)= pt^ 1'

\

G
t
I

2.15 db,

db,

and

L, pb

= L, ^

bf

in

(2.45) ^

(2.47), '

where

L,

bf

is given

by (2.31): 1
(2.53)

106.92

20 log d

dbu/kw

where

in

(2.31)

has been replaced by

in

(2.53).

Thus

is

224.3

millivolts

per meter at one kilometer or

139.4 millivolts per meter at one mile.

In free space, the

2-15

"equivalent inverse distance field strength",

tion resistances
r
ir
,

E^,
I

is the

same

as

If

the antenna radiar

ft

r
t

and

r r

are equal to the free space radiation resistances ^ '^

and

then (2.52) provides the following relationship between

^ =

E
1

kw

and

with

Gpt^(rj 1'

G
t

= 139.37

+ 20 logf

- L,

dbu/kw

(2. 54)

Consider a short vertical electric dipole above a perfectly-conducting

infinite plane,

with

E.R. P.

= 30

dbw, G^:=1.76db, and

t

G A^ .)

- 1- 76 db.

Then from
E,
I

(2. 52) the

rr t equivalent inverse distance field is

= 3.

db From

(2. 45)

30.39 dbw,

since

E+L, +L rr rt o
mv/m

109.54

20 log d

dbu/kw

(2. 55) '

*

corresponding to

300

at one kilometer, or

e
is

= 186.4

mv/m

at one mile.

In

this situation, the relationship

between

and

L^

given by (2.52) as

E,

ikw

142.38+ 20 logf

- L,

dbu/kw

(2.56) *
'

The foregoing suggests the following general expressions for

strength

the equivalent free space field

and the equivalent inverse distance

field

(P
t

rt

Gt )+

20 log d + 74.77

dbu

(2.57)

E^
I

rt

rr

dbu

(2. 58)

Note that
antenna in
nition of

L
its

in

(2.57)

is

not zero unless the radiation resistance of the transmitting

its

actual environment is equal to

free space radiation resistance.

The defi-

"attenuation relative to free space"

given by (2.35) as the basic transmission

loss relative to that in free space,

may be
- L,

restated as

= L,

bf

L,
f

E^
I

db

(2. 59)

Alternatively, attenuation relative to free space,

might have been defined (as

it

some-

times

is)

as basic propagation loss relative to that in free space:

A
t

pb

L,

bf

.-

--

rt

rr

db

(2.60)

For frequencies and antenna heights where these definitions

db, caution should be

differ

by as much as

used in reporting data.

For most paths using frequencies above

2-16

50

MHz,

L^^ +
L,

L_^^
is

is negligible,

but caution should again be used

It is

if

the loss in path

antenna gain

not negligible.

then important not to confuse the

"equivalei

free space loss

L^

given by (2.34) with the loss in free space given by (2.33).

2-17

3.

ATMOSPHERIC ABSORPTION
2

At frequencies above

GHz

attenuation of radio waves due to absorption or scatter-

ing by constituents of the atmosphere, and by particles in the atmosphere,

affect

may

seriously

microwave relay
1

links,

communication via
the total radio

satellites,

and radio and radar astronomy.

At frequencies below

GHz

wave absorption by oxygen and water vapor for

2 decibels.

propagation paths of 1000 kilometers or less will not exceed

rainfall begins to be barely noticeable at frequencies

Absorption by
but

from

2 to 3

GHz,

may

be quite

appreciable at higher frequencies.

For frequencies up

to 100

GHz, and

for both optical and transhorizon paths, this

section provides estimates of the long-term median attenuation

Aa

of radio

waves by oxygen

and water vapor, the attenuation

tion

due to rainfall, and the order of magnitude of absorp-

by clouds of a given water content.

[

The estimates are based on work reported by

[

Artman and Gordon

[

1954], Bean and Abbott

[

1957

],
[

Bussey
1959
] ,

1950

],

Crawford and Hogg

[

1956

] ,

Gunn and East

[

1954
],

] ,

Hathaway and Evans

[

Hogg and Mumford

[

I960

Hogg and Semplak

1961

Lane and Saxton

[

1952

],

Laws and Parsons

[

1943

],

Perlat and

Voge[l953], Straiten and Tolbert

[1947a,
b;

I960

],

Tolbert and Straiten

1957

and Van Vleck

1951].
3. 1

Absorption by Water Vapor and Oxygen

at a

Water vapor absorption has a resonant peak

frequency of 22.23 GHz, and oxygen

absorption peaks at a number of frequencies from 53 to 66

3.1,

GHz

and at 120 GHz.

Figure

derived from a critical appraisal of the above references, shows the differential absorp-

and V in decibels per kilometer for both oxygen and water vapor, as deter Y wo 'oo mined for standard conditions of temperature and pressure and for a surface value of
tion

absolute humidity equal to 10 grams per cubic meter.

those prepared for the Xth Plenary

These values are consistent with

U. S. Study

Assembly
is

of the

CCIR by
7.5

Group IV

1963d

except that the water vapor density

there taken to be

g/m

3
.

For the range

of absolute
is

humidity likely to occur in the atmosphere, the water vapor absorption in

imately proportional to the water vapor density.

db/km

approx-

The
is

total

atmospheric absorption
in one of

decibels for a path of length

kilometers

commonly expressed
^(r)

two ways, either as the integral

of the differential

absorp-

tion

dr

Jq

dr Y(r) ^

db

(3.

1)

or in terms of an absorption coefficient

r(r)

expressed in reciprocal kilometers:

10 log

exp

r(r) dr
-

=4.343

\
*^

r(r) dr

db

(3.2)

^0

3-1

The argument

of the

logarithm in

(3. 2) is the

amount

of radiov/ave

energy that

is not

absorbed

in traversing the path.

The

total

gaseous absorption

A
r

over a line-of-sight path of length

a

kilometers

is

A
cL

dr [y (h) + Y (h)] O W

db

(3.3)

where h

is the

height above sea level at a distance

from

the lower terminal,

total

measured
2A
db.
a

along a ray path between terminals.

For radar returns, the

absorption

is

Considering oxygen absorption and water vapor absorption separately, (3.3)

written

may

be

Aa
where
r

= Y

oo eo

+ y

wo ew

db

(3.4)

are effective distances obtained by integrating and r eo ew over the ray path.

y /y o oo

and y
^v

/y

wo

The effective distances

vati on angles

9=0,
.

and r are plotted versus ew eo 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, L and tt/
r
r

and frequency for eleo

radians in figures
d,

2-3. 4.

Figure
values

3. 5

shows the relationship between

o

and the sea level arc distance,

for these

of

may

be estimated from figures

I.

21 to
r

I.

26 of annex

I,

where attenuation relative

to free space.

A,

is

plotted versus
the ray

f,

and

ignoring effects of diffraction by terrain.

to its

For nonoptical paths,

or
9

from each antenna

horizon makes an angle

6.
1

or

with the horizontal at the horizon, as illustrated in figure

at distances
is the

of section 6.

The

horizon rays intersect

d
of

and d
values

from

the transmitting and receiving terminals.

The

total absorption

Aa

sum

A at
+

and

A ar

A
where

A at

A ar
(3.5)

i^

at

(f,

dj,
1

ot

ar

Aa

(f,

d^)
2

or

For propagation over a smooth earth,

horizon paths and the frequency range

9 ot
0.
1

=9
-

or

=0, and A

=
a

10

GHz,

a figure 3. 6 shov/s

2A

(f,

0,

d/2).

For trans-

plotted versus

distance over a smooth earth between 10 meter antenna heights.

3^2

3.2

Sky-Noise Temperature
source
of

The nonionized atmosphere

reradiator that
it

is a

radio noise, with the

same properties as
T
s

has as an absorber.

The effective sky-noise temperature

may

be de-

termined by integrating the gas temperature

T multiplied by

the differential fraction of re-

radiated power that is not absorbed in passing through the atmosphere to the antenna:

(K) =

T(r)r(r) exp r.
L

r(r')dr'[dr

(3.6)

where
stance,

the absorption coefficient

r(r)

in reciprocal kilometers is defined by (3.

2)

For

in-

assuming
T(r) = (288 - 6. 5h) "K

for

h < 12 km,

and
T(r) = 210K
for

h > 12 km,

figures.? shows the sky-noise temperature due to oxygen and water vapor for various angles
of elevation

and for frequencies between

0.

and 100 GHz.

the antenna pattern

In estimating antenna temperatures,

and radiation fronn the earth's

surface must also be considered.

3-3

3.3

Attenuation by Rain

The attenuation
the

of

radio waves by suspended water droplets and rain often exceeds

combined oxygen and water vapor absorption.

all

Water droplets

in fog or rain will scatter to the

radio waves in

directions whether the drops are small

compared

wavelength or

comparable

to the

wavelength.

In the latter case,

raindrops trap and absorb some of the

is

radio wave energy; accordingly, rain attenuation

lengths than at centimeter wavelengths.
In practice
it

much more

serious at millimeter wave-

has been convenient to express rain attenuation as a function of the pre-

cipitation rate,

which depends on both

the liquid water content and the fall velocity of

the drops, the latter in turn depending on the size of the drops.
that rain with a

There

is little

evidence
of

known

rate of fall has a unique drop- size distribution, and the

of

problem
is quite

estimating the attenuation

difficult.

radio waves by the various forms of precipitation

Total absorption

due to rainfall over a path of length

can be estimated by

integrating the differential rain absorption y (r)dr along the direct path between two intervisible antennas,

or along horizon rays in the case of transhorizon propagation:

A
Fitting an arbitrary

=
r

V
^r

{r)dr

decibels

Jq

(3.7)

mathematical function empirically

[

to theoretical results given by

Hathaway and Evans

1959] and

Ryde and Ryde

1945], the rate of absorption by rain y

may

be expressed in terms of the rainfall rate

in

millimeters per hour as

V
r

KR* r

db/km

(3.8)

for frequencies above 2

GHz.

The fmictions

K(f

and

0!()

are plotted in figures 3.8

and

3.9,

where

is the

radio frequency in GHz.

K=
a

[3(f^-

2)^ -

2(f^- 2)] X 10"^

(3.9a)

[1.14

0.07(f^- 2)^] [1 + 0.085(f^-

3.5)

exp(-0.006 f^)

(3.9b)

An examination
form

of the variation of rainfall rate with height suggests a relation of the

\/\g

exp(-0.2 h^)

(3.10)

3-4

where R

rs

is the

surface rainfall rate.

Then

'rs

er

db.

(3.11)

\s
where
v
'r s

KR^^ db/km,
rs

r_ er

dr exp

(- 0. 2

ah

km

(3.12)

is the

surface value of the rate of absorption by rain, and

er

is

an "effective
o

rainbearine ^ distance".

Figures 3.10 -3.13 show ^

er

versus

for several values of 9

and

a.

The curves shown were computed using (3.12).

A "standard" long-term cumulative

using

distribution of rain absorption is estimated,

[

some

statistics

from Ohio analyzed by Bussey

1950],

who

relates the cumulative dis-

tribution of instantaneous path average rainfall rates for 25, 50, and 100-kilometer paths,

respectively, with the cumulative distributions for a single rain gauge of half-hour, one-hour,

and two-hour

mean

rainfall rates,

recorded for a year.

The

total

annual rainfall in Ohio

is

about 110 centimeters.

Rainfall statistics vary considerably
to year,

from region

to region,

sometimes from year

For instance,
along the

and often with the direction

of a path (with or

across prevailing winds).

lie

in

North America, east-west systems seem particularly vulnerable, as they

path of frequent heavy showers.

For very long paths, the cumulative distribution

fall

of

instantaneous path average rain-

rates,

Rr

depends on how

Rr

varies with elevation above the surface and upon the cor-

relation of rainfall with distance along the path.

Figure

3. 14

provides estimates
0. 1,
1,

of the

instantaneous path average rainfall rate

exceeded for 0.01,

total

and

percent

of the

year as a function

of r

er

and normalized to a
in (3.12)with

annual rainfall
3.

of 100

cm.

To obtain A

from

(3. 11),

replace

from figure

14, multiplied by the ratio of the

total annual rainfall

and 100 cm.

These estimates are an extrapolation

of the results given

by Bussey
given by
fall rate

1950] and are intended to allow for the average variation of

with height, as
of

(3. 10)and

allowed for in the definition

of r

and for the correlation

surface rain-

with distance along the surface, as analyzed by Bussey.

3-5

3.4 Attenuation in Clouds

Cloud droplets are regarded here as those water or ice particles having radii smaller
than 100 microns or 0.01 cm.

Although a rigorous approach to the problem of attenuation by

it is

clouds must consider drop- size distribution,

of clouds rather than the

more

practical to speak of the water content

drop-size distribution.

Reliable

measurements
of the

of both

parameters
of a

are scarce, but

it

is

possible to

make reasonable estimates

water content, M,
of

cloud from a knowledge of the vertical extent of the cloud

and the gradients

of

pressure,

temperature, and mixing ratio, which

of

is

the ratio of the

mass

water vapor to the mass

dry air in which

it is

mixed.

The absorption within a cloud can be written as

A
where A

K M

db

(3. 13)

is

the total absorption attenuation within the cloud,

3. 1,

K
1

is

an attenuation coefficient,

values for which are given in table

and

is the liquid

water content of the cloud,

water,

measured

in

grams per cubic meter.

The amount

of precipitable

M,

in a given

pressure layer can be obtained by evaluating the average mixing ratio

by the pressure difference, and dividing by the gravity.

in the layer, multiplying

Using this method of obtaining

and the values

of

K from

table 3.1

it is

possible to get a fairly reliable estimate of the

absorption of radio energy by a cloud.

Several important facts are demonstrated by table
with increasing frequency
is
3. 1.

The increase

in attenuation
of

clearly shown.

The values change by about an order

6

magnitude

from 10

to 30

GHz

Cloud attenuation can be safely neglected below

GHz.

The data
These

presented here also show that attenuation increases with decreasing temperature.

relations are a reflection of the dependence of the refractive index on both wavelength and

temperature.

The different dielectric properties

of

water and ice are illustrated by the

difference in attenuation.

Ice clouds give attenuations about two orders of

magnitude smaller

than water clouds

of the

same water

content.

TABLE
One -Way Attenuation Coefficient,
Temperature
(C) 33

3.1

in

db/km/gm/m
GHz,
17
0.

Frequency,
24
0.

9.4
0. 0483 0.0630 0.0858

Water
Cloud
J

20
10

0.647 0.681 0.99

1.25

311

0.406 0.532 0.684

-3 35 X 10"

0.

128 179

0.

0.267 34

(extrapolated)
Ice
8.

0. 112 (extrapolated)

74X

10"^

6.

4.

36

10""^

2.46 X
8.

lO"'^

Cloud

10

2.93 X 10"^ 2.0

2. 11

10"

-3
-3

1.46 X 10"^
1.0

19 X 10"^

20

X 10"^

1.45 X

10"

lO"'^

5.

63X 10"^

3-6

SURFACE VALUES Xoo AND Xwo OF ABSORPTION BY OXYGEN AND WATER VAPOR
PRESSURE 760mm Hg TEMPERATURE 20<C WATER VAPOR DENSITY I0g/m3

O
UJ Q.

m o UJ a

a. a:

o en
CO

<

LU

0.0005

0.0002

0.0001

''

"

EFFECTIVE DISTANCES

Tqq

AND

rg^

FOR ABSORPTION BY OXYGEN AND WATER VAPOR

00=0.
0.01,

0.02

DUU

tt>-u
A.

-nni

R9A

-nnp

AQr\

Af\f\

^'
^
.

'-

...=

--

--

__

__

__^j

^ T,
oil

7
7
lCf\

^^ ,2 ^'^
.i

I-,Z^'

-'^
-

a]
_

7/^

ion

z; ^ ,^^ :'

^-- ^

-''
ji
-

280 /

^?<777

-,?2=^
-:

^''
;:; =

,^

$;?2^
S2i'

Zll-^''=

-- = -'D.2
T"

2^~L

...s
= = = = =

210

*7^^ ^^%-^-

7^J

'''"' = --=^=^

.__d^_:__ __________________ ______ = = == = = 32!

=

5nn

/^V^-y
'Aiyxy,^'
^ <^''' '

A'/y

y-\^^
"j,:-i'

-<:::. -'j

""C^-r-' '"

.,__,

'x E==- = =' = =

,

-Jki^^ '
icn

/yyj^.-^

Tf

J i 55

= =

=== = = = = = = = = = = = = = =

==== =

=== = = = =

= = =LL

M
RT

f-r

/^
-7^g^?fei
'^S^^'^^''''' --''
^ -.^%
|S

'^

-"""

L- --=

:._.5 =--->3

:^=^ = =
li!U

g^--

-Z--=;:--::- = -^=

3^ = "'

:-

::__::__=:__|to

9^

80

zt*
sis

Ar\

7
7

z*

J'

a^

^
40
80
120
160
<

00

240

280

320

360

400

440
To

48 3

520

560

600

640

680

720

760

800

840

880

RAY PATH

IN

KILOMETERS

ou

fiO
L-J-l.
1

Ir ,1-100

Tf""

40

9o

60=0.01
60 = 0.02

^
.^

20

'

^
_. I. ._ __

If

RECJUENCY IN G c/s J

00 LJ

O z < (V)

X
80

/
..-

|0!-ionL

_.
,.

__

7^
60

/
^.---'

Q UJ > Io

,/

20

'j^

^ ^ <^
40

,''
''

-^

.^ -^

_.,

_.. _.. _.,

1 I-IOO

20

60

80

100

120

140

180

200

220

240

260

280

320

340

360

380

400

420

440

Figure 3.2

3-8

'

EFFECTIVE DISTANCES

r^^

AND

r^^

FOR ABSORPTION BY OXYGEN AND

^C
=

WATER VAPOR
1 1

0.05.

0.1,

.2

e=o.c
80
e--oi

/
/

[01

G<-/='
1

p^

9=0.2
60

40

20

//

//

//
/

//
/

y
FREQUENCY
IN Gc/s|

|0 2|

A /x/
1

00

80

/ A A^
y

/^.'

'

^^
v^

^ -rr
^

rr
^

Gc/5'

Jrv?L pi

60

40

4 ^ A
^$. .._.J^ i^r"1

V / ^?^
^rt=>
1

/^

j<

^r

^l^'. ^^0-- ^z kcr ::!

^
S

^
:d
0.1
.

^ "^^
'

'

_ ~

M ^
1

r~

I2J.H5I
L.

-'
1

rri ss.1'.:

g^r:

r -- -- ._ .-. -- M Gc/s - .-- |02 __ - -- - ^- ._. - :[2] ^ " ... "~ " ""?
-|

Tw

[E

-'>IOt

20

Si" rn
\^\

^>inL i^

Ti
180

20

40

200

220

240

260

280

300

320

340

360

380

400

420

RAY PATH

IN

KILOMETERS

).05

So-<

50

EQUENCY
;

IN

Gc/s^
|oi-ioo|
1

^-
0.1-10 TJ

^,>^\

^^^

^r

yC\ J^ 2

- -"

"
L

0.1-100
4
6
3

IC

12

14

16

18

20

Figure 3.3

3-9

EFFECTIVE DISTANCES

Teo

AND

r^^

00 =0.5.

FOR ABSORPTION BY OXYGEN AND WATER VAPOR v/2

I.

^,

10

12

14

16

20

22

24

26

28

30

32

34

36

38

40

42

44

46

48

50

52

RAY PATH
en

Tq

in

KILOMETERS

00=0.5

0 = 7r/2

D.I-IOUl
1

(J

CO
LlJ

Q >
\i^

/- -^
/^
>'^ ^

"^
;.i-iu

ipPFOllFMrV
rJ

IM nr/cl

^^-

=
3

"

1 l-IO(
II

^
13 14

12

10

12

15

16

17

19

20

21

22

23

24

25

Figure 3.4

3-10

'

d, IN

SEA LEVEL ARC DISTANCE

KILOMETERS

"~

::::::inir;T;::::
CL
V^ 5s

I"

\^

ll>

's
^.

^
^
^
T^l

o~"
1

S X.
:

^ _-

^ J

'

-Ps-_ _ ^^'
ISIP

_5

5^:.:; cb '' --' S A"" kPWJ 1 ^41^-41 1 ---

'

^Mi-^
"

V
:-

^^T

J iw

H
....

'

^m

I^^W
^
I

;E!;;i!;;;;:

:::

=i

sl

"^

.^1

> -<

3)

II

iiz^\\u\\-w\
:;S;;::!i::;;

;;;;

e^
ii;

:i;4>s^S^^

'It'

^^fl

::

Hi
J
'

==^

33-'

^fc^
_^
1

iU-~

ij
CO

m
H I
^
::> II

'

m!

^'''^^S
!'

T^sij I
*
=; ->

HiiNJ.
j|lii[>
"

l\t'-i

:r
-c -n

>

X\

:::: :::

T lu I ii:; I I
'

k \s A ? k

s^
"

-^

"i
1

"

v
i;^ jL.m

\3 V
t

-n

> H X
6^
-z
TJ

SSp kl5

!~ --

::::::pr::^!a :::::::g-p;:i
liLLl

X. ^ \

-^r^**!^
-

^^"5

^,-^hS 'i

;
:;::;::::

--.-it .1

it

m
en OJ

<

:-::'::t

S
fl-

c ^
(6

i::;;;;::::;

O
-

A
=1

Bffl

^Sljlj,

:::;;

rr-j

^+*^ i+-

Oi

en

n o s m H m Zl w

II

___
Nl

MMM SI
-T
'-'-IT

'--'ilF'E^lL^ ^^

3^.EEE

i||

'CT It -i

m
?
it 45 '^

CO

iii'tr^r:;;

?-T-i;n-^

--

^^p-tSr :^ik Tl-S" J -5=3

.1-

-|::: ::i"::!f

CTWH/

o
^5
.

~
II

"V M-i^lp
==-3
--

t^" ^fl... ::z rt:

Ol

pfim ^d
-i:-..j

p m > OJ rm en < m r> X

tn

CO

^,-^ vTlxv

o *

:V:::::i:i--=^:
::::i::::::::::::::;;;:!

sis /]\
V
1

i\ jV
1

:::::: ::::::::
\
1 , 1

lur
\
1
J
*

'"

J 1

vi
ij

I T
1

il

M
::

Al
\i
i

il'il

'it

P "i

\f nTT i nh

-^
^'1

H > 2 m

TV

11

4* :::

::
:

"

ii El

iliWlW
::
:

fci E3E B=^'"*

::
'

S"'

lift

i!

^^

^ ag
'^

W^

[imi:!:|;;;:|in'!
1
i
ri

Tnff'

J"

=
[lull

= ==
:::::::::
:

|:^ ttftH
IJJlIj.

....
(ji

::|ttH||iH

111

ili fffa
-Jf:
.-:
:

=Siil III ffl o~ TiF

t:
;::

IL+
-

^^ ir^

P :g^; '^W

IB

ii

3-1!

OXYGEN AND WATER VAPOR ABSORPTION

IN

DECIBELS

IW
1

\ \^

1\
I

% \^ s^ \, ^ \C\ \ \ \\ A \^ \ \ w \ .\ \\ \ \ \ \ \ \ \\ \ \
\,

\\\

H
> H m o
-n -n

o
CO

O ^ m o
t>

> z
c,

\
\

\
'

\
\
-

> > < >

O
2

>
3:

o m
5 o O O
o s o

o o
I

ro

O O O

Ol

O O o

c c3 c>

10,000

m O m o -< z
I

O o
JO

TJ

>
CD 10

O
ZO -0

1
1

\
3-12

"

, .

SKY NOISE TEMPERATURE DUE TO RERADIATION BY OXYGEN AND

WATER VAPOR
DUU
i^:::
=

|
I;;;!!

eee;;;

W
::::::
:

|;;|;;;;;; 1 Eliii T-

ffF

--=-:-|::::::

e;:;;

::::

^--|1|^
d:
:::::

^Hb a a
4 Hi^

in
eIJ;
::

|::

TO
0.1

0.5

m HtHtp
100

ATER VAPOR AND 0XY6

:n
Ijr

*~

rtr^' "=

ig

--M^f^-""
i

1
I117
'

" L-

- 0.2

^3i J< 1/
*

'ii'
,

'

I
'

fl

5^U"
**"
''

^=^^'ttf f
1

t\'\\.\ T4 ^3 1X1 /-ilmjJir J tjS

i,

Wwi^l^^'-'

Si
iji

?- 0.3

?ft

-f+'*f

-'l! ....

\H^* as
~

---WW w^MMm T'

i

li

^-g: :|

1 IP
o.c
1

|eo=oS
Ss
j

it BJJ4Uiif^ ^n^'j
,,,:

4++*"

jili-f'

Spp
e =

jiU-bi 31!^
n'i

;=;};;

i-f H! jg fff'

y ^ J<^ J
"^
lU*

w1T,I /v /rif
?
:

^;ri-.l. 'In lit

|fj/^:^

^S

/
s

'

''-'-

;~

-fffK

"^ 1
^

-^

liii;

ii|

J=t

,:,
ml li-'i

50
UJ

1 ^1-1-= pi 1 i iF^i^^- ""wf^ M ^ jgfi.'ffff ;;^o.03ppga.

tt
"

;;r4|t

fc
:|:atl

II
i

TT/Z

f|:

_L

Sr Hiiri-^-m'i

/M
i/^-i-iff

-i^ 5

t;

:;::;

ee;;e;;;

1 ii;^=^4
ij?r^
.J
;

= IllU^illlli

O a.
20

IT

fe:,;:.

1
';'
:;

30.05|:

::2

I Q <
LU

4'
''

>--[JOl|--

:,::;.

n
h

ff
yinttum

ir s*

;;

^J
'If
tt:

'-Mh
EiBili

r::::;

1
f|

1 jpy

---('f

S:
i

=^^*i
;

^ffft

38-"

"i

---S-

!|:|
f11
=railn
-:=a^-

i;^^=i|
1
-i5

J]

^|r
'''"
1!

+j
I t
;

r
i*^
;M:

4*1
'it

--;;;

?\
R
r^-J^--l-

10

= ^"-"4?^r%

Kl
'

'

L
"

1
iil

A T%
i.'ii

jf

Tr
i/T
1

f
'

ii^-i l^#'
;i

-a

'"I
-"""

7M^^' j^ "^Mfy
"i

fH

Tf

iiisao2i = "-fVf

'.i^i

'f

h:!

^^i^^3
.-:|i;^_=,i?r
-"
-H:

TT#

m
ifi;. ;:i;

''D

-Ij-

f PT

Ey^/V/TP

SSletiS!?'
^:^^^^;i^

O > X o cr o
a:

Mffitrfi^^ifi ^a >_Iol f^ii '

-!

lli%r-

i^^

1 iii';?^.*^^iHiiaiiii

^^w_^S^
.iii:^:
:

-^
fS -gl [R^
-

fe:;;;ii;i

ii^^
fiStt

KS

^
=
1 \

/j

:iHij^ Y.
..
:

?! zz-.-A'i : :
:

:rj;i

:;

Hiilt^::
:

t
::::

-M-

- ^_
::'

m-*iHjiffiil^
2

4--

....

-,-1

^
^-.^:
/[
/

wm
.'
>

'4.

fe:M0.5f^ifffig^
-iirilii K;i
Si"."'^-";

19 i i iiU^. 1
if?

::::+n "^ ^ ^2-iisi; L^ ^ is" J

EE EEEE::!:

::

,^
..

.1}:

]
.

E^ii;
...1.

ii

3^-

m
;
1

^
Si -:
-fti

E:

r^i- Ipijiiii i
::::::t::.::
+

;:

'31:i

;:;,:::::
;!:::::::.
"
:

.\ :::::^
:

_... ""J,

!'+

"

[^

=1

?*-: =i=3- ^=|E|-=^SiSii^pSi. ;"K-

"tJ^ =

pij ^s. ^^Il^|.

i

....,:!...] ...:!.

^^

^^ j^
\

i '^a!^^^: -^' -^f^^

TTI

T ^-^-i

-n:::

::s:: Tte-

:;

--.-|:l
t

-^T
,1
1

.,

1
1

jf

"^ib T'^'' i\\\ ij;;

~r

Hp

-,^
tfi -

^4"

J.=-

-.

=i^

3
i-TT-;
-;
J
-,

f:?rn

gs-r- 1:^ -\\

--;

ill

-:;j

jf'Sp^
;;|;^;

-i*n-

-irT^r^-TZ-r

"
";
"

=nm

IrH l^u

=5= '
0.5

^ ^^
=
~i~"r

T^ 'i\^'--i^V^^-;::_;

i-;-i-^f]

^M
'-''
\

;;

^!'

r
,

'

'-,<['[

".!.;

i^n^^i
!

:g lS:--;g=|

iivn'|----i 1

-']"]']'"-

i-^^

EEE;::;i;;;;::;fe^j

^^\r^

IliiUlii

::::I41I

i---n
0.2

^M
EE
:

Il
_::::;:::::::

|Lj:^":r:i|j^
;
1

Mi
..

" :

..
[

.|

-J

::::

:1

il TOf

.+^4^
ii"i" T T T T
""''""ill

isfc
4
.

t""'if -]

::

:::

.-..II
1^

atti:

,t

ni

:::::i;::::;:::i:::
0.2 0.5

,,

20

50

100

FREQUENCY
Figure

IN

GHz

3.7

3-13

RAINFALL

ABSORPTION COEFFICIENT

K vs

FREQUENCY

7=

KPr"*

RAINFALL

RATE

db/km, WHERE Rr IS THE IN MILLIMETERS/HOUR

ee|!:::|:|m|

i::i|:i|||

IP
1

IP'l
41

Hin:;;;;;;;;;

^^^--^fe

_
F.=)=!=' '

i"^.

'0
,1: -.:.:r.-

III
"'-

::i;D
iii|'i

_:,:
:

:Mi^:.-.^^

'::
, :

_^;.;

::,-- :glr^:.
:

::-.:
:

..

^i
3

:^:.:^,

^: ;:;

p-^
,

'

\-l

_l'._
i/'i

;...

-::

-I':

'jj

=^
'-:-

\-

P :.
Z
'

tU

in
i

^::.

:
1

;;
--

"

i^

ii

..;

-s

Zt

rf
ntltff+r-

in"'

="

"^

10

_2__L_L_^__|_;l^
~'

ib mm
liifc""'

ifflWl
1
-;:

^:j
::|: l'r

'TT

tnTiTP'

Mlilili'l't
:;-r;;Ei;;H j=^;:ri;-;
;:

---"

dm
--^

"''''

f--^w
^;hI::
; i

i--H;Htr:;iii vyy-v

^,

, ^

-^^i4:ixr:rr-;H;,sa;

^_

::

:";:-;:

^ii;

-i^s;;;;; -.-

--

-:

...

-:---

Mm
i i^
r

,-2 : ::""^"^"^ 10

f pi
^
1

-:

1
l''i

'T,n

;""'
' ' , ,

^yyr---:----

ir

-I
^?
:;:
-!I

zirr

---"^

::::::f
SI'
-i

T
:

i^L

-.f.::

rr.

= =!== E:-r=r=F
-

r:

::^^i.:;

:.;

;:-^:;-

--

'

2i

-rriri-

;:

^ 1

'

:rS

:-::i^.::|
. :

i;::::^:,-

,.::::::::/,.
.

;.,

: !

'i^

:!
'-

-::fii

'

i
1

_.__L
-

-^

-n

-,

:a

''---y-y^:---^--^^-:-:-

---

-i
-^::,:

'

rr-

TT"

|;:tfflSi
-^j/h
[:::

^^^^^^^=W+^#tt
fG-2)^2{fG-2)|x|Cr'*
l

Mi^r
Ij;

r:i:

K(7

3(

f
f
...
1

r---'i
i

-?,

~'"

=
,,..

14-00 ?'(fe-2)''''
!,
,

+ 0.085(fc-3.5)exp(-0.006fG
-i:
r
: :

[L
;^^'ij

10
,.
;
:

|::i

:-

,r

- ,::
'

''! !'t!|i':t-

;;"

liij

H:::.,
;

,:

"

;.
1

;:

is

11
::i;
.til
.

t
1
^.
1

"'''H
..

.;!

/
i

.-a
:

;.

-/

--

'

Hi

'~
1'.'

:::!*,
10

4IN

'!'
1

mrrfr^f-'r
nil

..-4

_:::

Ti:.

;,,,

nil

JJlmil

FREQUENCY

fg

GHz

Figure 3.8

3-14

> Z
> 2 X
T^i
II

35

:d

-n J> |
I

>
OD

r"

o
Q. CT

o d
<o

3]
:3D

m o -<

m
^ ^ n

H O

-n

(0

^ X m m
33

CD

m X Tl O m
Q
1^ -n 33

X
N

m
33

X X o m
33

m O c m z: o -<

3-15

EFFECTIVE DISTANCE

fg^

IN

KILOMETERS

\
^

V ^
^
i^^
\\

1; \ ^.
\\

m
-n -n

1 \\ % \
\
\

3) > -< -n

k\
>

m o < m
o

1 \ ^
y

>

-n
(O

c ^ a

__

1 ^. \ \ ^ ^\
S\\

CO

^
Ol

^.

7^

<)

rCJ

^ m H m
Ji

w w
\\^
\

^ ^

o o m o (^ o ^

CO

1
1
\

>
^.

i 1\
\v v^ \ \\ \
\ iW

>
DO

CO

O zi
T)

w
kJ
1

"32
II

d"

h
-J

O b
\ISJ

<?

o o

cuKj t) io bo

Op

3-16

EFFECTIVE DISTANCE

Tgr

IN

KILOMETERS

\
^

\
w!
\ ^

VV
\

m m o H < m o GO H >
o
v

"n

W
Iwi
\

^
1
1

t>
-<

w! \

-0

1 S^
\ S\^
cr>
II

n o

H X

t>

^
2
7^

^V \v

w
1

\
\ \\

O O Ui
*

2 m
C

o^ -^

n C) S m H

**

zo CO

iV \ \\ \ l\
1

o "H O o 71
FN")

M > ^ >
CD CO

'

:d "D

o
?
ii

11

o
\)

O o
|tn

P OJ 00

OJ

p
00

--.--P P o OJro o to 00

3-17

EFFECTIVE

DISTANCE

Tgr

IN

KILOMETERS

r-^

en

c=>

\ \k
'^

en

ro

ro en

l'

C3

f
\

cn

:c

cn

> -<
"D
'CD

^ 1 1 k 1
11

m m o < m o
CO

Wv \\\
^\
\

-\

>
cn cn

k
iv
ftW \
\\

>

n
<a

I
6^

<s

2
7J

CD
CI3

UI ro

o ^ m H m
3D
<y)

CD cn

o>

\
1

cn

>
CD CO

cr>

o H O
XI TJ

cn

CD cn

.s

cn
tl
II

io

p cn

iLn

ge

OJ ho

t)

o p 03
Cd

3-18

EFFECTIVE DISTANCE
<=>

Tgr

IN

KILOMETERS

ro

>

"

<->

<n

s
Q
-^

SO

-i^

0*

>

<^

C3

6
R
IT
ro

ro

V
^.

R
3

m "n m o H < m o
in -n

-P>

1
en

\ 1

-<
T)

<=>

X
31
10'
<D

no

1
\

1
\\\
1\

o m
V

no 03

uu \

ro

o
-J)

m m
3) CO

ro 00

m \\
'

>
I

to

ro

OJ 05

>
C/5

V l\ \v \
1

>

o
"D

CD CO

333333 N
ro ro

33:5
ro

ro

OJ OD

.7380

ro
.1494

ro
.2156

<i^

\\
c m

.9817

8090

8894
V/(0.4a)

e=>

km

km

km

km

km

km

J ro

05

.4^

cn
OJ
r\3

O p p 03
(X)

3-19

PATH AVERAGE RAINFALL RATE, Rf, VS EFFECTIVE RAINBEARING DISTANCE, (TOTAL ANNUAL RAINFALL, lOOcm)
100

rer

50

___
""*"*

"^
cc ID

\\ _

O X
oc

V V,

PERCENT OF THE YEAR THAT Rr IS EXCEEDED

UJ 0.
(/)

^
>

cc UJ 1-

,So7;r7
1

"

Jm
1%
tr.

~-~-J^

_i _i

< tr
bJ

<

> < g

UJ

Is 7cJ
0.5

0.2
5

"

10

20

50

100

200

1000

EFFECTIVE RAINBEARING DISTANCE,

fg^, IN

KILOMETERS

Figure 3.14

3-20

4.

DETERMINATION OF AN EFFECTIVE EARTH'S

of a

R.ADIUS
is

The bending

radio ray as

it

passes through the atmosphere

In

largely determined
to

by the gradient of the refractive index near the earth's surface.

order

represent radio

rays as straight lines, at least within the first kilometer above the surface, an "effective
earth's radius"
is

defined as a function of the refractivity gradient,

AN,

or of the surface

refr activity value

N
s

= (n
s

1)

X 10

(4. 1)

where n
s

is the

atmospheric refractive index

at the surface of the earth,

In the United States the following empirical relationship has

been established between

the

mean N
s

and the mean refractivity gradient

AN

in the first kilometer above the surface:

AN/km

-7.32 exp(0. 005577

(4.2)

Similar values have been established in West

In this

Germany and

in the

United Kingdom, [CCIR 1963

e]

paper values of

are used to characterize average atmospheric conditions

In the

during periods of

minimum

field strength.

nortnern temperate

zone, field strengths

and values

of

reach minimum

values during winter afternoons.

Throughout the world,

regional changes in expected values of transmission loss depend on

minimum monthly mean

to sea level:

values of a related quantity,

which represents surface refractivity reduced

N
where h
s

N so
=

exp(-0. 1057 h
s

(4.3)

is the

elevation of the surface above

is

mean

sea level, in kilometers, and the

1

refractivity

read from the

map shown

in figure 4.

and taken from Bean, Horn, and

Ozanich

I960]
of the refraction of a radio

Most
to

determine

most important
horizon paths,
terminal, and

OS
and h
to the

ray takes place at low elevations, so

it is

appropriate

for locations corresponding to the lowest elevation of the radio rays

of a

geometry

propagation path.

As

a practical matter for within-the-

is is

defined as the ground elevation immediately below the lower antenna

No

determined at the same location.

For beyond-the-horizon paths, h

s

and and

are determined at the radio horizons along the great circle path between the antennas,
is the
if

N
and

average

of the
is

two values calculated from

150 meters below

(4. 3).
its

An exception

to this latter

rule occurs

an antenna

more than

radio horizon; in such a case,

h
s

No

should be determined at the antenna location.

a,
is

The effective earth's radius,

given by the following expression:

a=a

0.04665 exp(0. 005577

)l"'^

(4,4)

4-1

where

is

the actual radius of the earth, and is taken to be 6370 kilometers.

a,

Figure 1.2

shows the effective earth's radius,

plotted versus

The

total

bending of a radio ray

which

is

elevated

more than 0.785 radians

atmosphere
is
is

(45) above the horizon and

which passes
of

all the

way through
satellite

the earth's

less than half a milliradian.

For studies

earth-

communication ray bending

important at low angles.

is

At higher angles

it

may

often

be neglected and the actual earth's radius

then used in geometrical calculations.

Large values
is

of

AN
of

and

are often associated with atmospheric ducting, which

usually important for part of the time over most paths, especially in maritime climates.

The average occurrence

strong layer reflections, superref raction, ducting, and other

is

focusing and defocusing effects of the atmosphere

variability functions to be discussed in section 10.

taken into account in the empirical time

Additional material on ducting will be
,

found in papers by Anderson and Gossard

1953a, b]

Bean
[

1959],

Booker
[

1946],

Booker

and Walkinshaw
and Belkina
[

1946],

Clemow and Bruce-Clayton

[

1963], Button
19 52]
,

1961], Fok, Vainshtein,

[

19 58],
[

Friend

1945]

Hay and Unwin

[

Ikegami

1959]
[

Kitchen, Joy,
,

and Richards

1958],

Nomura and Takaku

[

1955]

Onoe and Nishikori

1957]

Pekeris

1947]

Schunemann

1957], and Unwin

1953]

4-2

4-3

EFFECTIVE
11600

EARTH'S

RADIUS, Q,

VERSUS

SURFACE REFRACTIVITY, Ng

11200

<
UJ

9200

8400

7600

260

280

300

320

340

360

380

400

N.
Figure
4.2

4-4

5.

TRANSMISSION LOSS PREDICTION METHODS FOR WITHIN- THE-HORIZON PATHS

Ground wave propagation over
a

smooth spherical earth

of

uniform ground conductivity

and dielectric constant, and with a

homogeneous atmosphere, has been studied extensively.

in

Some
[1959,

of the results

were presented

CCIR Atlases
[

19 55,

19 59]

Recent work by Bachynski

of

I960,

1963], Wait [1963], Furutsu

1963]

and others considers irregularities

electrical ground constants and of terrain.

of terrain

distinction is

made here between

the roughness
of radio

which determines the proportion between specular and diffuse reflection

is

waves, and large scale irregularities whose average effect

line or

accounted for by

fitting a straight

curve

to the terrain.

A
surfaces
is

comprehensive discussion
given in a recent book by

of the scattering of

electromagnetic waves from rough

[

Beckmann and Spizzichino

diffraction,
[

1963]

Studies of reflection
hills,
[

from irregular terrain as well as absorption,

and scattering by trees,

[

and

man-made
1958],

obstacles have been

made by Beckmann
[

1957], Biot

1957a, b]
[

Kalinin
,

1957,

Kuhn[l958], McGavin and Maloney

[

19 59]

McPetrie and Ford

[

1946]

McPetrie

and Saxton
.workers.

1942], Saxton and Lane

1955],

Sherwood and Ginzton

1955], and

many
in

other

Exan:iples of studies of reflection

[

from an ocean surface may be found

[

papers by

Beard, Katz and Spetner

If

1956], and

Beard

1961]

two antennas are intervisible over the effective earth defined

in section 4,

geometric

optics is ordinarily used to estimate the attenuation

relative to free space, provided that

the great circle path terrain visible to both antennas will support a substantial

amount

of

reflection and that

it is

reasonable to

fit

a straight line or a convex curve of radius a to this

portion of the terrain.

Reflections from hillsides or obstacles off the great circle path between

two antennas sometimes contribute a significant amount to the received signal.

against such off-path reflections

Discrimination

may reduce

multipath fading problems, or in other cases

antenna beams

may

be directed away from the great circle path in order to increase the signal

level by taking advantage of off- path reflection or knife-edge diffraction.

For short periods

of

time, over

some

paths, atmospheric focusing or defocusing will lead to

somewhat smaller or

much

greater values of line-of- sight attenuation than the long-term median values predicted

for the average path by the

5.
1

methods

of this section.

Line-of-Sight Propagation Over a Smooth or Uniformly Rough Spherical Earth

field at a receiving

The simplest ray optics formulas assume that the

antenna
r
,

is

made

up to two components, one associated with a direct ray having a path length

and the other

^
.

associated with a ray reflected from a point on the surface, with equal grazing angles
reflected ray has a path length
differs
r^ + r^.

The
ray

The

field arriving at the receiver via the direct

from

the field arriving via the reflected ray by a phase angle

which

is

a function of the

5-1

path length difference,

is

Ar

= r

+ r

illustrated in figure

5. 1.

The reflected ray fieM

(tt

also modified by an effective reflection coefficient

and associated phase lag

c),

which depend on the conductivity, permittivity, roughness, and curvature

of the reflecting

surface, as well as upon the ratio of the products of antenna gain patterns in the directions of

direct and reflected ray paths.

Let
the other,

^oi

and

g
02

represent the directive

to be

gain for each antenna in the direction of

assuming antenna polarizations

matched.

Similar factors

g
ri

and

g
r2

are

defined for each antenna in the direction of the point of ground reflection.
reflection coefficient

The effective

is

then
y^

e
1-0

cr,

= e

DR
01 02

exp

sin

d;\

(5.1)

where

the divergence factor

allows for the divergence of energy reflected from a curved

surface, and

may

be approximated as

2d d

%
2

a d tan

(5.2)
ij^

An expression
R

for the divergence factor,

D, based on geometric optics was derived by Riblet

of the theoretical coefficient, of a

and Barker, [1948].

exp[-i(TT -c)],

The term R represents the magnitude

for reflection of a plane

In

wave from
c

smooth plane surface

given conducis

tivity

and dielectric constant.

most cases

may

be set equal to zero and R

very nearly
III.

unity.

A
of

notable exception for vertical polarization over sea water is discussed in annex

Values

R and

vs

ij;

are shown on figures

III. 1

to

III. 8

for both vertical and horizontal

polarization over good, average, and poor ground, and over sea water.

The grazing angle

on figure
5. 1.

ijj

and the other geometrical parameters

a
,

d, 5.

and a are shown

The terrain roughness factor,

X.,

defined in section

1.2,

and the radio

i^

wave

length,

are expressed in the same units.

The exponent

(cr

sin

ij")/)^

Rayleigh's

criterion of roughness.
If

the product

DR

exp(-0.6

cr,

sini]j/\)

is

less than

*/sln

4i ,

and

is

less than

0.5,

groimd reflection

may

be assumed to be entirely diffuse and

is

then expressed as

%
R
ri
oi

r2
sinLJJ

(5.3)

02

where terrain factors

D,

R and

cr

are ignored.

The factor
to

r2

g
02

in (5.

3)

makes

01

approach zero when narrow-beam antennas are used

discriminate against ground reflections.

5-2

For a single ground

tained

reflection,

the attenuation relative to free space

may

be ob-

from

the general

formula
r

A-G =-10

log -^g g ^oi ^02

/ZirAr ^ cosl 1+R 2 -2R e \ \

r

db

(5.4)

where

the path antenna gain

1

may

not be equal to the

sum

of the

maximum

antenna gains.

At frequencies above
in

GHz

an estimate of losses due to atmospheric absorption as shown

(3.4)

should be added to

as computed by

(5.4)

or

(5.5).
1

Over a smooth

perfectly-conducting surface,

and
g
),

c = 0,

Ascuming also

that

free space antenna gains are realized, so that

to free space is

10 log(g

the attenuation relative

A
Exact formulas for computing
given in (5.9) to
(5.4)
(5. 13)

= - 6 - 10 log sin

(tr

Ar/X)

db

(5. 5)

Ar

are given in annex

III.

The appropriate approximations

If

suffice for

most practical applications.

methods

Ar

is less

than 0. 12X

may underestimate
Section
5.1.1

the attenuation and one of the

of section 5.2

should be used.

shows exact formulas for antenna heights h

and

above a plane
4"

earth, or above a plane tangent to the earth at the point of reflection.

is

The grazing angle

then defined by
tanip = hj/d^ =

h^/d^
d

(5.6)

where heights

ajid

distances are in kilometers and

and

d-

are distances from each

antenna to the point of specular reflection:

d^+d^^d,
The distances
heights heights

d^ = d(l +h^/h'^)"-^

d^ = d(l +h'^/h^)"-^

(5.7)

d h

and

may

be approximated for a spherical earth by substituting antenna

h
'

and

above the earth for the heights

and

h'

in

(5. 7).

Then these

may

be calculated as
h'^

h^

- d^^/(2a)

h^

h^

d^ l{2a.)

(5. 8)

ior an earth of effective radius

a,

and substituted in (5.7)

to obtain

improved estimates

of

and

Iterating between

(5.7)

and (5.8),

any desired degree of accuracy

may

be

obtained.

5-3

The path length difference between direct and groiond reflected rays

is

Ar

= ^^ d^

+ (h; + h^)^

\|

dS (h;

h'J^ -

2 h- h-

/d

(5. 9)

where

the approximation in

(5=9)

is

valid for small grazing angles. d

,

Referring to

(5. 5)

the greatest distance,

for

which

is

zero,

(assuming that
(5,9)

=1 and that free space gains are realized)

therefore:

occurs when

Ar

= \/6.

From

Ar S 2h'h'/d;

12 h'
1

hl/X
Z

(5, 10a)

This equation
solving
(5. 8)

may
for

be solved graphically, or by iteration, choosing a series of values for

h
,

d
o

h'

and testing the equality in

of

(5, 10a).

For the special case

distance
d
o

equal antenna heights over a spherical earth of radius

a,

the

may
Ar

be obtained as follows:

= \/6 =

^- [h

dio
-

d ^/(8a)l

2h^/d

ooo
-

hd

/(2a) + d ^/(32a^)

(5, 10b) '

*

where
d, 3 d^ = d/2,

h, = h- = h

and

h' =

d ^/(8a)

For
the angle
ij^

this special case

where

p h

over a smooth spherical earth of radius

a,

may

be defined as

tan

qj

2h/d

d/(4a)

(5. 11a)

and

Ar

d(sec

iJj

- 1) =

+ tan

i|j

- 1

(5.11b)

Let

9,

represent the angle of elevation

h
,

of the direct

ray

zontal at the lower antenna,

assume

that

< < h

< < 9a

ijj

relative to the hori2, and that the / 2,

a,

grazing angle,

i|'>

is

small;

then, over a spherical earth of effective radius

Ar ^ 2h

sin^J

2
'

h^j
L

,,

+4h/(3a)

+ e

(5.12)

whether

is

positive or negative.

For

=0,

'i-,

= 2h

/(34j),

5-4

Two very

useful approximations for

Ar

are

Ar

^[1

d d /d

^ 2h

sin

\\i

kilometers

(5. 13)

and the corresponding expressions for the path length difference

in electrical

in electrical radians

and

degrees are

ZirAr/X = 41,917

h'h'/d

41.917

f L^^

dd./d ^

42

h sin

4;

radians

(5.14a)

360Ar/\

2401.7

f h'Ji'

/d

2401.7

^^

^I'^J^ -

^"^^^ ^ ^i ^^^"^

degrees

(5.

14b)

where

is the

radio frequency in

MHz,

and

all

heights and distances are in kilometers.

if

The last approximation in (5.13) should be used only

as
it

h
1

is

small and less than h /20

^

involves neglecting

/(2a)
(5. 5),

relative to

in

(5.8)

and assuming that

=
15 db

d.

As noted following
Ar > 0.06
X.

ray optics formulas are limited to grazing angles such that

With

this criterion,

and assuming

= 1,

the attenuation

is

for

the corresponding

minimum

grazing angle

4j

mS

^y

0.03 \ d/(d,d7)~" radians

1

where antennas are barely intervisible.

A comparison
= 0)

with the

CCIR

Atlas of smooth-earth

diffraction curves shows that the attenuation relative to free space varies

from

10

to

20

decibels for a zero angular distance

(9 = 0,

il'

except for extremely low antennas.

of actual radius

Figure

5.

la shows
5.

how rays

will

bend above an earth

a
o

=63 70

kilo-

meters, while figure

fective radius

lb shows the

same rays drawn

level,
,

as straight lines above an earth of ef-

Antenna heights above sea

h'

and h
5. 1. 1.

are usually slightly greater

This difference arises from

than the effective antenna heights

and

h'

defined in

two circumstances:

the

smooth curve may be a curve-fit

to the

terrain instead of representing

sea level, and straight rays above an effective earth overestimate the ray bending at high elevations.

This latter correction

is

insignificant unless

is large.

5. 1.

A Curve -Fit

to

Terrain
It is

smooth curve
h'^

is fitted to

terrain visible from both antennas.

used

to define an-

tenna heights

and ray

h^,

as well as to determine a single reflection point

where

the angle

of incidence of a

r^

is

equal to the angle of reflection of a ray

tr
,

in figure 5,1.

This

curve
in

is

also required to obtain the deviation,

of terrain heights
h'

used in computing
0.16 \
for the

(5,1).

Experience has shown that both

h'

and

should exceed

following formulas to be applicable.

For other prediction methods, see subsection 5.2.

5-5

First, a straight line is fitted by least squares to equidistant heights

h.(x.)

above

'

sea level,

and

x. /(2a)

is

then subtracted to allow for the sea level curvature

line

1/a
fitted to

illustrated in figure

6.4.

The following equation describes a straight

h.(x.)

h(x)

21 equidistant values of
the transmitting antenma.
to either antenna

for terrain

between
and

x. =

and

x. =

kilometers from

The points
not visible

are chosen to exclude terrain adjacent

which

is

from

the other:

h(x) = h +

m(x

- x)

(5. 15a)

20 20
17

h.(i-lO)

lyK..

;=^^^. r.-^S^

(5.15b,

Smooth modified terrain values given by

y(x) = h(x) - x^/(2a)
(5. 16)

will then define a curve of radius

which

is

extrapolated to include all values of

from

to

= d,

the positions of the antennas.

of the

The heights

antennas above this curve are

h^ = h^^ - h(d)

h'^ -

h^^ - h{0)

(5. 17)

If

h'

or

h'

is

greater than one kilometer, a correction term,

is

Ah,

defined by

(6. 12)

and shown on figure 6.7

used to reduce the value given by

it

(5.17).

Where terrain
single curve,
it is
o-

is

so irregular that

cannot be reasonably well approximated by a

is

is

large and

= 0,

not because the terrain

very rough, but because

usefiil.

irregular.

In such a situation.method 3 of section 5.2

may

be

5.1.2

The terrain roughness factor,

(t,

(r

The terrain roughness factor

modified terrain elevations,
y.,

in

(5.1)

is the

root-mean-square deviation of
(5..

relative to the

smooth curve defined by

16),

within the

limits of the first Fresnel zone in the horizontal reflecting plane.

The outline

of a first

Fresnel zone ellipse

is

determined by the condition that

^ll+^21
where
r

"l+"2+'^/2
from a point on
the

+ r

is the

length of a ray path corresponding to reflection

r

edge of the Fresnel zone, and

+ r

is the

length of the reflected ray for which angles of

[

incidence and reflection are equal.

Norton and Omberg

1947

give general formulas for

determining a first Fresnel zone ellipse in the reflecting plane.

Formulas are given

in

annex

III

a b the first Fresnel ellipse cuts the great circle plane.

for calculating distances

and

x,

from

the transmitter to the two points

where

A sample

calculation of

o"

is

given in Example
5-6

of

Annex

III.

5. 2

Line -of -Sight Propagation over Irregular and Cluttered Terrain

optics formulas described in section
5.
1

Where ray

are not applicable, a satisfactory-

estimate of'line- of- sight transmission loss

may sometimes

be

made by one

of the following

methods:
1.

If

a slight change in the position of either antenna results in a situation

where ray

optics formulas

may

be used, then
9 =

A may
is
fit to

be estimated by extrapolation or interpolation.

useful set of calculations for

2.

given by

Domb

and Pryce
5. 1,

1947]

Instead of a single curve

fits

terrain as in

in

some cases

the

method may

be extended to multiple curve

3.
If

and multiple reflections from these curves.

it

terrain is so irregular

cannot be reasonably well approximated by a single

7

curve, the line-of-sight knife-edge formulas of section

4.

may

be applicable.

Interpolation between curves in an atlas,

I,

or standard propagation curves such as

those given in appendix

5.

may

provide a satisfactory estimate.

of interest

Empirical curves drawn through data appropriate for the problem

may

be useful.

For example, the dashed curves

of figures 1,1

-1.3 show how

values of attenuation relative to free space vary with distance and frequency for a large

sample
ures
I. 1

of

recordings of television signals over random paths.

to a

The data shown

in fig-

-1.4 correspond

more

careful selection of receiving locations and to a

greater variety of terrain and climatic conditions.

The effects of refraction, diffraction, and absorption by trees,

obstacles are often important, especially
if

hills,

and man-made
is

a receiving installation is

low or

surrounded

by obstacles.

Absorption

of radio

energy

is

probably the least important of these three

is

factors except in cases where the only path for radio energy

directly through

some

build-

ing material or where a radio path extends for a long distance through trees.

Studies

made

at 3000

MHz

indicate that stone buildings and groups of trees so dense

that the sky cannot be seen through

them should be regarded as opaque

At 3000

objects aroiond which

diffraction takes place

[McPetrie and Ford, 1946].

MHz

the loss through a 23-

centimeter thick dry brick wall was 12 db and increased to 46 db when the wall was thoroughly soaked with water.
a wet one were usual values.

loss of 1.5 db through a dry sash window, and 3 db loss through

The only objects encountered which showed a loss

of less than 10 db at

3000

MHz
huts.

were

thin screens of leafless branches, the tr\ink of a single tree at a distance exceeding

30 meters,

wood-framed windows,

tile

or slate roofs, and the sides of light

wooden

Field strengths obtained

when a
6

thick belt of leafless trees is between transmitter and

receiver are within about

db of those computed assuming Fresnel diffraction over an

obstacle slightly lower than the trees.

Loss through a

thin

screen of small trees will rarely

If

exceed

6 db if the transmitting
is the

antenna can be seen through their trunks.

greatest expected loss.

sky can be seen

through the trees, 15 db

5-7

The following empirical relationship for the rate

given by Saxton and Lane
[

of attenuation in

woods has been

19 55

A^=
where
f

d(0.244 log

f -

0.442)

decibels

(f

> 100 MHz)

(5. 18)

A,^

is the

absorption in decibels through

meters

of trees in full leaf at a

frequency

megahertz.

The situation with a high and a low antenna in which the low antenna
distance

is

located a small

from and

at a

lower height than a thick stand of trees

is quite

different

from

the

situation in which both antennas

may

be located in the woods.

Recent studies at approx-

imately

500

MHz show
],

the depression of signal strengths

below smooth earth values as a

fxinction of clearing depth, defined as the distance

from

the lower antenna to the edge of the

woods

[Head, I960

Expressing

this

empirical relation in terms of a formula:

A^=52-12 1ogd c
'^

decibels

(5.19)

where

A^,

is the

depression

of the field strength level

below smooth earth values and d

is

the clearing depth in meters.

particularly interesting application of

some

of the in the

smooth-earth formulas given in

design of space-diversity configuraDiffraction theory

this section is the

work

of

Lewin [1962] and others

tions to

overcome phase interference fading over

line- of- sight paths.

may

be used to establish an optimum antenna height for protection against long-term power fading,

choosing for instance the

minimum
Then

height at which the attenuation below free space

the

is

20 db

for a horizontally uniform

atmosphere with
the

maximum

positive gradient of refractivity

expected to be encountered.

formulas

of this section will

determine the optimum

20

diversity spacing required to provide for at least one path a similar

db protection against
of

multipath from direct and ground- reflected components throughout the entire range
tivity gradients expected.
In general, the refractive

refrac-

index gradient will vary over wider

ranges on over-water paths [Ikegami, 1964].

GEOMETRY

FOR WITHIN -THE-HORIZON PATHS

Figure

5.1

5-9

6.

DETERMINATION OF ANGULAR DISTANCE FOR TRANSHORIZON PATHS

9,

The angular distance,

is the

angle between radio horizon rays in the great circle

is

plane defined by the antenna locations.

as well as in forward scatter theory.

This important parameter

used in diffraction theory

Angvilar distance depends upon the terrain profile, as

illustrated in figure 6.1, and upon the bending of radio rays in the atmosphere.

Figure 6.1

assumes a linear dependence on height

of the
If

atmospheric refractive index,

n,

which implies

a nearly constant rate of ray refraction.

heights to be considered are less than one kiloa,

meter above

the

earth's surface, the assumption of a constant effective earth's radius,

for ray bending.

makes an adequate allowance

Atmospheric refractivity

N
to

= (n-

1)

10

more than one kilometer above

tially with height

the earth's surface, however, is

].

assumed

decay exponen-

[Bean and Thayer, 1959

This requires corrections to the effective earth's

radius formulas, as indicated in the following subsections.

To calculate
horizons.

6,

one must first plot the great circle path and determine the radio

6.1

Plotting a Great Circle Path

For distances less than 70 kilometers, the great circle path can be approximated by
a

rhumb

line,

which

is a line

intersecting all meridians at the

same

angle.

For greater disis

tances, the organization of a

first plotted on an index

map
to

study

is illustrated

on figure 6.2.

Here, a rhumb line

map

show

the boundaries of available detailed topographic sheets.

Segments

of the actual great circle path are later plotted on these detailed

maps.
is

The spherical triangle used for the computation of points on a great circle path

shown on figure
inals,

6.3,

where

PAB

is

a spherical triangle, with

and

the antenna

term-

and

the north or south pole.

B has

a greater latitude than

A, and

is in the

same hemisphere.

The triangle shown

is for the

northern hemisphere but

is

may

readily be

inverted to apply to the southern hemisphere.

B'

any point along the great circle path

from

to

B,

and the triangle

PAB'
and $

is the
,

one actually solved.

The latitudes of the

points are denoted by

while

and

are the differences in longiZ'

tude between

and

B and A and

B', respectively.

and

are the corresponding

great circle path lengths.

The following formiilas are practical for hand computations as well

Equations
[l.

as for automatic digital computers.

(6.1)
,

to
],

(6.4)

have been taken, in this form,

from a well-known reference book

The
initial

T. and T.

1956

where they appear on pages 730-739.

bearing (X from terminal A,

and Y from terminal B) are measured

from true

north, and are calculated as follows:

tan

cot

(6.1)

6-1

tan

Y + X

cot

$_
r

[{''

A\
Y-X

(6.2)

Y+X^Y-X +
5
5

^, Y

and

jY+X

(6.3)

The great circle distance,

Z,

is

given by

tan

S
^

tan

Y + X

Y-X

(6.4)

To convert the angle

following
is

obtained in degrees from (6,4) to

level earth's radius of

ixnits of

length, the

used, based on a

mean sea
d,

6370

km:
(6.5)

km

111.18 Z

The following formulas show how

point on the great circle path,

to calculate either the latitude or the longitude of

when

the other coordinate is given.

to

The given coordinates

7.5

correspond to the edges of detailed maps, and

intermediate points usually about

minutes apart, so that straight lines between points will adequately approximate a great
circle path.

For predominantly east-west paths, calculate the latitude

difference

for a given longitude

C:
cos Y' = sin

X sinC
= sin

sin $

A
.

cos

cos

(6.6)

cos

^-r,,

cos $

/sin Y'

(6.7)

For predominantly north-south paths, calculate the longitude difference

given latitude

for a

sin Y' = sin

cos $

/cos

(6.8)

cot

C -

tan

Y' -

X ^

cos

B'

B'

(6.9)

Where

the bearing of a path is close to 45 degrees, either nciethod

may

be used.

6-2

6. 2

Plotting a Terrain Profile and Determining the Location of Radio Horizon Obstacles

This subsection explains

how

to

determine the sea level arc distance,

h
J-jt,

d^ Lt, r

from an

antenna to
level.

its

radio horizon obstacle, and the height,

of this obstacle
(d
Xjt
, )

above
j

mean sea
)

The horizon obstacles are represented by the points

the great circle plane containing the antennas.

These points

may

h and (d h in Lr Lt Lr be determined by the tops of

itself.

high buildings, woods, or hills, or

may

be entirely determined by the bulge of the earth

All of the predictions of this paper replace the earth by a cylinder

whose elements are pergeneral irregular and

pendicular to the great circle plane and whose cross -section

is in

determined by the antenna and horizon locations

in elevations of antenna
to

in the great circle plauae.

When

the difference

and horizon greatly exceeds one kilometer, ray tracing

of radio

is

necessary

determine the location

Elevations
h.

horizons accurately [Bean and Thayer, 1959

].

of the terrain are

read from topographic maps and tabulated versus

their distances

x.

from

the transmitting antenna.

The recorded elevations

The terrain profile
of the
is

shoiiLd include

those of successive high and low points along the path.

plotted on linear

graph paper by modifying the terrain elevations to include the effect

of the radio ray path
at a distance
x.

average curvature
of

and of the earth's surface.

The modified elevation

y.

any point

h.

from

the transmitter along a great circle path is its height above a plane

which

is

horizontal at the transmitting antenna location:

2 y. = h. - X. /(2a)

111

(6. 10)

where

the effective earth's radius,

of

a,
.

in kilometers is calculated using

(4.4), or is

read
(4. 3),

from figure 4.2 as a fvmction

where

The surface refractivity,

is

obtained from

N is estimated from the map on figure 4.1, A plot of y. versus x. on linear graph paper

is the

desired terrain profile.

the

Figure

6.4

shows the profile for a line-of-sight path.

The solid curve near

(h =

bottom

of the figure

indicates the shape of a surface of constant elevation

km).

Profiles for a path with one

horizon
6.5

common
6.6.

to both antennas

and for a path with two radio horizons are shown in figures
of these three figures are

and

The vertical scales

exaggerated in order to provide a

Plotting terrain elevations

s\afficiently detailed

representation of terrain irregularities.

vertically instead of radially

from

the earth's center leads to negligible errors

where vertical

changes are small relative to distances along the profile.

On a cartesian plot
ray from each antenna
horizon elevations
is

of

y.

versus

x.,

as illustrated in figures

6,4,

6,5,

and

6,6, the

to its horizon is a straight line,

provided the difference in antenna and

to be followed

less than one kilometer.

Procedures

where

this is not the

case are indicated in the next subsection.

6-3

6. 3
If

Calculation of Effective Antenna Heights for Transhorizon Paths

is

an antenna

located on another structure, or on a steep

cliff,

cliff

or mountainside, the

height of this structure,

or movintain above the surrounding terrain should be included

as part of the antenna height.

To obtain
h

the effective height of the transmitting antenna, the

average height above sea level

mitter and
31 evenly
its

of the central 80

per cent of the terrain between the trans-

horizon

is

determined.
h

The following formula

.

may

be used to compute

for

spaced terrain elevations

for

= 0,

1,

2,

.... 30,

where
h
t30
=

h h

is the
:

height

above sea level of the ^ ground below the transmitting ^ antenna, and,
27

Lrt

h
t

TT-r

25

, '

h = h
t

h
t

for

h <
t

Z7

ti

ts

h ^
to

(6. '
\

Ua) -i

i=3

otherwise

h
t

=h ts

h to

(6.11b)

where
is

is the

height of the transmitting antenna above

mean sea

level.

The height h

similarly defined.
If

h
t

or

as defined above
r

is

less than one kilometer,

h
te

= h
t

or
6.7,

re
is

r to

For antennas higher than one kilometer, a correction

reduce
h
t

Ah, read from figure

used

or

h
r

to the value

te

or

re

h
te

h
t

Ah{h
t

Ns

re

h
r

Ah(h

Ns

(6. 12) '

*

The correction

Ah was obtained by ray tracing methods described by Bean and Thayer

h

1959

].

For a given effective earth's radius, the effective antenna height

given horizon distance d
1-jt

corresponding
h
t
.

to a

is

smaller than the actual antenna height,

and
h <
1

Over a smooth

spherical earth with

<

km

km,

the following

approximate relationship

exists between effective antenna heights and horizon distances:

\e
If

<t/('^)

'

\e
is

4r/(^^)
from

^'' '^^^

the straight line distance

d,

between antennas

substantially different

the sea level

arc distance

as in communication between an earth terminal and a satellite, the effective

antenna heights must satisfy the exact relation:

h
t6

a[sec(d
XjU

/a)-

1]

TQ

= a[

sec(d
jT

/a) -

1 ]

(6.13b)

6-4

6.4 Calciilation of the Angular Distance,

The angular distance,

and
is the

9,

is

the angle between horizon rays in the great circle plane,

minimum

diffraction angle or scattering angle unless antejana

beams are
in

elevated.

Calculations for cases where the antenna

beams are elevated are given

annex
9

III.

In calculating the angular distance, one first calculates the angles

et

and

er

by
as

which horizon rays are elevated or depressed relative

to the horizontal at

each antenna,

shown on figure

6.1.

In this report, all heights and distances are

measured
d9

in kilometers,
is less

and angles are in radians unless otherwise specified.

When

the product

than

2,

9=9 oo
where
"a"
in
(6. 14)

d/a +

0+9 er et

(6. 14)

is the effective

earth's radius defined in section 4.

The horizon ray

elevation angles

and

may

be measured with surveying instruments in the field,

6.5

or determined directly

from

a terrain profile plot such as that of figure

or

6.6, but

are

usually computed using the following equations:

h^
9

h
_

=
et

-i^iJi
d^

Lt

-^ 2a

d,

=_i;L^_ er d,
Lr
h
ts
,

h^

d^

LL

2a

(6.15)

where
all

h^

Lt

h^

Lr
is

are heights of horizon obstacles, and "

level.

above

mean sea

horizon obstacle

or (h of a id. Lr Lt Lr Lt determined from the terrain profile by using (6.15) to test all possible
(h
) )

As

a general riole, the location

rs d ,

are antenna elevations,

horizon locations.
er nearest zero.
et
^

The correct horizon point

is

the one for which the horizon elevation angle

or

is

a maximtim.

When

the trial values are negative, the

maximum

is the

value

For a smooth earth,

9
et,

er

-/Th
te,

re

7a

for

te,re

<

km

At the horizon location, the angular elevation

greater than the horizon elevation angle ^ ^
or
et

of a
:

horizon ray,

9
ot

or

or

is

er

ot

= 9

et

+ dT^/aLt

nr

=0 er

+ d

Lr

/a

(6.16) *
'

If

the earth is smooth,

9 ot

and

or

are zero, and

= D /a s

where

s
Figure "
6,8, valid only ^ for
9

= ^ -

^Lt

^Lr
^
9

(^- 1^)

ot

+9 or

= 0,
'

is a

graph of ^

-versus

Ds

for various values of

surface refractivity,

6-5

In the general case of irregular terrain, the angles

61

and

(3

shown

in figure

are calculated using the following formulas:

J_
2a

h
+
9

h
(6.18a)

oo

+
et

-il_^f
d

=^+e GO 2a
8

h
er

h
(6.18b)

-ZlJ^
d

These angles are positive for beyond-horizon paths.

refractivity gradient,
the angles

To allow

for the effects of a non-linear

a
p

oo

and

oo

are modified by corrections

Aa

and

AS "^o

to give

and

whose sura
s;

is the

angular distance,

9,

and whose ratio defines a

path

asymmetry

factor

= a

oo

+ Aor

(6. 19a)

= P + P '^o oo
'

Ap
'^o

I9b) (6. ^ '

e = a

+ p

s =

Of

'^O

/p 0*^0

19c) (6. \ I '

The corrections
and of the distances
rays.

Aa

A 0*^0
and
d
6

are ftinctions of the angles

to the

ot

and

or

(6. 16), '

d
St

and

sr

from each horizon obstacle

crossover of horizon

These distances are approximated as

d
st

=dp oo /9 oo -d Lt
is the

sr

= dor

oo

/9

oo

-d Lr

(6.20)
'

'

The Slim of
(6.17).

sr Over a smooth earth

St

and

distance

s
.

between horizon obstacles, defined by '

st

=d sr

D s /2

Figure
Similarly, '

6.9,
is

drawn

for

Ns

= 301,

shows

Aa

as a function of

and
ot

st

Ap '^o

read from the figure as a function of "

other than

301, the values as read

from

8 and d . or sr figure are the multiplied by

For values of

C(N

Aa (N

OS

C(N
s

Aa

(301)

Ap (N
o

C(N

so
)

Ap

(301)

(6.21a)
'

(N^) = (1,3

N^

60 N^)

X 10"^

(6.21b)

For instance,
shows

C(250) = 0.66,
plotted versus

C(301) = 1.0,

C(350) =

1. 38,

and

C(400) = l,84.

Figure

6. 10

C(N
s

N
s

6-6

'

'

For small
^^^^

no correction

\a

or^p

is

required for values of d

..

100

Km. When

both

\a

and

^p
e = 9

St, r

less than

are negligible:
=

oo

oo

+ S
'

oo

(6. 22) '

which

is the

same as

(6, 14).

If

or
ot

or
d'

is

negative, compute

= d
st

I '

a 9
ot

St

or

d'

sr

= d

sr

|a9
'

or

(6.23)

substitute
If

d'

for
9

d
st

or or
9

d'

st

sr
is
9

for

sr'

and read figure

6.9, using

9
ot, r

=0.
radian,

either

ot

or
for

greater than 0.1

radian and less than

0.9

determine

Aa o

or

=0,1
ot

radian and add the additional correction term

Ns
The bending
the

(9.97

cot 9

ot, r

[ -

exp(0.05 d ^

st, r'

X lO"

radians

of radio rays elevated

more than
less than

0.9

radian above the horizon and passing all

way through

the

atmosphere

is

0.0004 radian, and

may

be neglected.

Other geometrical parameters required for the calculation of expected transmission

loss are defined in the sections

where they are used.

s

Many
where
a

of the p

graphs in this and subsequent sections assume that

are defined by
(6. 19a)

a fp

<

1,

and

and

(6, 19b).

It is

therefore convenient, since

the transmission loss is independent of the actual direction of transmission, to denote as the

transmitting antenna whichever antenna will

natively,
in
s

make

less than or equal to unity.

t

Alter-

may

be replaced by

1/s

and the subscripts

and

may

be interchanged

some

of the

formulas and graphs, as noted in later sections.

6-7

PATH

GEOMETRY

SCATTER

VOLUME

DISTANCES ARE MEASURED GREAT CIRCLE ARC.

Ds

IN

KILOMETERS

ALONG

_d_

Figure

6.1

6-8

> O> H OS X H d m H <^ i> O O Ol I m uu _?_ < m UJ. ,n-^^ S !^ o 00 (jJOJ. _| > Nl
l- rl\l
'

j:=.

III

(>J

o
c

O> H H Z ^ 33 > OJ O ^ 00 ^ CO ^ H S: f^.Soo H m m iji ro JO r"^

1- |3>
isi
(-kl
,

ID

rvi

H ^S m
=

-^ IM

6-9

(POLE CORRESPONDING TO HEMISPHERE OF B)

SPHERICAL TRIANGLE FOR GREAT CIRCLE PATH COMPUTATIONS

Figure 6.3

6-10

MODIFIED TERRAIN PROFILE FOR A

LINE-OF-SIGHTPATH

20

30

40

DISTANCE, X.IN KILOMETERS

Figure 6.4

6-11

MODIFIED TERRAIN ELEVATION,

IN

y,

KILOMETERS

o o
31

m o m
>
CO

H > o m
-z.

o
m
CM

n
C
0) Ol

^
:^

o
:^3

Tl

o m H m
C/)

o o
1

IT
U. n
r-

OL
II

O.

T\

+
n -n
Q.

N 2
U
J>

6-12

MODIFIED TERRAIN ELEVATION

IN

KILOMETERS

o
m o m
>
CO

H o J> 2 O m
-p

o
m
Tl

-r\

X
-~~

to

c
CD

-z.

CT> <=>

01

7^

o
o o

O)

o m H m
:d CO

OO C3

cz DD

m X o N o
I

Tl

6-13


REDUCTION OF ANTENNA HEIGHT FOR VERY HIGH ANTENNAS

1000

500

// /. ^ /'/
'

/ / /// r /' / / / / / /
,
'

//

/
/
200

100

/
/

///
/
'

/ / '/
/

/ V/
/

//

/
,

50

n
^

.^/^/
20

/ / / // / / //

/ ?r / ./
/ / /

/
/
/'

/
/

VV
/

YP //

^v
f

CO

o
i

/ '/ / // '/ '/ /, v \

/ /, /, /

/ / \/

//

f
y

'

/ /

'

J
/

y
1

05

/
/
/

// '/ /
f

/ /
/

h,5

h,-Ah

/ //

/
/ /

0.2

/ />
0.1

// V
f
'

^^\\
^>..
a
\

/
0.05

/
/

/A 7/ /
/

//

1
/

/a

0.02

1
1

\i
0.01
I

7
/

//
/
/

1 1
,

II

/
1

0.005

/'/
I

1
0.002
1

/
r

10

20

50

100

200

500

1000

2000

5000

10000

HEIGHT, h^,

IN

KILOMETERS

Figure 6.7

6-14

ANGULAR DISTANCE,

9,

IN

RADIANS

O
CO -i

> z o m
O)

m H m m o JO M o z
00

<e > in o ^ CO

m H

Q5

y)
CO

o
2
t-

O m H m
:a Ui

om
33

6-15

CORRECTION TERMS Aao, AyS^ FOR

Nj

301

20

40

50

60

90

100

0o,^r

IN

MILLIRADIANS

Figure 6.9

6-16

THE COEFFICIENT ClNg)

Figure 6.10

6-17

7.

propagation path with a

DIFFRACTION OVER A SINGLE ISOLATED OBSTACLE common horizon for both terminals may be
In

considered as

having a single diffracting knife edge.

as discussed in section
7.
1;

some cases,

reflection

from terrain may be neglected

in

in other cases,

ground reflections must be considered as shown

section 7.2 and appendix

III.

In actual situations, the

common

horizon

may

be a mountain ridge
[

or similar obstacle, and such paths are

sometimes referred

to as "obstacle gain paths",

Bar sis

and Kir by, 1961; Dickson, Egli, Herbstreit and Wickizer, 19 53; Furutsu, 1956, 1959, 1963;
Kirby, Dougherty and McQuate,
not provide an ideal knife edge.
[

1955;

Rider,

1953; Ugai,
of

1963]

ridge or mountain peak miay

is

The theory
[

"rounded obstacles"
[

discussed by Bachynski
,

I960], Dougherty and

1958, 1959],

Maloney

1964], Neugebauer and Bachynski

[

I960]

Rice [1954], Wait

and Wait and Conda

1959]

Furutsu

1963] and Millington, Hewitt, and Immirzi

In

1962a] have recently developed tractable expressions for multiple knife-edge diffraction.
relatively

some cases, over

smooth terrain or over

the sea,

the

common

horizon

may

be the
8.

bulge of the earth rather than an isolated ridge.

7.
1

This situation

is

discussed in section

Single Knife Edge,

No Ground Reflections
is

A
A(v,
0) is

single diffracting knife edge

where reflections from terrain may be neglected

represents the knife edge.
v,

illustrated in figure 7.1,

where
7.1

the

wedge

The diffraction loss

Schelling,

shown on figure
and
is

as a function of the parameter

from

Burrows,

and Ferrell [1933]

defined as

v = 2'4~KtJ\

sHJd tan a"'~taiip~)7)r~

(7. la)

or in terms of frequency in

MHz:
v
= 2.583 e\/Td~^'d^d'

{7.1b)

where the distances are

all in

kilometers and the angles in radians.

The distance

Ar =

r, + r, - r
1
<i

=9
and

d d^/(2d)

12
d

is

discussed in section

5,

and the distances

d.

from

the knife edge to the trans\, is in

mitter and receiver, respectively, are shown on figure 7.1.

the

The radio wavelength,

,

same
6,

units as the total path distance,

In this case, '

d.

The angles

a
o

and

are defined in

section

(3

are required.

= d = 0, A a or and si'ice d no corrections Lr' o St sr For the line-of-sight situation, shown in figure 7,1 and discussed in

Lt

= h^

section 5,2,

the angles

a a

and and

(3

are both negative, and the parameter are both positive and

is

negative.

For transhorizon paths,

is

positive.

7-1

If

is

greater than

3,

A(v,

0)

may

be expressed by:

A(v,

0)

12.

953 + 20 log v

db

(7.2)

The basic transmission loss,

adding
A(v, 0)
to the free

L, jj

^^ot

a knife-edge diffraction path

is

given by

space loss:
db

Hd
where

^^^' ^bf +

(7.3)

L
bf

is

given by
(3
.

(2. 31).

For frequencies above about

to (7.3)

GHz, an estimate

of the loss

due to absorption
If

1)

should be added

and (7,4).

the angles

and

|3

are small, the basic transmission loss over a knife-edge

:

diffraction path

may

be written as

L
which, however,

= 30

logd + 30 logf + 10 log a

+ 10 log p

+53.644

db

(7.4)

is

accurate only

if

v >

3,

d >>\,
is

and

(d/X.)

tan a

tan 6

> 4

For many paths, the diffraction loss

(7.2),

greater than the theoretical loss shown in

is

(7.3),

and

(7.4),

because the obstacle

not a true knife edge, and because of other

possible terrain effects.

For a number

of paths studied, the additional loss

was

about 10 to 20 db.

The problem of multiple knife-edge diffraction

is

not discussed here, but for the

double knife-edge case, where diffraction occurs over two ridges, a simple technique

may

be used.

The path

is

considered as though

it

were two simple knife-edge paths,

ridge-second ridge -receiver.

(a)

trans-

mitter -first ridge-second ridge,

tion attenuation

and

(b)

first

The diffrac-

A(v, 0)

is

computed for each

of these paths,

and the results added to obtain

v
is

the diffraction attenuation over the whole path.

When

the

parameter
resiilts.

positive and rather

small for both parts of the path, this method gives excellent

Methods for approx-

imating theoretical values of multiple knife-edge diffraction have been developed by Wilkerson
[

1964]

7-2

7,2

Single Knife Edge with

Ground Reflections

Theoretically, received fields

may

be increased by as

much

as

db due to enhance

ment, or deep nulls

flection

may

occur due to cancellation

of the signal by ground reflections.

Re

may

occur on either or both sides of the diffracting edge.

When an

isolated knif.e

edge forms a

common

horizon for the transmitter and receiver, the diffraction loss

may

be

estimated as:

A
where

= A(v, 0)

G(h^)

G(h2)

db

(7.5)

h,
1

2.2325 B^(K,b)

(f^ /a

J3 hte ^
s

5.

74 (f^/a,
'

) 3

h
te

(7.6a)
h^ = 2,2325 B^CK.b") (f^/aj^ ^
2
2

re

5,

74 (f^/aj^ h
Z

re

C^-^b)
)

a,
1

= d

Lt

/(2h

te"

),

a, = d, /(2h
2

Lr'^

re'

The parameters b", K, and B(K,b'') are defined

A(v,
"

in subsection 8.1.

The knife-edge attenuation

0) is

shown on figure 7.1, and

shown on figure 7.2.
6.
)

the function

G(h) introduced by Norton, Rice and Vogler

[19551 is
-'

Effective antenna heights

h
re

re

and the distances

d^

Lt

Lr

are defined in section

In these and other formulas,

is the

radio frequency in

MHz.

_
The function G(h
represents the effects of reflection between the obstacle and the

transmitter and receiver, respectively.

of the terrain

These terms should be used when more than half

horizon cuts a first Fresnel zone ellipse which has
lies in the great circle plane.

between an antenna and

its

the antenna and its horizon as focii

and

Definite criteria are

not available, but in general,

if

terrain near the middle distance between a transmitting

1/

antenna and
used.

its

horizon

is

closer to the ray than

d
J_ir
,

0.5(Xd
Xjt

kilometers,
G(h,)
2

G(h
1

should be

The same criterion, depending on

details of terrain are not known,
if

determines when

should be used.

When
used

an allowance for terrain effects,

G(h
1.

),

0,5(\d

%
Lt,

should be

Lr

>

|h
'

Lt,

Lr

h 1/2, where all distances and heights are in ^ ts,rs'

kilometers.

When
tennas
is

the reflecting surface between the diffracting knife-edge and either or both an-

more

than the depth of a first Fresnel zone below the radio ray, and where geoapplicable, the four ray knife-edge theory described in annex
III

metric optics
to

is

may

be used

compute diffraction attenuation.

This method

is

used when details

of

terrain are known so

that reflecting planes

may

be determined rather accurately.

Using the four ray theory, the

received field

may

include three reflected components, with associated reflection coefficients

and ray path differences, in addition to the direct ray component.

7-3

7. 3

Isolated Rounded Obstacle,

[

No Ground

Reflections

Dougherty and Maloney

1964

describe the diffraction attenuation relative to free

space for an isolated, perfectly conducting, rounded knife edge. considered to be isolated from the surrounding terrain when
j_

The rounded obstacle

is

k h[2/(kr)]^ > >

where

= 2tt/\,

is the

radius of curvature of the rounded obstacle, and h

2

is the

smaller of the two values

[{d

+ r

2/^
)

- r

and

[(d

2 ^2
i"

- r ],

The diffraction loss for an isolated roxuided obstacle and irregular terrain

shown

in

figure 7.3 is defined as:

A{v, p) = A{v,

0)

+ A(0,p)

+qvp)

db

(7.7)

where

is the

usual dimensionless parameter defined by

r

(7, 1)

and

is

a dimensionless

index of curvature for the crest radius,

in kilometers,

of the

rounded knife edge:

rp

= 1.746 e(fr)^
y^

^''^^

1 -i 0.676 r^f ^ [d/lr^r^)]

(7.9)

where,
r

is the

radio frequency in

MHz,
For

is the

path distance in kilometers, and

,r

shown

in figure 7.3

are the distances in kilometers from the transmitter and receiver,

all practical applications,
^ ,^
-,

respectively to
by
r
,

the

rounded obstacle. rounded obstacle

is:

may

be replaced

d d
for a

Where

the

is the

broad crest

of a hill,

the radius of curvature,

symmetrical path

r =

D s /e

(7. 10)

where

Ds

d,

Lt
9

- d^

Lr

is the

distance between transmitter and receiver horizons in

(6. 19).

kilometers, and

is the

angular distance in radians

Where

the ratio

a /p
a
r

the radius of curvature is defined in

in the next section,
(8. 9),

terms

of the

harmonic mean

of radii

a
t

and

defined

and shown

in figure
2

8. 7:

D
s

r =

St sr^
^

^^^^^^

eld

\ St

^+d sr

In

(7,7), the

term A(v,
7.1.

0)

is the

diffraction loss for the ideal knife edge

(r = 0),

and

is

read from figure

The term
p

A(0,

p)

is the

magnitude
7.4.

of the intercept values

(v = 0)

for various values of

and
is

is

shown on figure

The last term

U(vp)

is a

function of the product,

vp

and

shown on figure

7.5.

7-4

Arbitrary mathematical expressions, given in annex

figures 7.1, 7.3, 7.4, and 7.5
for use in

III,

have been fitted to the curves of

programming

the

method

for a digital

computer.

The diffraction loss

A(v, p)

as given by

(7.7) is applicable for either horizontally

or vertically polarized radio waves over irregular terrain provided that the following conditions are met:
(a) (b)

the distances

d,

and

are

much

larger than \,

the extent of the obstacle transverse to the path is at least as great as the width
of a first

Fresnel zone:

^^^d;--TTTd^,7d)

(c)

the

components

and

of the angle

are less than 0.175 radians, and

(it

(d)

the radius of curvature is large enough so that

r/Xf > >

In applying this

method

to

computation of diffraction loss over irregular terrain, some

is to

variance of observed from predicted values

is in

be expected.

One important source

of

error

estimating the radius of curvature of the roxinded obstacle, because the crests of hills

or ridges are rarely smooth.

to be greater at

Differences between theoretical and observed values are apt

UHF

than at

VHF.

7-5

7.
If

4 Isolated Rounded Obstacle with Ground Reflections

a rounded obstacle has a small radius and is far

from

the antennas,

(7o7)

may-

neglect important effects of diffraction or reflection by terrain features between each anteaona

and

its

horizon.

Such terrain foreground effects

10 exp(-23p)
p =
1,

may

be allowed for, on the average, by adding a term,

to

(7o7)

The effect
is

of this

term ranges from

10 db

for

to
)

db

for

When some information

available about foreground terrain, the

if

G(h

1> ^

terms

discussed in section 72

may

be used

more than

half of the terrain

between an antenna

and

its

horizon cuts a first Fresnel zone in the great circle plane:

A
where
A(v, p)
7,2.
is

= A(v, p) - G(h^) - G(h2)

db

(7.12)

defined by (7.7),

by

(7.6),

and the functions

G(h

_)

are shown

on figure

When

details of terrain are

known, and the reflecting surfaces between the rounded

obstacle and either or both antennas are

more than

the depth of a first Fresnel zone

III

below

the radio ray, the geometric optics four -ray theory described in annex

may

be appli-

cable.
field

In this case, the

phase lag of the diffracted field with reference

in addition to the

to the free

space

must be considered
$(v, p)

ray path differences of the reflected components.

The phase lag

of the diffracted field is defined as

2

$(v, p) =

90v

(j)(v,

0)

(t)(0,

p)

ct)(vp)

degrees

(7.13a)

where the functions

respectively.

(j)(v,0),

(J'C^jP)'

s^-'^^

(^(^p)
= 0,

^''^^

shown on figures

7.1,

7.4,

and

7.5,

For an ideal knife-edge,

the phase lag of the diffracted field is

$(v, 0) = 90 v^ +

4)(v, 0)

for v >

(7.13b)

and

$(v,

0)

(|)(v,

0)

for v

<

(7.13c)

7-6

KNIFE

EDGE DIFFRACTION LOSS, A(v,0)

C/5

UJ UJ (T

O UJ Q

Figure

7.1

1-1

G(hi_2)
^"""T " ~" ~
35

""

~~"

i\

"T"
k

~~~

p-

~"

" r- - -

^ u

^
ff
'

^ Ty
!

_ 7: -

~i' r
- ?;

^^ s ^ L
L

b b

aID
1

-_

t -| cr E - U"

b 2

~ [ ^ ^ S, L
^

-Ju

u
=
E
0(D

_J

=i
11 1 1 1

"~1

-J

= =
=

'

ts>
:

1
^

"''

E
1
-i

1
=:

5
:

=
z:

ID

i 1

11

II 1 1 1 1
1 111

= ID

^ 00 ? E
;

--:
^^
;.=
:

=t=:

1 R-^
E
!--::-:^

-j

-i--;

.;

;::..!

^ ^ V
te
-:
;

rt

=^

:V v=^ N
\

1
I

i1 i

==
-

ii^i S
T.-r:

\ \
\
W

1 1 ^ = i= li 1 1 i i i = = = =E = = s s s = P p H_i^ = = = = = ^^ = = = = = 3

\
^

s,

\ \

>

*
V
1

V \

X m
JO

==

=
= E

m
CO

==
1 11 1

-)

=E
=r= =
i

1
^ i=

1 1 1^
i
lA

c >
1

1 1!M

111111
----tz^

7^ lA

II

i1

= ^ i
i

X m

I ==

1
1

1 ji
'

I H
O" CD
II

no

<0
^

> ^
-r\

.<:i

___ 1

____

00

= = =
1

1
II
r^
=

1 i
~T~ _]_

1 1

1
11

= E
1 1 =

C Z H

11

1
1

1
1

= 11

li
II

11

r
7"
" 11

g 1 1 =
1

11
1

j
1

o
1

1 !

1
111 ^

11
1

1
1 11

W\
1

ll

1 11 1 i 1 1 ^1 1111

=
ll

1
1

1 1

/
f

:^

=]

rr ^=

a
j

-_

r:z- r r

^ q- J= r- J r
~

:- = - - = = z = - - ~ :r
---

1
n4^^^
:3~:^!-h_

=
=^

e| =
i\

1 11
i|

11

1i

:.

._L,..,^_---

. .

^- -i
-

='

'

"t~

~^: ;

11 11 ^ ~j " ^u :^ :^ ^ ~ j
i
i 1

li
^ f_

11

B^ i ^
,

1 -^ ~^
!

==
]^ E
L^:

_.:_.

--

::-.
!

Mi/
^y!
.-

"

-^
=

r^

_1

/
;/:
1

44^
i

^ p^i

IT ^
^

.i=

,.1

,-

'

:.:

'Ill

-4/^

"'

'-j

;
;

'

':T:!=

irr^ii^;;^

fe3=fe=

-^rj

r-

N=^=^iEH=

m^^i^^^m^^ ^.._:

^^nH"-- -I^MM^

=iEa??4^

7-8

DIFFRACTION LOSS,

A(v,/o),

FOR A ROUNDED OBSTACLE

Figure 7.3

7-9

INTERCEPT MAGNITUDE AND PHASE FOR DIFFRACTION OVER A ROUNDED OBSTACLE

UJ
cc
LlI

o UJ Q

bJ to

< X

<
O

Figure

7.4

7-10

UNIVERSAL DIFFRACTION CURVE FOR A ROUNDED OBSTACLE

en _i LJ

m o UJ
Q

>

>
-e-

Figure 7.5

7-11

8.

DIFFRACTION OVER SMOOTH EARTH AND OVER IRREGULAR TERRAIN

7.

Diffraction attenuation over an isolated ridge or hill has been discussed in section

The following nnethods are used

irregular terrain.

to

compute attenuation over

the bulge of the earth and over

The methods are applicable

to the far diffraction region,

where

the dif-

fracted field intensity

may

be determined by the first term of the

1949].

Van der Pol-Bremmer

to well

residue series [Bremmer,

This region extends from near the radio horizon

beyond the horizon.

criterion is given to determine the

In

minimum

distance for which the

method may be used.

proximation

some

situations the first

term

of the series

provides a valid ap1964].

to the diffracted field

even

at points slightly within line-of-sight [Vogler,

A
cal

simplified graphical method for determining ground wave attenuation over a spheriin this far diffraction region
[

homogeneous earth

was recently developed by Vogler

in section 8.
1

1964],

based on a paper by Norton

1941]

The method described

is

applicable to either

horizontal or vertical polarization, and takes account of the effective earth's radius, ground
constants, and radio frequency.
In section 8.2, a modification of the

method

for

computing

diffraction attenuation over irregular terrain is described, and section 8,3 considers the

special case of a

common

horizon which

is

not an isolated obstacle.

the attenuation due to gaseous absorption should

For frequencies above

1000 MHz,

be added to the diffraction loss.

8. 1

See

(3. 1)

and figure

3. 6.

Diffraction Attenuation over a

to free space

Smooth Earth

The attenuation relative

the distance dependence,

may

be expressed with four terms; one contains

two represent the dependence on antenna heights, and the fourth one

depends on electromagnetic ground constants, the earth's radius, and the radio frequency:

A
where
x

G(Xq)

F(x^)

Fix^) -C^(K,b)

db

(8.1)

X, = d^
1

Lt

X.,

= d^

Lr

B
-

(8.2a) ^ '

= f^

B(K,b),

(8497/a)^
' '

(8.2b)

The distances
in sections 4 and 6,

d,
f

Lt
is

d^

Lr
o

and the effective earth's radius,

a,

have been defined

and

the radio frequency in

megacycles per second.

of the radio

The parameters

K
and

and
e,

depend on polarization
tr
,

wave

and the
8.1
o-

relative dielectric constant,

8.2

and conductivity,

of the ground.

Figures
e

and

show curves

of

versus frequency for combinations of

Figure

and

corre-

sponding to poor, average, and good ground, and to sea water.

a = 8497

8.1

shows

for

km.

For other values

of effective earth's radius,

K(a) =

K(8497)

(8. 3)

General formiilas for

section III.4 of

K
III.

and

for both horizontal and vertical polarization are given in

annex

The parameter
8.3.

B(K, b") in (8.2b)

is

shown as a function

of

and

in figure

The limiting value

- 1.607

for

K
in
(8.1) is

may

be used for most cases of horizontal

in figure 8.4.

polarization.

The parameter

C (K,b)

shown

The function

G{x

in

(8.1)

is

shown on figures

8.5

and

8.6,

and

is

defined as

G(Xq) = 0.05751 Xq

10 log

Xq

(8.4)

and the height functions

F(x
1,

^
,

are plotted in figures

F(x)
is

8.5

and 8.6 versus

and

For large values

Because

of

or

x
is

approximately equal to G(x)

first

this

method

based on only the

term

of the residue series,

it is

limited to the following distances to insure that

is

accurate within approximately

1.5 db;

x^

x^(Ax^) x^(Ax^)

x^i^x^) > 335

for

B
B

= 1.607,

(K<0.01)

(8.5a)

x^{Ax^) > 115

for

0.700, (K

>

10)

(8. 5b)

For values

of

lying between these two limits, linear interpolation between the

8.6,

A(x)

curves of figure

and the two

minimum

values in (8.5)

gives a fair approximation of the

range of validity of

(8.1).

Using linear interpolation:

x^-x^A(x^,B)-X2A(x^B)
where
X

>x^^

(8.6)

mm
.

335

242.6(1.607 -B)

(8.7a)

A(x,B)

= A(x, 1.607)

+ 1.103(1.607 -B)

A(x, 0.700)

A(x, 1.607)]

(8.7b)

A(x, 0.700)

and

A(x, 1.607)

are the values read from the upper and lower curves of

Ax

in

figure

8.6.

The basic diffraction transmission loss,

L. j*

is

obtained by adding the attenuation

to the free space loss

.

defined by

(2. 31),

including an allowance for atmospheric

absorption when required

8-2

8, 2

Diffraction over Irregular Terrain

To compute diffraction attenuation over irregular terrain, the single effective earth's
radius,
radii
a,

used in (8.2)
a_
Li

is

replaced by four different radii as shown in figure 8.7.

The
a
t

a,
L

and

of the terrain

between the antennas and their horizons, and the radii

of

and

of the terrain

between radio horizons and the crossover point

horizon rays are

defined by

^
%
The distances
h
re
,

<t/<2^e)'

^2^<r/(^^e)

(' >

s V^^'^sr)'

^
d^
,

^s^sr/(9V
the effective antenna heights
6.
,

^^'^^

D s'
,

d
St

and the angular distance

Lr Lt are defined in section

sr'

>

d^

te

and

Four values
spending to
of

of
,

C
,

are computed from {8.2b) with C

oi

C
02

C
ot

and C

or

corre-

and

respectively.

These are used

B,
1.2,t,r
_

in

(8.3)

to obtain values

l,2,t,r figure 8.3 corresponding to each value of

K-

for the corresponding '^ " earth's radii, and

are then read from

K.
is then:

The diffraction attenuation relative to free space

A
where

G(Xq)

F(x^)

F(^^)

C^ (K^

^)

A^

(8. 10)

A
1

is the

atmospheric absorption defined by

and
8.4:

(3.1),

and

is

negligible for frequencies

,b)

less than

GHz,

(K

is the

weighted average of

G (K

and

C, (K

b)

read from figure

S^^l,2^
X,
1

[x^C^(K^) +x^C^(K2)]/(x^ +X2)

(8.11)

=B,C oi2
1

f^d^
Lit

X,

2 =B_C 2 02

f^

d^

Lr

(8.12) ^

X
o

B C
t

f ot

d
St

or

sry

f'

+ X, + x^
1

(8. 13)

C
01
,

^ 02, ot,

or

=(8497/a,

, ^ r l,2,t,

)'

K,

, ^ l,2,t,r

01, 02, ot,

or

K(8497)

^l,2,t,r
This method

^(^l,2,t,r'^)

is

applicable to computation of diffraction attenuation over irregular

terrain for both vertical and horizontal polarization for transhorizon paths.

The method may

d

be somewhat simplified for two special cases: diffraction over paths where

and for most paths when horizontal polarization

is

used.

i-3

8. 2.

Diffraction over paths

where

d
St

sr

For paths where the distances

be defined in terms of

sr and the corresponding earth's radius

St

and

are equal, the parameter

a
:

may

x^

OS

f^

Ds +x 1 +x^ 2
C
OS
=

(8.14) ^ '

D=2d St =2d sr
s

a=D/9, s s
K(8497)
,

(8497/a
s

)^

(8, 15a) '

K S =C OS

B
s

=B{K,b)
s

(8.15b)

where
using

x^

and

x^

are defined by

(8. 12).

The diffraction attenuation

is

then computed

(8. 10).

8. 2, 2

For horizontal polarization

100

For horizontally polarized radio waves, at frequencies above

K(a)

MHz, and with

1.607,

<

0.001,
=

the

parameter

B(K,b)

approaches a constant value,

and

B w

and

C.(K,b)

20.03 db.

Assuming

= 1.607

20,03,

the diffraction attenuation may-

be computed as follows:

G(x
i

F(x
2

F(x^)

20.03

db

X, = 669
1

f
x

d^
Xjt

/a^
1

x, = 669
^

f^ d^

^2
/a,
JLjr
L.

(8. 16a)

(8.16b)

1
=

669

f^ e^

D str

+ X, + x^
1

(8, 16c) '

where

str

= (d

st sr

)M\ d St
^

1/1 + d

l^ / ^ ^ /(d + d sr )* sr// st

The parameter

D str

is

shown
d
st
=

in figure 8. 8
,

as a function of

d
st

and

sr

oaths where For ^

simplifies to

sr

using horizontal polarization, the parameter

x
"

669

f^ (9

D s )^

+ X ,+ X.
1

(8. 16d)

8-4

8. 3

Single -Horizon Paths, Obstacle not Isolated

regiilar terrain or over the sea, a

In

some cases, over rather

common

horizon may-

be the bulge of the earth rather than an isolated ridge or mountain.

path distance,
in section
8.2
d,
is

For such paths, the

is just the

sum

of

d
L,t

and

d^

Lr

and in

this case, the

method described
The para-

simplified to one with only two earth's radii instead of four.

meters
is

x^

and

x^

are defined by (8.12), and

(8. 10).

^q

"^

^1

"*

^2

'

The diffraction attenuation

then computed using

The diffraction loss predicted by

this

method agrees very well with observed values

over a niimber of paths in the United Kingdom and the United States where the

common

horizon

is

not isolated.
of short to

For transhorizon paths

medium

length,

when

it is

not

known whether

diffraction or scatter is the dominant propagation

mechanism, both
to

diffraction and scatter

to

loss should be computed.

The next section shows how

compute scatter loss, and how

combine the two computed values when they are nearly equal.

8-5

"

"

'

'

CARRIER FREQUENCY

IN

MHz

o o
C3

V O o ^
l>

"

^
ffi

---.

^^^
"

^^

C3 C3 ro

'<=>

qm
C3

m 5 O m
q/ft

<

o O o ^ o 30 o
r\3

q
-

-- ^
-"

^-'
t^

DO J>

qm gr\3
,

3)

^
*

.^
,,--^
,.^00''^

-"'

'''

p b O

p rr O o^ o b""
^
1

^-^

"^^^
^,^.*'^

^^

^
^

^.^

^^ -^ ^

o o
C3

<=>

L *

'
^

.i*'^^
_ ,,,,*'^
^^***^

^^ o ro

\,
\

"T .^
\

^^

-^^

,<.^

-*

^
--

*"

'
^-

V"-^ ^
C=9

^^ ^'
-^
.^"^'^
,

'

u
II

c m

C3 cn

^
*

-<
\

^ ,-^
'

00
-fc>

<^
\ _2^
\

^ "^
,

-rf*

Pi**.

A
\

GO

10
->l

*'

5r

3
c> ro

^- ^'

^
-*

-'
'

---

"' *'

\ \
\

,,

\
\

"

'^
t

r^^

\ /^
y

C3 Ol

r' ^ _,y
y

^^' ,^ ^-^

,^'

^
/'

\
\
POOR

\
1
GOOD

\
AVERAGE

y
4

/
> f'

y
GROUND

z^'

q e

0=15

a e
=25
=0.02

<=> r\3

X /
4

'1
C/)

O o o

=4
=0.005

/
/

^'
cn

or

(AIR

IN

m
33

)
=1
IRTICAL
Tihos/meter

o
cn

DRIZONTAL

/
/'

8-6

1
b
IN

DEGREES

1 1

H w
1 1

df
jjf

l\
\
\

1
1

jj \
\

M.I.
II
1
1

\ \

X m

\
\
\

q
^

n\

q
0.001

(T\

O
DR //

/
7

>

4
\

\
)

\
>

\
\

7
/

\
-r\

o
-n

X m c m H O -< 2
1 1 1

\
\

\
\
/'
\

1
S

i
mhos/meter

ground

//

^
7
/,

/ /
/
\
\i
1

> X) > o m H < m m XI

3)

/
t\

\
V
\ L
d"

4
1

^i >r \
,

//
\

CO
"0

S
">v

/
\

\
Yt

A
/

\
>

7
\

a> ro

q
15
0.005

^
:rage

^ A
\

q m
" "

\
\^

\
>
'

/
/

15
0.005

^ m
RAGE

6
\
\
\

1
ground

\
k

/
/

/ /
/

mhos/meter

/
(

/s
^

ground

I CD m o o c > o m > m D

mhos/meter

I
1

s
\

\ \
\

ll
\i

\
\

u
\
1

)
\

no

q
7;

iTN
11

CO rn

o
8 3

g o i

1
J
J

/
'

5 \) 1
T> 1>

rj "''=' CD
0"=

o
e
=
SEA

o
t r
5i
1
WATER

XI

II
\

II
11

5
mhos/mete

8l

\ \
>

S ^
/

ll

11
1

"
\|

\
\

l! 1'
1

/
y
-

/
1 1

/
/ /

/
^

III

VERTICAL

/
\
\

HORIZONTAL

II

/
/
/
POLARIZATION

\
\

/
\
\ i

1 1
POLARIZATION

II

1
1
\

\
...

1
1
1

8-7

THE PARAMETER B(K,b)

iiljhiil;:-!^
0.1

0.2

0.5

10

Figure 8.3

8-8

Ci(K,b)

"
I\3

O b
C>1
i

[
\\

//
/

11

\ '

yj y ^
<^\.

X m
T)
\

c K^
CO

rr

^^v.
11

\^
^v^

/ / /
N.

> > m H m
zi

^X N
Vc y^
\\

/
'

o
o

\
\
\
1

Vi \ \\

\\\

1
'* ""
11

1
1
1
1
1

8-9

THE FUNCTIONS
E^

FCxj.Xg)

FOR LARGE
70

x:

FOR K^O.I AND F(x)~G(x)

GCXq),

m o a

0.6

Xq.Xi

OR *2

'N

KILOMETERS

Figure 8.5

8-10

;:

"

THE FUNCTIONS F(x,), FOR THE RANGE

3U

F{X2)

AND
I

GIXq)

<K<
_
I.

r^;
-

zn
2)- i-nrv.D:,

d|_2(km)=4 I225^/h|,2
20

(m)/Co' ,CoS(4/3k)'''

^r

k=aeff/6373,aeff;EFFECTIVE EARTH RADIUS

E: EH e:
L
.

1
;

b=0 b=90
.

FOR LARGE x:F(x)~G(x)

G(x)s0.05751)i-I0 log
-

bP=

180

-1

10

i
2

'
-

e;:
e:e'e; :e-;
;
;"

n
-

i'^
'

E
F

r ; ;
-

::e
-^r:

-'
-

^ / 9
_

f/j

^r
E-.

H
:

f/l

i
E

"'

:E'
;

E E;

:-

E;H E:
- -_

If
'-

r:

:,

E
-10

r^ V^:\i:

^'

EEJEt
.-

E'

rl\

E^: ;: EE.-: he:E-.

E---:
_

Er Ei
K=
1

E;
;,

EEEE^ ^:^;"; EEH^- :h-,

^
(^ !?

/^

"'-^

-;::
--V:
'

r\K
=

=--:::
=:;;:

" ~ ~*
1

"

G(o'
_4_

<i.\0

-20
i ,

0.5

-~ >-

*
-

^y
^-:
>

^ / ^ /
^

Bi
-30

-. U.3
1

EE
, ,

E:
-

.:---

,,,,

-."

_
^
_J

s^

'-.

:-

S;

^
,

EE
^

--*+.

T-'

;i-;

K
-

0.2
-^--"i

^^
.::;

-^
^\ y.
"

=E;;r ::E1
= rr_. rrrii

^^
E
--

^
r.

--

^
.

= 0.l

^
:yy

~-

^
'y

^r
'i

\0^
.-:

E-

/
\

^:-

E;^lt
i

^">

E; ^=
-

z <
CM

;?Ef

;e

/(^
)'

^^=-_-;-_-

^'
r-_

S
1l

-50

EiTSE -EE" EEE-^ E ;eh:

=
E
;
_
,-

eee:
-^:-,
"--""

^^

r
"

K = 0.0l

.i^ . -^^ _.-- < ^ /-_.--. -^^

.,

.*

y 'A y
y v/
]':"'
'
/.

^ ^

^ ^ -

'y^yf

--

:;

-60

^^^E
-.

:r
E^'^^:
'

eee:
:-;-"^^

ii ;^ =i^^' ^EE = "^Er -70 E=

-

-^ r

t/

-::-:.
-1--
IE

^:.s.'
'

'f-

E:

^^^--^
fEE
;:;
E-:

_yj

e=

ERROR

IN

At<

l.5db;

1
~-H:_t
'

|-335,K<O.I

eh:
;=,
_^
"'"

V ^J^
J

'0-'^""'-2^'2'~iM5.K>IO
_:
1

^.4 iv < '^^ /

j^' jr^

--- -'

.^

Z-r- T

--~-z:

=r

___ ,^^

_ - -'
""
"li.

r-'
yc

E3.\

: :^= --

K =0.001
--fEE

"^

#" ^y^-- EEE

E_EEE_
: :

_--E-:

-eIeh:

ee:

:h

H
::

-""-:^-

e;-:t-

/
_

^y
---

^^

-"'

1.0
1

-80

SeI

-iii^'E:

^ee;
e""E
= L

rrE;

=E Bzi
:e:

-^

=-; EE;
jJ;^ E._

E
;E

T,

/ /f /
^f /
je:
-~i

y ;^

h:
E e:
-

-.=^"Z

~~- E^

K >i n

/
-"

^
-'

0.8

S^

"e
-90

:';

^^'

'--'-^
3^irr:-

HE
--=^
e:-.:;-

:-:

^:ir;=HE

z:r^^

"-

i^^
='^

= 0.(

500

lE > V, '^
^^

--"-H

E_

~~7 '^^

EL
.-:.:

-7
E i-i
h:'
;e -t;"-

^^
--

^^-' 1^- ^_i

~->
^e::::eee
'

-I00~

i=->-

==^
=E l^
.^^

beT; k = o S3-; ; E>/

-110

/ /

7 /

^V

//
;--:

K^ EH
:

r^ -"-^-"__
. "

e:^ ^3^
-^
:."r::t

E; y'^-

H.
m-

7
/

E^^ZTrrr-r

n^H

":::."

E-::

H -;ee

H
--;::

Tfrzi

/ EE;.

HE

n 0.6

E;:

.-

--

^^ EEHr HE' ^~ HH H E-H EE=E-S ErE; EEE l-:i EE EH EH

-^^T: ^-r:
:

_E--

-'-

4=== h:: EE- H^

= SEE EE
:hh =::
't^.

^-^E -in^ EEE

:5: EE e: e;

f-

:h: .H"

/
1

EE-

-:

-":1

ee:

^ EE

i;r

ee:

H E EE H ~; E
;::

^^
:ee/

'EE- ^;;
^-^-"^
^^"-5^;

n ^"
k"

>

E fi

n
;:

:'

?
/
^;: '^:

0.4

e;;

H;^
EE_

_;^.
'

:/
J.xiH

^ k
/

^n;
ee^e;

~-

/ \y
^iri
1

0.2

' T-E
l-

EE;L;-

-,_=^-: =;-;-:
E:-,

EE EJ ^j;
EE Ej ;-E

==--"

-120

EEIt^ -^-"-^ ~2

^^1

.^
2
1

<<

^^.

-L-'--...
'

^:
20

y
^-

^ i_

l(

50

10 00

0,

1,

^0^0

'/3

^MHz do.|,2(km)

Figure 8.6
8-11

GEOMETRY FOR DIFFRACTION

OVER IRREGULAR TERRAIN

Figure 8.7

8-12

'sU

m
5 > m m
JO

00 00

cz>

c^oc^*:

dsr (OR djf

8-13

9.

FORWARD SCATTER
A method
is

This section gives methods for calculating reference values of long-term median radio

transmission loss over paths that extend well beyond the horizon,
bining diffraction and scatter transmission loss estimates

given for

com-

where

this is appropriate.

The

methods

of this section provide a

reference median value of basic transmission loss.

of

Em-

pirical estimates of the

median values and long-ternn variability

transmission loss for sevIII.

eral climatic regions and periods of time are given in section 10 and annex

For long tropospheric paths

the propagation

mechanism

is

usually forward scatter,

especially during times of day and seasons of the year

when ducts and elevated layers are

becomes more coherent
it is

rare.

Often, for other periods of time, as scattering

more

properly called reflection.

path during

The examination

of

transmission loss variation over a particular

tilts,

some period

for

which detailed information about layer heights,

and inten[

sities is available

can be very illuminating; see for instance Josephson and Eklund

a turbulent

1958]

Sometimes no distinction can be made between "forward scatter" from

and "incoherent reflections" from patchy elevated layers.
in

atmosphere

The
,

first viewpoint is

developed
1957],

papers by Pekeris
[

1947]

Booker and Gordon

1950a, b]

Megaw[1950,
[

1954,

Millington

1958], Staras[l952, 1955], Tao[l957], Troitsky

1956,

1957a], Villars and

Weisskopf
is

1955], Voge

1953,

1955], and
[

Wheelon
I960,

1957,

1959], while the second viewpoint

,

ennphasized in papers by Beckmann

[

1957,
,

196la, b]

duCastel, Misme, and Voge

[

[1958], Friis, Crawford and Hogg

19 57]

Starkey, Turner, Badcoe, and Kitchen

19 58]

and

Voge

19 56, i960]

The general prediction methods described here are for the most part con-

sistent with either viewpoint, and agree with long-term

median values
IV.

for all available data.

brief discussion of forward scatter is given in

Annex

The reference value,

L,

bsr

of

long-term median basic transmission loss due

to for-

ward

scatter is

L,

bsr

= 30 log

20 log d + F(9d) -

ox

00a
+

db

(9.

1)

the For most applications ^-^

In (9.1)
f

first three

terms

of (9.

1)

are sufficient for calculating

L,

bsr

is the radio frequency in

MHz, and d

is the

mean

sea level arc distance in kilo-

meters.

The attenuation function F(9d),

the scattering efficiency

term F

and the frequency

gain function
defined by
(3,

H
1)

are discussed in the following subsections.

3. 6,

Atmospheric absorption,

A
be
in-

and shown on figure

may

be neglected at lower frequencies, but

may

more

than 2 db over a long path at 1000

MHz, and becomes increasingly important with

creasing frequency.

For ground-based scatter

distance,
r
,

links the sea level arc distance,

d,

and the straight line

loss
/ d)

between antennas are approximately equal.

satellite,

To estimate transmission

between the earth and a

where

is

much

greater than d, a term 20 log(r

9-1

should be added to the reference value L, Annex III contains a discussion of transmission bsr loss expected when antenna beams are elevated above the horizon, or directed away from the
.

great circle plane determined by the antenna locations.

The median forward scatter transmission loss, L

L,

is the

basic transmission loss,

bsr antenna gain,

to

minus

the path antenna gain,

L
gP

when

Section 9.4 shows hov7 to estimate the loss in path p there is a loss in antenna gain due to scatter. Section 9. 5 shows
.

how

combine diffraction and scatter losses.

Following Arons [1956], the scattering of dif-

fracted fields and the diffraction of scatter fields are ignored.

9.

The Attenuation Function F(9d)

of the

The attenuation function F(9d) depends upon the most important features
path and upon the surface refractivity,
to data available in the

propagation

The function includes a small empirical adjustment

frequency range from 100 to 1000 MHz.

9.
1

For most land-based scatter links figure

the product 9d for

may

be used, where F(9d)

d, is in

is

plotted versus

= 400, 350, 301

9,

and 250.

The path distance,

of

kilometers and
1

the angular distance,

in radians.
of

For values

9d < 10 the curves of figure 9.

are valid

for all values of

s

s.

For values For

s

9d greater than 10 the curves

1

may

be used for values of

from 0.7

to unity.

greater than

use 1/s

in

reading the graph.

1 1

For highly asymmetrical paths with 9d >

are used to obtain F(9d).

10, figures III.

to

III.

14 of

annex

III

Annex

III

also contains analytical functions fitted to the curves

F(9d) for 0.7 <

<

for all values of the product 9d and for

= 250, 301, 350,

and 400.

Using the expressions for the function F(9d) with

= 301,

the reference

median basic trans-

mission loss

is

For 9d s

10:

L,

bsr

s 135.8 + 30 log o

+ 30 log o 9 + 10 log D d + 0.34 9d

(9.2a)

For

10 < 9d

50:

L^

bsr

s 131.4 + 30 log

+ 35 log 9 + 15 log d + 0.27 9d

(9.2b)

Reference values

may

be computed in a similar manner for other values

in (9. 2)

of

N
s

The approximations

do not make any allowance for the frequency gain function,

at

For usual cases

of

transmission

frequencies above 400

MHz

the approximations

in (9.2) give

good results.

For

the higher frequencies an estimate of atmospheric absorption

should be added.

For lower frequencies, or low antenna heights, ground-reflected energy

tends to cancel the direct ray and the approximation in (9.2) will underestimate the transmission
loss.

9-2

9.2
It is

The Frequency Gain Function,

H
of

assumed

that

if

antennas are sufficiently high, reflection

energy by the ground

doubles the power incident on scatterers visible to both antennas, and again doubles the power
scattered to the receiver.

As

the frequency is reduced, effective antenna heights h

/k and

re

/\ in wavelengths

become smaller, and ground-reflected energy tends

of the

to cancel direct-ray
is

energy at the lower part

frequency gain function
loss.

common volume, where

an estimate

scattering efficiency

greatest.

The

in (9. 1) is

of the

corresponding increase in transmission

This function first decreases rapidly with increasing distance and then approaches a
constant value.
of

For h
re

te

/\ > 4 a/d

and h
is

re
o

fk > 4a/d,
6 +

is

negligible
:3

The upper limit

as h
te

and h

approach zero

H ^
of

db,
o

where A

the diffraction atten-

uation

over a smooth earth, relative to free space, at

be estimated from the

9=0. For

frequencies up to 10 GHz,

may
o

CCIR Atlas

Ground Wave Propagation Curves [1955, 1959].

should rarely exceed 25 db except for very low antennas,

The frequency gain function, H

lengths, path

depends on effective antenna heights

t)

in

terms

of

wave

asymmetry, and

the

parameter

shown on figure 9. 2 and defined as

T]

=0.5696h

+ (0.031

2.32

N
-

x 10

+5.67N xio" )exp(-3.8h

km.

x lO"

)]

{9.3a)

sde/(l + s)^

{9.3b)

The parameters

and

are defined as

r,
1

=4-n-9h
te

/\,

r^ = 4119 h
2

re

/k

(9.4a)

where
the

is the

angular distance in radians, and the effective antenna heights h

te

same

units as the radio

wave

length,

X.

In

terms

of

frequency

and

may

are in re be written

r,
1

=41.92efh
te

r^=41.929fh
2

re

(9.4b)

where

is in

re For the great majority of transhorizon paths,

te
s,

radians,

in

MHz,

and h

are in kilometers.
s

is

within the range

0. 7

<s <
III.

The

effect of very small values of

with a

may

be seen in figures

III.

15 to

19,

which have been computed for the special case where effective transmitting and receiving antenna heights are equal.

9-3

a)

For

ri

greater than or equal to

o
(r,)
1

1:

Read H

and

(r.)
Z

from figure

9.3; then H. is o

H
where

[H
"

(r,) +
1

(r,)]/2
o
2
-^

+AH o

(9.5)

aH

6(0.6

log

r|

)log

log

q.

If
T|

-n

= 4,

s =

SOS
s

a /p
o

q = r /sr
d
L

>

the value of
1

for

ti

= 5

is

used.

The correction term

db,

AH o

is

zero for

1,
t)

or q =

and reaches a
of

paths

when

=1.

The value

maximum value, aH =3.6 AH may be computed as shown

s

for highly

asymmetrical

or read from the

nomogram,

figure 9.4.

straight line between values of

and q on their respective scales intersects

the vertical line

marked

,
ri

This point of intersection when connected by a straight line to

intersects the

the appropriate value of

AH
in

scale at the desired value.

The following limits should be applied

If If

determining

AH
1.

s
s

> <

10 or
0.
1

q > 10,

use
1,

= 10

or q = 10.
1

or q s 0.
(r

use

s = 0.

or q = 0.

If
If

AH AH o

o^olJ
> [H
1:

H 02-' (rjl H
o

/2,

use

would make

negative,

H =H(rJ oi o use H = 0.
o

+ H(r,).

o2

b)

For

ri

less than
s

First obtain

H
is

for

ti

as described above, then read

's

for

ri

=0 from

figure 9.

5,

's

The desired value

found by interpolation:

The case

r\

corresponds to the assumption

h
=
te

of a

constant atmospheric refractive index.

in

case
s

A H
1,

special case,

re

r
1

= r
2

occurs frequently
1

systems design.
r;

For
5

this

has been plotted versus

0. 25,

in figures III.

5 to III. 19 for

= 1,
is

2,

3,

4,

and for

= 0.

0. 5,

0.75 and

1.

For given values

of

r|

and
is

s,

H
=0)

read directly from the

r|

graphs using linear interpolation.

of

No correction term
9. 3

required.

For
is

<

the value
9. 5

OS
(ti

= 1)

is

read from figure

with

r
1

= r 2

and

(n

read from figure

as before.

9-4

9.3

The Scattering Efficiency Correction, F

o
in (9. 1) allows for the

The correction term F

reduction of scattering efficiency at

great heights in the atmosphere:

F
where
and h

= 1.086{ti

/h )(h
s

h
1

h
J-it

h
J-ir

db

(9.7)

r|

are defined by (9.

3)

and h

is

defined as

h^

is
n

sD

9/(1 + s)^,

D
S

= d

d_
JljZ

d^
i-iX

(9.8)

The heights
in section 6.

of the

h and the horizon distdncies d, ^ Lt Lr Lit, All heights and distances are expressed in kilometers.

horizon obstacles,

d^
l,r

are defined

The correction term F

large that h

exceeds

2 3

decibels only for distances and antenna heights so

exceeds h

by

more

than

kilometers.

9-5

9.4 Expected Values of Forward Scatter Multipath Coupling Loss

Methods

for calculating expected values of

[

forward scatter multipath coupling loss are

given in several papers, by Rice and Daniel

[

1955],

Booker and de Bettencourt

1955], Staras

1957], and Hartman and Wilkerson

1959]

This report uses the most general method available

[

depending on the paper by Hartman and Wilkerson

1959]

As explained

in section 2,

the path antenna gain is

G
p

G
t

L
gp

db

(9.9)

where G

and G
of

are free space antenna gains in decibels relative

to

an isotropic radiator.

The influence

antenna and propagation path characteristics in determining the loss in path

antenna gain or multipath coupling loss

separately.

are interdependent and cannot be considered

This section shows how to estimate only that component of the loss in path antenna gain

which

is

due

to

phase incoherence
9. 6

of the

forward scattered
r|
,

fields.
(9. 5),

This quantity

is

readily ap-

proximated from figure

ii

as a function of

defined by
If

and the ratio

Q/H, where

= 25 =
t

is the effective
,

half-power antenna beamwidth.

values of

the antenna

beamwidths are equal,

antenna beamwidths

ii

ii

and

if

s = 1,

are not equal the loss in gain

may

from figure 9.6 are exact. gp be approximated using f2 = sjn

When
f2 t r

The relation between

tropic radiator and the half

the free-space antenna gain

= 25

in decibels relative to an iso(2. 14) as:

power beamwidth K

was given by

G
where

= 3. 50 - 20 log 5 = 9. 52 - 20 log

f2

db

and

fl

are in radians.
efficiencies for both antennas,

Assuming 56% aperture

e/n s e(nii^)"'^ = 0.33 5eexp [0.0576 (G

+G

)]

(9.10)

where

is the

angular distance in radians and

2

are the free space gains in decibels.

Section be computed as

shows
:

that the gain for parabolic dishes with

56% aperture efficiency may

(2. 16)

G
where D
is the

= 20 log

+ 20 log
f

f -

42. 10

db

diameter

of the dish in

meters and

is the

frequency in MHz.
< 25,

For dipole-fed parabolic antennas where

gives the following equation for the antenna gain

10 <
:

D/\

an empirical correction

(2. 17)

= 23. 3 log

+ 23, 3 log

f -

55.

db

9-6

The general method for calculating L

V =

requires the following parameters:

r|

/2,
S

|i

= 6
r

/5^
t
:<

(9.11)
1,

For

sii

>

1,

n = a /6^.
o
t

For

S(i

n =

(3

/6

(9.12a)

fi

(n+ 0.03v)/f(v)

(9.12b)

f{v) = [1.36 + 0. Il6v]

+ 0.36

exp(-0.56v)r^
r

(9. 13)

where

r\

a
o

and

have been defined,

5
t

and

are the effective half-power semif ( v)

beamwidths
(9. 13) is

of the transmitting

and receiving antennas, respectively, and

as defined by

shown on figure 9.7.

Figure 9.8 shows L

versus n for various values

.

figure 9.8 for l/(sn)

gP instead of sn

of the ^ product

sn

For

six h-

<

read

9. 5

Combination

of Diffraction

and Scatter Transmission Loss

slightly

For transmission paths extending only very

will be the

beyond line- of- sight, diffraction

dominant mechanism in most cases and scattering

may

be neglected.

Conversely,

for long paths, the diffracted field

field,

may

be hundreds of decibels weaker than the scattered

In internnediate cases, both

and thus the diffraction mechanism can be neglected.

to be considered

mech-

anisms have

and the results combined in the following manner;

R(0.
5),

Figure 9.9 shows a function,

which depends on

the difference

between the
)

dif-

fraction and scatter transmission loss.

Calculate this difference

determine R(0.

5)

from figure 9.9 and then determine

cr
,

sr dr the resulting reference value of hourly

(L

in decibels,

median transmission loss, L

from

the relation

L
the difference

cr

dr

R(0.5)

(9.14)

If

between the diffracted and the scattered transmission loss values exceeds

15 db, the resulting value of

if

this is the

smaller value.

will be equal to L if it is smaller than L or to L * cr dr sr sr In general, for most paths having an angular distance greater
, ,

than 0.02 radians the diffraction calculations

may

be omitted; in this case,

cr

= L.

sr

9-7

THE ATTENUATION FUNCTION, FiOd)

d
IS IN

KILOMETERS AND 6 (0.75^S<1)

:'-..
!-

IS IN

RADIANS

lU

-.4.

80

^:^- "s. --/\ ^*V= V \ >. ^ii^ S^ ^?t; ^\ i^P' ig.^ -N s

i'-L-.
,.

;|.

---

\
L

-;,

..;|
,

.'^

jl;

\
--

.-=:--

'

,.,

:.::

''.:%

-^

:-

^
y >
k!
':

V \ g?F=^ \, \
3ESr-E
s^3;~3"^

E./:-;

220

90

S NK, ^> \
^,.

\S
N ^
,

=::

\
y-|Ns = 30l

230
;!i

3;^-"

^3==
5?
100 asr...

'^yE V *v \ \ V^ ^^

s.

^ 1^

^ =N^
E

r:'

"
!:

A
1

:- E

,::

i=y

-E

ii

..

~^'
3=:::.

S^

\
S

>FT
eS.
It

N,
/s s

240
V

\ \ *^ ^Ns = 400[i:

s^^V
=Nv

110

="r -;.'t

b; -:
=E^'-;.

\ s \ ^\ ^y?d1

T _i=1350|-

\ \ \
13011^
1250
>.
-E.:
I

"

'

.'

'li

':^.

i
:
'

r^

=
.

~= ._
=

N, ,\
=

120 T

.-

E
.;

K\S
--

N.

\
E.

"ji

250

\.
^ ^

s,.

s^

;
-

s,

~H

"

-:
-1.=

=;;.

^NV
--.:

\- N,

s
"^

V \,

EE

"

-1e.-..-

\
"s

260

T3

E=:
130
=;."^-.
3^.;';.".
^"

\N
V

^v
s
''s

S^
S.

--:.

_i UJ

e:^;: =i^^-: iB^-

;...

140
E;r:-

:-,;t^

ss S =
T--n

;::-

" ':

V \ x,

::.

Q Z

^,150

CD
Li_

^=;

.:.

-.E-:.

\ 'v"^ ^.V VyN \, s^ s s, X a > vs s -\\,\

s
-

270

-s

\
\
1

280
-

V
v
.

::_.,

...

-:^

Ik

E=tE7

4^
=fcl=t

\
s

s>

1.;^:

-.

-:=--

3.-:

==fet= =-V

160

;-;-

.._U- E-E ~- ;^E^S^E __g^. -.-p~" E --~ =?EiF -E;;

s\

V
s-^""

,;

V
:.JT

.::

^, s,
rE

\
tk

A
. :

100
S\'''

200
:tt-

500
,3m:
-7:-

:
Ei:
.

LU 290
^:
:

.3
:

4.

'4^

s
^

\, ^V

S
X

:
.

---l
^

~r
=-'-

:
"

-^-:

\^ \A \
S.^^V
"^.

170

#E E; #F
=?;.

u M
=
;

'

\ \ \
\ s\
A.

jj

:1
'
.

=r::;

=
;e

".

^y\\ V \
\,

"i"

g==-:
33?;-'/
:
; ;

i;5E

o
-S,

~s

BE

E---i
.

:|

iiii

i'rJ
'=:'.
i
;

-:|
V:
.

180

is;-

EE:-:

E-E
:

^ ^^
i

El IE

m^ =^

m p~
rUl

,=E

?
E.|
fij"
:

^EE--

;-

"33 BS iia
iiE E:

1^

'

Li/s:

== S=;
4|S

'^

190

JjT

V tE jo ? w
iE:

\^

EEE

EE.

E-";
-

E-E

Jj.*^

ElEE

'.':

"='
si
'^
:

^3=; Sh: ?!ir

200

m
=1^::-.:

(^t ;= s^.e:
' Si

"

"17

V EE wA
^
':

..

'i

T.

=^^--rt^-r
"
:

:iry

ft^

'".i
.

\'r
yfcs
.-

"i
a

:',;:
'"-

210

__^.
:

1^ =^ffT
oin
0.01

.^-i:
t..-:~-

"'\
!;.

''i!

r\

i
.:|

r.

4^

,..J-E:1 ::_:
1

,_

:E^Ey
20 50
100

0.02

0.05

0.1

0.2

0.5

10

9d
Figure
9.1

9-8

THE PARAMETER

T^glho),

USED TO COMPUTE

H^

y
7

y r

/
T^s
\

=0.5696 h,j|li- (0.031- 2.32NsXlO"^+ 5.67n| xlO')e"'''o'"

1
\ \

ho
6

Sd/(i +

S)

km

,.

>
5

/ f

A '^X

/
r

/
-

\ s77s =0.5696 ho

/
jt

./ f
f

^J

^75

/ A // A
1/ 'A tjT
.

/ ^A 7 / A A/
f

/^

//,

J
/ 2

/y ^^ii^
j^'

/, ^^
/, K- & y /^ V /^ T A^ r

^ r ^
/

/ Cu

'~~~^

Ns

-400
1

350
301

-_J___]__|

^ ^

250

10

12

13

hQ

IN

KILOMETERS

Figure 9.2

9-9

THE FREQUENCY GAIN FUNCTION, Hq

.''2

Figure 9.3

9-10

O
-c
I I I I I I I I I I I I

C3 en
I I I I I

oo
1

CO CD

II

mil

III

X3
II

[> T.
II

O)
^-^

ro

"^

>
Cn
-C*

fN3

w -<

<T>

o
(O

w
OJ

-d
v>

^5

ro OJ

-c*

cji

II

liriiliniliriiliiiiliii

Tiiilmiliiiiliiiiliiiii

o o

o
lO
(/)

o
IQ

o
CD

3]

c
(0

o m H m m
>
o
0.7

0.5

0.3

0.2

0.1

9-11

/ /
:

/
(0.7<S<
1)

'

THE PARAMETER Hq FOR 773=0

J-LJJ,
3a^^^:-

--;

.-:-:

3-r-E
-:-:;

5
~Y

'

"-

^-'
Ei^zi.
Ei=":l
10

y^ :^

-:;< ^ y ^
_,i>^i

, -"-"
::
:
:
ill:

jy
> ^
/
,
'

^^^ >

"''

=!""

ilTr^
H':-

// y

/^
''

<^

^P ^-'---'

^"
'

^^--TT
_^_--.-

='^^r
3^_.:-:.

y/

=n:T-"
15

?-.

/^ / / ^/ / // /
[r 2
-

.*'^

"

lUU

^y

|T=i

iTTr'::-*
z^--::

/;.

r^ //
E;=^^
;
-"

-;|y

/ / y^ ^'' y^ / y// / /
/
''/

// ^/ y
'

x''
^^

yy^ ,^ " >^ ^^

^,,,^

_-

'""

---'

.^^

---^

"^

-> -r*-"
-

20

"

//^
/-y/

/ '/
'/
/

//
'/

yyy
/
"^

v*

_^

_,--<^
,-

.--

Jev^'

=:^ ;
E>:-

"

i?r-=^'

25
E"
-

=
Z
30
T

-.

|0 3f

/.'

o
o
35

E"-:;_^

^/,

y ^ / [0 ^^ r2 77~/A yy / / y' y -'^ ^,-J04t ^ / '"' / y ^' ^.< r/W ^o^^ / ,^ ^// / M' .'^^ X. y ^ ^, % ^7> ^ -7" y ^ ^ i^? ^v^r / y ,'y y ^ ^"-"""^ __^ -^
[0

7--/K/

yy/
^

// /^ / /

_^-'^ \.-'^
''''

'

-^
_^__ -rr

'

^/)

^:Z

Ej'Erz

ErE-rZ
-J-

^:^::-

40

45

50

55

/ / ^7^^ y y'^y -^ ,^^ -^/ / 'A / / ^ y / \^-^^ ///f /' 7 r / y //A '^ ^ J i^ "^ / / / / / ^^, / ///} y z''-^^, Jv ^-^ / \0^ ff/// '//a / >* y // / f/^// 7 y ^/ / / .y ^ - VX / -;?^^ / //// / / / / / f/ / "'" ^^y /// // / / / <fy ^ /// / / / / / / / y' / ^s^ y /// / / ^, / <^3>^ r // Vy // ^ ^ J V / / A/ / /y ^ / y / ^ ^ / / / / ^>^ / / / / // / / V / X / //) / / / / y / / /
'---

/J/ / y' y' -^' ^^""^ 6 /// / / / // / / / /y^^y/ '^' y /'.-^ ^ ^^-"^ / / ^~^ // y^ y / /' ^-'' y / '/ >v4 > /v ^
^

'/j

v) '/f ^

/xx

'

y yy

^^

.^-

--=
;i

^,

i:;
;

' 1

''

'

//

-"^

//(

'^
'

,''^::

/'

^""

,/

____

,''

__

-"

_;

---'

/.

'

/'

z'

'

'

//
en

//

/ //
i
1

/ ry

/
V

;::

:
.

;
'1
:

/ /

/
..

J
0.2 0.5
I

0.01

0.02

0.05

0.1

Figure 9.5

9-12

IN ANTENNA GAIN, Lgp assuming equal free space gains G^ and Gr at the terminals of a symmetrical path

LOSS

D.^ =Cl^

,s =

P^"

0.1

0.2

0.5

5
to

10

20

50

100

Ratio 0/ii of angular distance

half-power antenna

beamwidth, XI

Figure 9.6

9-13

Hv)

m o o H > q
31

c a
C/)

9-14

LOSS
60
56

IN

PATH ANTENNA GAIN

vs

ft

52
1

/
/

CO _J LU go

48

7
/

s^-/-V/'
44 40
S/J,=

o Q

s/z=2~~

<

36
32

7 \
'/

w
J r
\

LU h-

<
I-

28
I

T\ /
/
//

< a.
CO
(7)

24
t

20
1

/ A /
\
1

o
Q.

16
t

12

^ ^ ^^^
0.1

^ i^
I

4 V

i
//

0.2

0.5

10

20

50

100

A n
Figure 9.8

9-15

P.{0.5)

IN

DECIBELS

-D l
C/)

m m m
>

2 2 > H 71 T) J> en l> ^ * "D K |- | ~n m H

C7^
II

31

r-

c
(0

Q.
1

O m o CD m
rco

:o

Ul

u m -L m (n H n J> c ^ no c C" 1 t: l > m in H H > < > H m u H H "n m "n H

::d

C/)

-z.

C/5

-^ in

T)

:d

>
n

CD

T) H H in m n (J |
<;:

m
r~

-n

H X

9-16

10.

LONG-TERM POWER FADING

The variability

of

tropospheric radio transmission loss depends upon changes in the

atmosphere and upon complex interrelationships between various propagation mechanisms.

Short-term variability or "phase interference fading,
curing
"

associated with simultaneously oc-

modes

of propagation,

is

discussed

in

annex V.

The effects

of this type of fading ex-

pected within an hour are allowed for by determining an hourly median rms- carrier-to- rmsnoise ratio which defines the grade of service that will be provided.
identified with the variability of hourly

Long-term power
is

fading,

median values

of

transmission loss,
of

usually due to

slow changes in average atmospheric refraction, in the degree

or in the intensity of refractive index turbulence.

atmospheric stratification,

An estimate
tant
to

of the

long-term power fading

to

be expected over a given path

is

impor-

insure adequate service over the path.

The possibility that unusually high inter-

fering fields

may

occur for an appreciable percentage of time places restrictions on ser-

vices operating on the samie or adjacent frequencies.

The basis

for the

mainly empirical pre-

dictions of long-term variability given here needs to be well understood in order to appreciate
their value as well as their limitations.

An increase
scatter fields but

in

atmospheric refraction increases long distance diffraction or forward

lead to multipath fading problems over short paths.

may

Increased turbu-

lence of the atmosphere

may

result in either an increase or a decrease of radio transmission

loss depending on the geometry of a particular path and on the dominance of various propagation

mechanisms.

Increased stratification favors propagation by reflection from elevated

of

layers and sometimes the guiding

energy by ducts or layers.

Such stratification usually

increases long-distance fields but

may

be associated with prolonged fadeouts at short distances.

Just beyond radio line- of- sight, fading rate and fading range depend in a very complex

manner on

the relative importance of various propagation

mechanisms.

During periods

of

layering and ducting in the atmosphere, transmission loss shows a tendency to go into relatively

deep fades, with durations from less than a minute

fraction signal fades slowly
if

to

more than an

hour.

Ordinarily a difduration and very

at all,

and the fades are

of relatively short

deep.

A tropospheric forward

scatter signal, on the other hand, exhibits the rapid and severe

fading characteristic of the Rayleigh distribution.

the scattered
tion, for

An intermediate

type of fading results

when

power

is

nearly equal to power introduced by some mechanism such as diffracin

which the variation

time

is

usually very slow.

Aircraft reflections introduce

of

rapid, intense, and relatively regular fading.

Meteor bursts and some types

ionospheric

propagation add spikes to a paper chart record.

Space-wave fadeouts [Bean, 1954] may represent power fading due

radio energy in

to defocusing of

some regions

of

space, (radio holes) accompanied by a focusing effect and

10-1

signal

enhancement

in other

regions [Doherty, 1952; Price, 1948], or

In

may correspond

to

phase interference fading phenomena.

temperate continental climates, space-wave fade-

outs are likely to occur primarily at night and

most frequently during

the

summer months;

they are

more frequent

at

UHF
in

than at

VHF, and

their occurrence can be correlated with

the occurrence of ground-modified refractive index profiles [Barsis and Johnson,

19621.
Or-

Such fading predominates

geographic areas where layers and ducts occur frequently.

dinary space diversity does not appear to be helpful in overcoming this type of fading.
ing periods of uniform refractive- index lapse rates, prolonged fadeouts are

Dur-

much

less in-

tense or do not exist.

Sometimes those

that do exist are caused by multipath reflecti Dions

in the laps 3se

which arrive

in

such a phase and amplitude relationship that a slight change

rate will cause a large change in the resultant field.

The

latter type can be

overcome

in

most instances by either relocating

General discussions
reports by Bullington
[

the terminal antennas or by the use of space diversity.

of
,

of the
,

time fading
[

VHF

and

UHF
[

radio fields will be found in

[

1950]

du Castel

1957a]

Chernov
[

1955 ],Gr osskopf

[

1958]

Krasil'nikov
of the

1949]

Troitski [1957b], and Ugai

1961]

Silverman

1957]

discusses some

theory

of the

short-term fading

of scatter signals,

Bremmer

1957] discusses signal

distortion due to tropospheric scatter, while

Beckmann [1961b] considers

related depolari-

zation phenomena.

The observed correlation

discussed by Bean
[

of

radio data with various meteorological parameters

[

is

1956,

1961],

Bean and Gaboon

1957]
b,

du Gastel and
c,

Misme

1957]

Josephson and Blomquist [1958], Misme [1958, 1960a,

[I960], and Ryde [1946].

1961], Moler and Holden

Meteorological parameters such as surface refractivity and the

height gradient of refractive index have been found

more

useful as a basis for predicting

In this report

regional changes than for predicting diurnal or seasonal variations.

meteor-

ological information has been used to distinguish between climatic regions, while radio data

are depended on to predict long-term variability about the computed long-term median value
in

each

of these regions.

The basic data used

recorded
in

in developing these estimates of

long-term power fading were

Path

various parts of the world over

more than

a thousandpropagation paths.

distances extend from within line-of- sight to about 1000 kilometers, and frequencies range

from 40

MHz

to 10

GHz.
is

As more

data are collected, particularly in regions where little information

cur-

rently available, these estimates should be re-examined and revised.

Allowances should

sometimes be made
reflections

for predictable

long-term variations

in

antenna gain, interference due to

from aircraft or

satellites,

and variations in equipment performance.

3,

Micro-

wave attenuation due

to rainfall, discussed in section

should be allowed for in estinnating

10-2

the variability of transmission loss at frequencies above 5

of

GHz

The long-term variability

oxygen and water vapor absorption

It is

may be important above 15 GHz.

an estimate
dependence

often desirable to specify rather precisely the conditions for which

is

of

power fading characteristics

desired.

For instance,

the average frequency

of

long-term variability over a given type

of profile

depends critically on the relative dom-

inance of various propagation miechanisms, and this in turn depends on climate, season,

time of day, and average terrain characteristics.

Climatic regions

may

be defined in several different ways

(3)

(1)

by geographic areas
of

on a map,

(2)

by average meteorological conditions,

(4)

by the predominance

various

propagation mechanisms or

by averages

of available data.

In various so-called "climates," at

different times of day or seasons of the year, different propagation

mechanisms may be
of a reof

dominant.

For example,

in a continental

temperate climate the characteristics

ceived signal over a given path

may

be quite different in the early

morning hours

May

than during the afternoon hours in February.

Probably the most serious obstacle

characteristics of tropospheric propagation
cribing layers and ducts.
tion well
It is

to

improvement

of

methods for predicting the

is the

lack of adequate parameters for desto

important to learn how

describe atmospheric stratifica-

enough

to predict the intensity of focusing,

defocusing, and reflection as well as the

of

percentage of time such phenomena are likely to be important for a given region, time
or season, and radio frequency.

day

When

it

becomes possible

to describe the actual

it

inhomo-

geneous, stratified, and turbulent atmosphere

more

adequately,
of

should also be found

of

worthwhile to "mix" predicted cumulative distributions

propagation mechanisms.

transmission loss for a variety

Based on our current knowledge

of

meteorological conditions and their effects on radio

propagation, the International Radio Consultative Committee[

CCIR

1963f

has defined several

"climates."

large amount of data

temperate climates.

available from continental temperate and maritime Other climatic regions, where few data are available, are discussed in
is

annex

III,

The division into climates

is

is

somewhat arbitrary, based on present knowledge

of

radio meteorology, and

not necessarily the

same

as meteorological climates

Haurwitz

and Austin, 1944]

Three important effects

of the

atmosphere on radio propagation have been considThese are: bending

of
of the

ered in defining the various climates.

radio rays, the effects of

atmospheric turbulence, and the degree and stability

atmospheric stratification.

The

amount a radio ray

is

bent and the intensity of atmospheric turbulence are usually correlated

10-3

with the surface refractivity,

The intensity and

stability of various types of stratificaof

tion are often not well correlated with

N
to

Rather stable and extensive layers

for

dry

warm
The

air over relatively cool

moist air tend

form ducts

VHF

and

UHF

commiinication.

sharp moisture discontinuity bends the radio rays, which then tend
interface.

to follow the

dry-moist air

The phenomenon

of

super -refraction, associated with the occurrence of radio

Inland,

ducts close to the surface of the earth, is essentially a fine weather phenomenon.

during fine weather, ducting

is

most noticeable

at night.

Over

the sea super-refraction is

is able to

most marked where

the

warm

dry air of an adjacent land-mass

extend out over a

comparatively cool sea.

Rough terrain and high winds both tend

Areas
of

to

increase mixing in the at-

mosphere and consequently reduce super-refraction.

for elevated duct formation, appear to be

divergence, usually favorable

to

most persistent over ocean areas from 10

[

40

north and south latitude, especially during winter months.

1964]
.

Moler and Holden, I960; Randall,

Elevated ducts are usually less important for tropospheric propagation than those

close to the surface.

World maps
monthly mean

of

minimum monthly mean N

III.

figure 4.

1,

and of the annual range of

figure

31,

may

be useful in deciding which climate or climates are

applicable in a given region.

defined.
In

The boundaries between various climatic regions are not well

be necessary to interpolate between the curves for two clito the

some cases

it

may

mates giving additional weight

one whose occurrence

is

considered more likely.

of

Some important characteristics of the climatic regions for which estimates

variability are shown, are noted below:
1.

time

Continental Temperate characterized by an annual

mean N

of about

3 20

N-

units with an annual range of monthly

in a large land

mean N
of

of 20 to 40 N-units.

continental climate
to

mass shows extremes

temperature in a "temperate" zone, such as 30

60 north or south latitude.

Pronounced diurnal and seasonal changes in propagation are exrange

of

pected

to

occur.

On

the east coast of the United States the annual

N^ may be as

much

air as 40 to 50 N-units due to contrasting effects of arctic or tropical maritime

the

masses

which may move into

2.

area from the north or from the south.

of

Maritime Temperate Overland characterized by an annual mean N^

about

320 N-units with a rather small annual range of monthly

climatic regions are usually located from 20

mean N^

of 20 to 30

N-units.

Such

to 50

north or south latitude, near the sea,

These where prevailing winds, unobstructed by mountains, carry moist maritime air inland. North America and Europe conditions are typical of the United Kingdom, the west coasts of
and the northern coastal areas
of Africa.

10-4

Although the islands

of

Japan

lie

within this range of latitude the climate differs in

showing a much greater annual range

of

monthly mean

about 60 N-units, the prevailing

rugged.

winds have traversed a large land mass, and the terrain

expect
to find

is

One would therefore not

radio propagation conditions similar to those in the United

is
s

Kingdom although
probably more

the annual

mean N

310 to 320 N-units in each location.

2 in this

Climate

is

appropriate than climate

sea areas of Japan.
3.

area.

Ducting

may

be very important in coastal and over-

Maritime Temperate Oversea coastal and oversea areas with the same general
2.

characteristics as those for climate

The distinction made

is that a

radio path with both

2 is

horizons on the sea is considered to be an

Ducting
is

oversea path; otherwise climate

used.
in

rather

common between

the United

Kingdom and

the

European Continent, and

summer

along the west coast of the United States.

4.

Overland characterized by an annual mean N Maritime Subtropical

of
S

about

3 70

N-units with an annual range of monthly

mean N

of 30 to 60 N-units,

Such climates

may

be found from about 10

to 30

north and south latitude, usually on lowlands near the sea

with definite rainy and dry seasons.

for a considerable part of the year,
5,

Where

the land area is dry radio-ducts

may

be present

Maritime Subtropical Oversea conditions observed

Typical of this climate

in coastal areas with the

sanae rajige of latitude as climate 4.

6,

is the

northwest coast
of

of

Africa,

Desert, Sahara characterized by an annual

mean N

about 280 N-units with

year-round semiarid conditions.

80 N-units.
7,

The annual range

of

monthly mean N

may

be from 20 to

Equatorial maritime tropical climate with an annual

of

mean N

of about 360

N-\inits
20

and annual range

to 30 N-units,

Such climates

may

be observed from 20

S latitude and are characterized by monotonous heavy rains and high average

N to summer

temperatures.
Africa.
8.

Typical equatorial climates occur along the Ivory Coast and in the Congo of

Continental Subtropical typified by the Sudcin and

of
s

monsoon climates, with an anto 100 N-units.

nual

mean N

about 3 20 N-units and an annual range of 60

This is a hot

climate with seasonal extremes of winter drought and

20

summer

rainfall, usually located

from

to 40

latitude

10-5

continental polar climate, for which no curves are shown,

to

may

also be defined.
is

Temperatures are low

moderate

all

year round.

The annual mean

Ns

about 310 N-

units with an annual range of monthly

mean N

of 10 to 40 N-units.

Under polar conditions,

which

nn.ay

occur in middle latitudes as well as in polar regions, radio propagation would be

expected to show somewhat less variability than in a continental temperate climate.

Long-

term median values

of

transmission loss are expected

to

agree with the reference values

cr
It is difficult to

predict the percentage of time that high fields due to ducting condi-

tions

may

be expected to occur.
[

Some

of the

better-known maritime areas

of

super-refraction

listed by

Booker

1946] are:
a)

summer months;

British Isles, Atlantic coasts of France, Spain and Portugal and

the

Mediterrean Sea
Sea, Gulf of Aden,

b)
c) d)

Red

Persian Gulf
of Australia,

west and south coasts

New

South Wales and

New Zealand

Pacific coast of United States and Canada, Atlantic coast north of

Chesapeake Bay
e)

coasts of China and Japan

polar regions, although

f)

some sub-refraction may also be expected

of Africa,

all year;

a) b)

west and southwest coast

especially

marked

in

summer

west coast

of India

and the Bay of Bengal except during the south-

west monsoon
c)

northern part

of the

Arabian Sea, especially during the Indian hot

season
d)

north and northwest coasts of Australia except during the north-

west monsoon
It is

apparent that the most intense super-refraction

in trade

is

encountered in a tropical (not equain

torial) climate,

wind areas over the oceans, and

most

of the principal deserts of

the world.

High mountain areas or plateaus in a continental climate are characterized by low

values of

N
s

and year-round semiarid conditions.

The central part

as

of

Australia with

its

hot dry desert climate and an annual range of

much

as 50 to 70 N-units

may

be inter-

mediate between climates

Prediction
of of

and

6.

long-term median reference values

of

transmission loss, by the methods

sections

to 9,

takes advantage of theory in allowing for the effects of path geometry and

radio ray refraction in a standard atmosphere.

tinguish between climatic regions.

Meteorological information
of data available in
of a

is

used to disregions
is

Median values

each

of these
)

are related to the long-term reference value by means

function of an "effective distance,
"

parameter V(50,d

which

defined below.

Long-term fading about

.

the

median

for each climatic region is plotted in a series of figures as a function of d

For regions

where

a large

amount

of

data

is

available, curves are presented that

III

show frequency-related

effects.

(Seasonal and diurnal changes are given in annex

)

for a continental temperate

climate.

10-6

10.

The Effective Distance,

d
e
,

Empirical estimates

of

long-term power fading depend on an effective distance, d

to other

which has been found superior

parameters such as path length, angular distance,

distance between actual horizons, or distance between theoretical horizons over a smooth
earth.

The effective distance make allowance for effective antenna heights and some allow-

ance for frequency.

Define 9
si

as the angular distance

where diffraction and forward scatter transm.ission

a - 9000 kilometers,

loss are approximately equal over a smooth earth of effective radius define d
si

and

as 9000 9
SI

Then:
J

d
SI

65(100/f)3

km

(10. 1)

The value

of d
SI

is

compared with

the

smooth-earth distance, d

so

between radio horizons:

so

^ d-3\l Zh. te

3^^Zh.

re

km

(10.2)

where the
in

effective antenna heights h

f

and h
in

are expressed in meters, the path length d

kilometers and the radio frequency

It

MHz.
is

has been observed that the long-term variability of hourly medians

only slightly greater than d
.

greatest on
is

the average for values of d

The effective distance d

arbitrarily defined as:

=130/fl+(d

-d
si

so

)/dl
-'

km,

for d

so

<d SI

(10.3a)

d
e

= 130 + d

SO

-d si km, ford >d si SO

(10.3b)

10-7

10.2

The Functions V(50, d

e

and Y{p, d
e

The predicted median long-term transmission loss for

characterized by a subscript
n,
is
)

a given climatic region

(50),

related to the calculated long-term reference value

cr

by means of the function V(50, d

e

(50)

=L cr

(50, d

db

(10.4)

where

(50) is the predicted transmission loss exceeded by 50 percent of all' hourly medians n in a given climatic region. V (50, d ) is shown on figure 10. 1 for several climates as a
L,
.

function of the effective distance d

For

the special case of

forward scatter during winter

= L.

afternoons in a temperate continental climate, V(50) =

the calculated

long-term reference value L

cr should be adjusted to the median

and L(50)

.In

all

other cases,
L,

(50) for

the particular climatic region or time period considered.

The function F(9d)

in the scatter

prediction of a long-term reference median contains an empirical adjustment to data.

The

term V(50,

provides a further adjustment to data for

all

propagation mechanisms and

for different climatic regions and periods of time.

In

general, the transmission loss exceeded (100-p) percent of the time

is

L
where Y
For
d

(p) =

(50)

(p,

d
e

db

(10. 5)

(p,

is

the variability of

(p)

relative to its long-term

median value L

(50).

a specified climatic region and a given effective distance, the cumulative distribution of

transmission loss

may

be obtained from (10.

5).

In a continental

temperate climate trans-

mission loss

is

often nearly log-normally distributed.

The standard deviation may be as

much

as twenty decibels for short transhorizon paths where the

mechanisms

of diffraction

and forward scatter are about equally important.

When

a propagation path in a

maritime

temperate climate

is

over water, a log-normal distribution

may

be expected from L,(50) to

L(99.9), but considerably higher fields are expected for small percentages of time when

pronounced superrefraction and ducting are present.

10-8

10.3

Continental Temperate Climate

long-term

predicting Data from the U.S.A., West Germany, and France provide the basis for power fading in a continental temperate climate. More than half a million hourly
of

median values

basic transmission loss recorded over

some two hundred paths were used

temperate

in developing these estimates.

10 Fieure o

shows basic estimates Y

(p)

of variability in a continental

climate.

Curves are drawn for 10 and 90 percent

In the

of all

hours

of the

day for summer, winter

and

all

year for a "typical" year.

northern temperate zone, "summer" extends from

May through October and

A "frequency

"winter" from
f)

November through

April.

factor" g(p,

shown

in figure 10.3 adjusts the

predicted variability to

allow for frequency- related effects:

Y(p) = YJp,d^)g(p,f)

(10.6)

The function
100
2

g(p,

f)

shows a marked increase

at

in variability as

frequency

is

increased above
values at
f)
1

MHz

to a

maximum

400 to 500

MHz

Variability then decreases

till

or

GHz are

similar to those expected at 100

of the

MHz.
of

The empirical curves

g(p,

should not

be regarded as an estimate

dependence

long-term variability on frequency, but rep-

resent only an average of

many

effects,

some

of

which are frequency- sensitive.

The apparent

frequency dependence

is

a function of the relative dominance of various propagation

of day,

mechan-

isms, and this in turn depends on climate, time

terrain profiles for which data are available.

season, and the particular types of

a heavily forested

For example,

low altitude

path will usually show greater variability than that observed over a treeless high altitude
prairie, and this effect is frequency sensitive.
is

An allowance

for the year-to-year variability

also included in g(p,

f).

Data summarized by Williamson

et al

[1960] show that L,(50)

of

varies

more from year

of Li(50)

to

year than Y(p). Assuming a normal distribution

to year,
L.

within each

year and

from year

would be normally distributed with a median equal

to Li{50) for a "typical" year.

Y(p)

is

then increased by a constant factor, which has been in-

cluded in g(p,

f).

Estimates
10.6).

of

Y(10) and Y(90) are obtained from figures 10.2, 10.3 and from equation
to obtain a predicted

These estimates are used

cumulative distribution using the fol-

lowing ratios:
Y(O.Ol) = 3.33 Y(10)
Y(O.l)
Y(l)

Y(99.99)
Y(99.9)
Y{99)

2.90 Y(90)
(10.7)

=2.73Y(10)

=2.41Y(90)
= 1. 82

=2.00Y(10)

Y(90)

10-9

For example, assume

100

MHz,

d
e

=110 km, and

)

a predicted reference

median

= 1.9 db, (figure 10.1), Y (10, d ) = IbO db, so that V(50,d basic transmission loss, L, o e e bcr = -5.8 db, (figure 10.2), g(10, f) = g(90,f) = 1.05 (figure 10.3). Then = 8 db, and Y (90, d
)

Y(10) =

1.

05

(10) = 8. 5 db,
=

and Y(90)
=

= 1.

05

(90)

=-6.1

db.

Using the ratios given above:

Y(O.Ol)

28.3, Y(0.1)
=

23.2,
-

Y(l) = 17.0, Y(10) = 8,5,

Y(99.99)

-17.7, Y(99. 9) =
is

14. 7,

Y(99) = -ll.l, Y(90)

= -6.

1.

The median value

(50)=

bcr

V(50) = 178.

db

and the predicted distribution

L(O.Ol)
L(90)
= =

of basic

transmission loss
=

is;

149.8, L(O.l)
=

154.9, L(l)

161.1, L(10)

169.6, L(50)
=

178.1,

184.2, L(99)

189.2, L(99.9)

= 192. 8

and L(99.99)
of

19 5.8 db.
p,

These values are plotted as a function

a

time availability,

on figure 10.4 and show

complete predicted cumulative distribution

of

basic transmission loss

For antennas, elevated above

munication, less variability
is

the horizon, as in ground-to-air or earth- to- space

com-

expected.
f(9,
)

This
is

is

allowed for by a factor

f(9,

discussed in

annex

III.

For transhorizon paths

f(9i-)

unity and does not affect the distribution.

of elevation
of

For

line- of- sight paths

is

nearly unity unless the angle

exceeds

0.

15 radians.

Allowance must sometimes be made for other sources

power fading such as atten-

uation due to rainfall or interference due to reflections from aircraft that

may

not be adethe distri'

quately represented in available data.

For example,

at

microwave frequencies
is

bution of water vapor, oxygen, rain, snow, clouds and fog

important in predicting longto

term power fading.

sources
j.

Let Y

Y
p..

represent estimates corresponding

each

of

these
i

of variability,

and

let

be the correlation

between variations due

to

sources

and

Then

the total variability is

approximated as:

Y^(P)

^
i=l

Y^^p) +

^
i,j = l

Y.Y.p..

(10.8)

where Y(p)
percent.

is

positive for p < 50 percent, zero for p

3

50 percent, and negative for p > 50

r

Section

shows how

to

estimate

(p)

and Y

(p)

for atmospheric absorption by

oxygen and water vapor, and for rain absorption respectively.

Let

p
la

be the correlation

between variations

of available data

and variations

due to microwave absorption by

10-10

oxygen and water vapor.

p
"^la

Let
ar

be the correlation between

Y and Y

Assuming

that

= 1,

= 0. 5,
ir

and

=0,

Y^(p) = (Y + Y

)^ +

Y^
r

YY
r

(10.9)

This method was used to allow for the effects

of rainfall at
I.

frequencies above
I,

GHz,

for 99

and 99.99 percent

of all

hours in figures

I.

6 to

11 of

annex

Figures
function of d
e

10. 5 to 10. 10

show variability, Y(p) about the long-term median value as a

88-108,

for period of record data in the following frequency groups; 40-88,

108-250, 250-450, 450-1000, and > 1000

values of
dians are:
Y(p)

MHz.

The curves on

the figures
in

show predicted
These me-

for all hours of the year at the

median frequency

each group.

47.1, 98.7, 192.8, 117, 700, and 1500

MHz

for data

recorded

in a continental

temperate climate.
curves in figures 10.

Equation (10.6) and figures 10.2 and 10.3 were used to obtain the
5 to 10. 10.
)

An
annex
III.

analytic function fitted to the curves of V(50,d

and Y

(P.

is

given in

Diurnal and seasonal variations are also discussed and functions listed to pre-

dict variability for several times of

day and seasons.

10-11

10.4 Maritime Temperate Climate

Studies

made

in the United

Kingdom have shown appreciable differences between

Data from maritime

propagation over land and over sea, particularly at higher frequencies.

temperate regions were therefore classified as overland and oversea, where oversea paths
are categorized as having the coastal boundaries within their radio horizons.
tend over a mixture of land and sea are included with the overland paths.

Paths that ex-

The data were divided

into
I

frequency groups as follows:

II

Bands

and

(40-100 MHz)
{

Band

III

50-250 MHz)

Bands IV and V

(450- 1000

MHz)
median value
is

Long-term

variability of the data for each path about its long-termi

shown as

a function of effective distance in figures 10. 11 to 10. 16.

of data for
to 10. 16

Curves were drawn through medians

90, 99, 99.9, 99.99.
6)

each percentage
that
it is

of

time

p = 0.01, 0.1,

1,

10,

Figures 10,11

show

not practical to use a formula like (10.

g(p,
f)

for the

maritime temperate
,

climate, because the frequency factor

Y/Y

is

not independent of d
e

as

it is

in the

o
of

case of the continental temperate climate.

The importance

tropospheric ducting in a mari-

time climate

is

mainly responsible for

this difference. in all

These figures demonstrate greater variability oversea than overland

groups.

frequency

The very high

fields noted at

UHF

for small percentages

In

of

time are due to per-

sistent layers and ducts that guide the radio energy.

cases

of

propagation for great dis-

tances over water the fields approach free space values for small percentages of time.

Curves

have been drawn for those distance ranges where data permitted reasonable estimates.
curve
is

Each

solid

where

it is

well supported by data, and

is

dashed for the remainder

of its

length.

10-12

10. 5

Other Climates

A
1963f
]

limited amount of data available from other climatic regions has been studied, [CCIR
.

Curves shov/ing predicted variability

figures
it

in several climatic regions are

shown

in

annex

III,

III.

25 to lU. 29.

At times

may

be necessary to predict radio performance in an area where few

In

if

any measurements have been made.

such a case, estimates

of variability

are based on

whatever

is

known about

the meteorological conditions in the area, and their effects on radio

If

propagation, together with results of studies in other climatic regions.

a small

amount
of

of

radio data

is

available, this

may

be compared with predicted cumulative distributions

to

transmission loss corresponding

somewhat similar meteorological conditions.

In this

way

estimates for relatively unknown areas

10. 6

may

be extrapolated from what

is

known.

Variability for Knife-Edge Diffraction Paths

The variability

of

hourly medians for knife-edge diffraction paths can be estimated by

considering the path as consisting of two line-of-sight paths in tandem.

The diffracting knife-

edge then constitutes a

common

terminal for both line-of-sight paths.

is

The variability

of

hourly median transmission loss for each of the paths

tion and characterized by the variability functions

computed by the methods

of this sec-

V^(p) = V^(50) + Y^{p)

db

V^lp) = V^(50) + Y^ip)

db

During any particular hour, the total variability function

expected
to

for the diffraction path

would be

be the

sum

of

plus

To obtain

the cumulative distribution of all values of

applicable to the total path a convolution of the individual variables

,

and V

may

be

employed [Davenport and Root, 1958]

Assuming
is the

that

and

are statistically independent variables, their convolution

cumulative distribution

of the variable

The cumulative distribution

of

may

be obtained by selecting

butions of

cumulative distribution

12
(p)

n equally-spaced percentage values from the individual districalculating all possible

and V

(p)

sums V

=V

and forming the

li

2J

of all

values

V,

obtained in this manner.

Another method
ordering
for
of only

of convolution that gives

good results requires the calculation and

n, instead of

2
,

values of V.

As before V

.(p)

and

(p)

are obtained
to the other of these

n equally spaced percentages.

n

Then one

set is

randomly ordered compared

so that the

sums V

= V.

are randomly ordered.

The cumulative distribution

.

sums then provides

desired this
is the

rhe desired convolution of

V
.

and

If

the distribution of

is

convolution of

and -V

10-13

V(50,de)

IN

DECIBELS

_
c=>

'
/

PO

/'
f

1% s. r\ A. \ \U
1

i
ii

>

/
J 0)
CM

K
j
1 /

;
M
/

4. 3. 2. i.

1 1

/
1
i

\
\ \
01

X m c z o

1
MARITIME

MARITIME

MARITIME

CONTINENTAL

1 1 1

^
/

/;' \
J

/
1

V
TEMPERATE TEMPERATE
SUBTROPICAL

/
/

1
\
TEMPERATE

.&

/
1

[/

1 1
OVERSE
OVERLAh

<n

<
01

a
31

Q.

1
OVERL

/
/
1

1
/'
/'

r-^
1

CX
<D

(D

O m H m
JO

i 1

/ /
03
-si

t-

^
-n

<
01

01
<=>

> H m
DESERT

1/ /

fi

01

PI to

3
00
|2:

Q.

\\
1
MARITIME

^^
a. o-

1 1 1

1 1

CONTINENTAL

EQUATORIAL

2 ^

,
i
1

> H
3 m

SAHARA

1
SUBTROPICAL

j
: 1
SUBTROPICAL

1 1 1

-^ <s

~i
j 1

1
OVERSEA

z w

1
1

1
00
1
1
1

1 1
1

<o

10-14

"

Yq (90, de)

IN

db
F^" ipilE- ^^

YodO.dg)

IN

db

E w ?-;-;
::i;

m^:
:i^

\^.\^:i..
'

?5 S7

-T

^
i

-HrT!!^ i*Hi^^!-=
.

'

T--

} i
.!

.^ "nf?^=FT
!

=?t5^
"

=^ -rp ^1-^ n= ^
-

.
..; 1.:
1

::

s.:-

1 .

-l|^
.

:S.
1":
:

s,'
-1 i i

i-

|-:

,^:^

;-i: i

->.
*^
!

.:

.^
--

-i
. ! .

i.-

r_\.

"T
I

"
-:

:;'
:

"
1
.

"
-;::
:*

'-'-.

^
^

^
''

,>^
:

^^
r

.^'
--

:
i ;

^<
1

r-

i.
i

; !
:

*>v.

^--^
:

:
1

p
!

^"r
1 1

-!-

^ "^
;

:!

-'V1.

-i-^::
].
::

=;
.^1":

..1

"^
i

Ifl:;;;|;;::}=.:
:-'':
-

'/-

- 4"- -T^

- -:--i...
'

:^|s -.;

-I
---!-..
1

'""!"":
i
:

,]~
:

4
1

.- '~|

---JT- --4
;: -:

.;:

l>
!

:.-l'^
i \

\y.

^-:!"
:i

i?;

?;
;;;;

r-

-:

\-\

|:

'"

SH 5^;
:

'A:

]-

""i:

J.:

"
:;,.
i'

7"
]..

.:- ;::]

-:^^ -^-::-.
-;;!=;=?
;:::(::.;

-..!-.-

:-!:

>-

/
:-i

:f
:;:

--1

-!
::
1

^::i:i;;

:.]'

c \~^

..

^;: ;-";:

:-:

1
[

;.': M.

r;^

;t;

=1;:^

r^:-^
::

Hi--

2;
4-^

'"""\"''l
.

ifes r-E
::

S ii^r 3B ^S
3S '4111-1
mg=
:'"l^:

;[:;:

-T '!
j

."

-;;l.i/

/ ""
'

::;

.::.;
.-j-...

"\
..

,;
i .

;:
1

...

-r.:

il'::

;:.^,

l4;
-H;;;?;:;:
;-;i
:..

':!

~|":
:zy-l_:
-

V
_4
:

H"-^:
.'^'T

:^i^^

:;i:

^^

J^
1;--.

-if
irrr

i:-i:

ii:.:.

:.

::p.;;

-V f-r
^
/

i:

"!-"

-t"

r='S

';;:

^;;

^irh;

":.!.:"
'

:;_:;

:"|
;

-m o c <
>
'

----

.1

::;il:;i

:::_
:

....
'::':~:

...

^
/;

---

-~- ~~^.
-;--;
-7"-^.

-V
-i---:;;;i

\ \ W
i;

"1

J f
.
i 1 1 .

:Tn""

/
;:
.

m
^

--r- - ---\-

SiD
---^

M1

-':

^'\:

,v
'

'tH;;;; r:

'.

J._

-J

-1

/
/ t
'

'

:.:i::.|::.|^:;
1

);!.
y

:i
.,.;:,l
.

i.

"l:

---i-

^r

p^
--!
1

""-1^"
1 .

^:

1
i

..'
1

/
----f
'.

-4;^
'

o
CO

"!"

\r

/'"
.i.

V
\

--

"---[-"-j-^- ----1 ; ;
.

-::|-

4-
.

l
'

/1

'-.'-}:'''

"

-,:-:
"-":

..

i.

n^
I
1

;.J...
1

::-"l_

--:- -:-'-

-4--i
-

~"
;

:-/
-/

. i

:.

2 z H H ^ m m J) z. ^ -i > n T) ^ H m m
1

-h.:
1

-^z ^H:
"-;

:..::.

I
\:,.\-::

-.l;.;
_

":-""(:

-j~ -
~:
..
1

':":-

1:

f=
-::
:

::::j:;:

?
;:

A
i

\
'-\

::

.j: ;.-

4 I
t

"4'
' i

-J
_.......

-.

:-.:b:
i

^
-1-.:;

4
1
1

|:::: .

p.

5 m H m
to

|\

]-:

i- ..!:
--y:.

-4- ..;:j

7-

m ^ JU n -^ H m
t>

-u

:t:':-

-n <-

!'.:
I

i\
1

;:-^--

4:;-:

'"'1

it
.

-y1
1

....,_..

;-=-:;

~:

-!..:..

\
1
1

-i
-:l:
. 1

~^.
-;}--

~~
.
--i-.
'-i

.;:
-:---:;.' .:
;i :.

-.::.
.

;::|:

1^
1
i

;:!

z ^ H > H m 2
-<

..1

;;..L.:i

i___:__:_

---;..
.

---

--"
--j-----

1
..

:-"--

.._

::
:

:::

t..:

-.
^.1.
-

;--

::'::::

;;;;!^-

-f].
^

--

"

:!.-

--1'.

' i

.:

T'"^
::"{-;:"
:;:.

"l"-

-:r-

-:^
"""

--

^:^

-
r~

----:- ..

::
.;-

."
1
1

..
^

..
:: 1^

:T
~-

'-;:"-

;:

.
1

i(

-"
;

-.-f-;'
1.

:
:;,

z H S C m 33 m 3]
C
-]

o
(/I

-'
-:f:-]..
-

"

--"!

H m
^
.

.
i

.1?
:;

t".-:

^; !
1

!.-:
!

ji
Ix
.
: :

~
...
j .

----

"i
f^
j.
1

.:';
1

"
1

;:

-I-:

'
1

i
1

-.

<
.

-.

4--

:,

ui

iiii

m,

liil

jiiHiif

ak

'111

in

Q
ttii ifii

.:

..

PP

fii ii

5
itli

:
ii
iHi

II

MiiL

\t^\\

MfttI

iltiff

1M

i!li jii

10-15

'

POWER FADING ADJUSTMENT FACTOR

BASED ON
U.S.

g(p,f)

OVERLAND DATA
--=== = 5

""iiS^I'
1.8

^rf
'1

^ =-q=n=='TTrt
3

TTlTtiTTffll

=?t;

_l_4-,_^+^. ffffTO Ij-U

CT

rrrnfl
E1

(-4
:-::::::T:lt
[jj
1

y
pi*
;::1f::

111

!::::::::::
iff

up

iii tW w|

IVz

'"'

1E

E~|=^
E
E
1 1

1.2

=
v

Ti'
:

ttii
illl

1;

E
nv to

^
j _l
-1
i

iiniiiiilit

^^
f.-.v-

|:

SUMMER bS'^
;;;:;:!!
!::::-:

g ALL YEAR
t-

WINTER

J ->
-=:::|::

||||[iil|i ^
itti
400
i

--L

1.0

U^Mrl^'4 ^;lii 1 = ^ =^+Hi'H

1 l^i

==

&j4ffl^
2000

40

200

1000

3000

FREQUENCY

IN

MHz

40

100

200

400

1000

2000

3000

FREQUENCY

IN

MHz

Figure 10.3

10-16

EXAMPLE OF A CUMULATIVE
I4U

DISTRIBUTION L|^(p) VERSUS p

150

160

CO

o _1 o
en en
170

\\,
\

< q: H O

\\

s.

\\

<
03

180

s ^

\ \.

^^ V

190

'^
\.^ ^N
0.1

200
0.01

10

20

30

40

50

60

70

80

90

95

99

99.9

99.99

TIME AVAILABILITY, p

Figure 10.4

10-17

"

Y(p)- VARIABILITY

ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

X**

"'SJte'
en
<=>

f,r
xts"

// // / /
f

,^

r ><

d
yi

/
/
e" ox

^^ \- S^ ^;:--N^ \ \ \^ X
^s
'

N
^

N,
\

\
X

\
)

/
\

'

ro o>
+

/'
r

\
r>o

:/ X
I

+J
,

o
OJ C3

\\

\\

\
\
y

\
\"
o

/ / / // /
/
/

'-o
r j

m m o
m o
CO

y
\

/
/
X

/ /
'

\\
OJ
c=>

\
\
\

/
/
1

H <

o
-Ck.

-c*

\\ \ \\ \
\\

/
/

y
1

/
/

o m
Q.
CD

o
CJl

'

/
1

\\

\\

5 lo m H

en
<=>

CT> <=

c
3)

JAIA
VIEDIAN

cr>
:;

"

5-^
>;
ts>

cz>

SP S = 9
c/,
Y(IO)

1 3
rKUlVl

Yd)
Y(OI)
Y(O.OI),

J C3

3 o<
t

50009
~^
99.99)

^ b ^ O OJ
;<

Iji

o -^
,

ui

,
FREQUENCY
(.-(JIM

OJ

Tl 3D

Y(90)

Y(99)

Y(99.9)

Y(

1
INtN

=
1

CD

lAL

cn

47.1

U.b.A.

CSO

<

B 3 1

tD ID

^^
11

to to (O ID ID
iXI
II 11

MHz

m W
-p

oo

r\)

fM

o
CO

)5
1

q<^
B
D
-'

00 tv

4^ ;o

;< ji ID lO to

i -0

CO cn
(D

(X) <X>

(X)

\oiS)

lO

CD lOlO

10-18

Y(p)

VARIABILITY ABOUT THE LONG-TERM MEDIAN,

I

IN

DECIBELS

-H5

^
/ 5^
tyj

55V.

/if '^/
/ /'
1

M
i-i
XO
" o
"j

\ '^^:: -^
N,
^

\P
<

/Y
'

i X >s

*^

^1
s
V
>

5Otj
o X

y .\ r> \ +
t>

+
^

+ p
.bP^<
^
(

^
3C
c
)

0(S >s^"

>A

r'
'1
1

/?'

XX

'\
3

O aS

"

fty

%/

\V
\
CM

\'\\

**/>

^/
+

+
f

'

"-/

Ir
>o 1,?

\ .V.J I o'

>
X

/
?'
f-

\\
\I

/ /'

/
/

t,
1

f
1

m c>
-\

\\
\

<

\,
\\
\
\
1
' 3

1
)

1
1.
'

/ /
j,
'

-n

m o en H >
o m ^
Q.
CD

c
<B

^^

o
0)

1
'

o m O H C5 m
CT5

-.

o
o X
MEDIAN

7 >

en

lO

c5

- o P

+
1

'

O X

X o - ro OJ g -g c3 b w ^ 5 c O o <^:^;< S 2 ^222
-^
iND LfJ
ji
<-J

-2

Lo
c/,
Y(IO)
Y(O.I)
Y(O.OI),

D
J
Y(l)

,
, ,
Y(99)
Y(99.9)
Y(99.99)
FREQUENCY

>

Y(90)

1 = >
98.7

MHz

(.

D to to to D to JO to
-

m CO to to ^ to o -n ro ro
_;<

cD c;i

oo
r\3

j:^

< ;<
c

to o
;<

"O ^^

^ o o o

to to to

'

'

to to to to J^ to to to to too

10-19

Y(p) =VARIABILITY
I I

ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

CM

ro

_l_

O xoi
J

xoJv

Swl.**

>"/fl

,p

^
^

^
X

/" '^. /
XO

/
/"

/ /
4

//
*

p
<00

\ K'c: c\ \ X k \ \
,.^>

>-.

o X

s.

/I

1
'

O
yi

)
CD

V
m m o -\ < m o
-n
<=5

.; *,^

^oo

Qjo
off

>

J: XX
+

/..

\ \ N

A \\
\ \ \ \ \
\

/-

I>

/
/
I

/ /
/

/ /

\- o
:

A
\
\
I

X
\

t
<^

6
K

/
1

c 1

/ / /
/
-

\\ \
C3

c
a>

H > 2 O m "
CL

in

\ \^
i\

/
'

.C^

cn

cn

p
-J

E
t: ^

cn cn
C=5

|-

O ^ m H m
3)

X
<=> <=>

o X
K

c
(
1

3
UAIA

en

en cn
1

5- OP 2-39C/,
ffi

MEDIAN

+
1 3
rrtUIVI

Y(l)
Y(IO)
Y(O.l)
Y(O.OI),

i>i c
1 1

rsj

iNj '-^

tM
ijJ

ro

^ :i :< o o o E 5" ~ O H m
"^

b O

o g
,
Y(90)

, ,
Y(99)
Y(99.9)
l.,UI\l

oJ

^ u -^

>

FREQUENCY

Y(

1
IINtlM

=
99.99)

2.

lAL
192.8

t (

<^ D to 3 U3 '
'

:i :i
to to to (O ID CD
" "
IN3 (,

n tr m
C/)

U.S.A.

MHz

1 1

"

-n

ro
'f,

o <;<;<;<-1

^ CD ^ w 5 3
to

- o-o
lo

:<

o o o

to

1 1

lO to to io <0 to to to

(0

10-20

Y(p)- VARIABILITY

ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

<=

en
<C3

c^

/
/
c:>
'

/ <. //

^^
/
f
/
/

^ '/
/
1

^ V ^v
'/
^

^
>

/ /
/
1

\ ^
N
X

^ V>\;

>>r^

^
\v S,
) /

s.

\V

/ /
no

/
X

\
ro on

'

l
)

OJ

\\ \ \
\\

\ \
k\>

O
,

/
o
c

/K
X

/ / / "/ /
/
/

"
/
'

f
\

m
-n

/
1
1

m o m
CO

OJ en

\
\
'

\
\

/
/
/

<

/
/
/
1

-c*

\\

\ \
i
'

X^:

-z.

o
p
OD
CD

\ \
\
^

Ql
en o>

-z.

en

O m H m
CO

<=>

o^ C5
<
-i

o
^ ^ <
X
c: JO
UAIA

"

"

MEDIAN

en

<r>

c
l-KUM

Y(I0)

Yd)
Y(O.I)
Y(O.OI),

c=>

^ D o<
I.

5
c

o g X K -< X o o o o
r\)

CM oJ OJ oi

b ^ o OJ

,
Y(99.9)
Y(99.99)

FREQUENCY

Y(90) Y(99)

LUNIINtNIAL

en
<

o
I

417

oo
<=>

D U) to Ig C 3 u) (X> to rn

<

^^ n
IX) (X)

U.b.A.

MHz

- ^^
-

CO
-n

00 ro

i\)

r\)

oo
cr>

J"^

4^

o <<;<:<
D to
CD

to

:i "o

O O O

to to

CO
<=>

:2.

-'

CO en
to to
(X)

1-

'1

<1

)-

to to to to

(X) (X3

to

O
10-21

Y(p)

VARIABILITY ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

>

o kO

A^

Oo
^^

xx
X

vjL

/, / /
f

^ ^ ^'7
V

Vfe A
,

c ^ vT^ ^ \
)

V
o

*^

//
i
1

/
/

3
)

\- **#

A
o/
xl /

V
s

ft

\
)

\ \\
)
/

\
/
/

'
I

o
r>

/
>

/
r/

\ \

\\
\

\
\
\

I
'\
\
o

ct

7
/
/
,

/
f

1
y

/
f

\}

\
m o H
< m o H > 2 o m
C/5

en

\
y\
\
1

/
/
i

/ / /
/

C3
<=>

\ ,\
1

/
/

\
\
'

-n

/
en
-b
/

^ a

\
o
/

,\
en

O
(0

Q.
CD

o
' '

Ps

en en C3

o ^ m H m
XI en

C3 C3
>
c

o
-9

<* er.
(

s.

>

c
CS

g m 2
;::

c3
t
^

- p p

C]

^
)

T
cji

^
"D

- O O X

S
1 )

o'

<K^-<a3

OJ

:<:<

5000^
3.

^ ^

rn

-^

:<

^
-g

z
1

03U)U3"^r'

H m

(O

-H t
<"

D U) 05 U3 3 CD CD to
' -'
'

(-

N p

m
en
"n

to ID K) ID

t"

"

M 00 ji ro
(X)

.:$

T3

o <;<;<;<
p:
<-

ID ID

5
a>

(D ID lO iO ID CO 10 to CD

'D

10-22

Y(p)

VARIABILITY ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

V
/

^u#
.y

/^ X
o
0?0 ^35

t^y

$^ s^
+

N-

en
c=>

^ '7 / /,
w
\
1

7
**

f X X

O O

>l
1

o o
o

A \ V \
\

^\ \
S,

\.

>

\
)
/
f

>

ro

O O O

o
<=>

1 \
\
y.

**

po en

\^
\\

o O

'

\ \"\l-u

:f
1

J 7 /
o

1
/

CKl

c^

o
m
-n -n

\ \\\ \^

/
>.

/ 1
/
1

^ H < m -o C3 o O
(/)
fl

OJ en

\
\

l-v

fto

o ^<'>

\ \y

i
1

\\
\\\

o
c
^

H T> 2
C-)

-C^

C3

\v
V
'

'

m
Q.
Ol C3

<e

p O

<D

z
7^

tn <=

cn

r~

o ^ m H m
3) CO

C5

o a
o X

UAiA

MEDIAN

+
rrcuivi

<35 cz>

P 2-3P
=;

E^l

'

ro ro Ol

^ o g
Tl

Y(l)
Y(iO)
Y(O.I)
Y(O.OI),

c-">

d<
1

X 33 ~ ~ "^ m o o (-, o 5 5
-C
.~

O ^

,
Y(99)
Y(99.9)
Y(99.99)

FREQUENCY

^..uinj

C>J

CaJ

Y(90)

minjuin

i
al

o
c

1500

<^ D CD 3 CD ~
-'
1

.:<

Ji
CD CD CD
II

t3D

1 II

^-o
II

CD CD CD

^ c m
1/3

u.s.a.

MHz

-n

ro j^

po

ro
CD

o"
c

<;<;<;<

CD o

;^

T5

D j:D

O O O

CD CD

m
:?

CD CD CD CD CD (X) CD CD CD CD

O P b 10-23

Y(p)

VARIABILITY ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

c=>

c:3

c3

CD

c:

c3

c3

c3

C3

L
cn C3
?

O C3
_
<=>

/ <-;

^ /;
o

# /
o o

i^ V
i9

*i
}%

^- V >C? i^ >^
1>,

/
ro
<=>

/y f
1

X
.

/"
f

/
/

Y
lo

f^ 1.% N
r

"

3
'

/o
io
i

:/

X V
/ +
X
1

> 3J
m m u m > H m o
^

111
I

ro
<=>

\^

\
I

\ \
OJ

^
\ \
\
\
1

o
m
-n -n m o H

\ \
^

\ \

I
1

1
:

J J T
K
E

3D

OJ c^

\^ y
\ \

\\
y
\

m
ri

<

o o

.:k

Z>

\ \ \
I
1

H > o S
Q.
cn
<=>

\
1

-pk

1
1

m o < m
> o
r

cn
c=>

3*

o m H m O
<7>

>

XI CO
DATA

> O M > o
C/)
1=1

DD

o>

-1
FROM

Yd)
Y(IO)

-^

Y(O.I)
Y(O.OI)

o
UNITED

C3

,
Y(99.99)

Y(90)

Y(99)

Y(99.9)

o o

KINGDOM

ao
<=>

OS CD

CO CD

CO cn
U3

iHDlX>tO to toio

10-24

Y{p)= VARIABILITY

ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

cn

o
C3

^ '^:
y,
/

>^ 7 V <^ C^v.v^ \^^^1 S

5**ss,^
'
1

A M

N,

DATA

~^\~^
Yd
Y(IO)

FROM
Y(O.I)
Y(O.OI)

<I3

/
/
1

/
\
J

/
1
1

\
".

\ \ A "\ \
I
1

)
, ,
Y(90) Y(99)

, ,
Y(99.9)

ro c?

Y
(99.99)

UNITED

1
1
'

.\
no

"o
\

O
OJ

\ \ \ \ \
n\ V

X
/

KINGDOM

\
\

o
m
OJ
<z>

\\
\ <>
\\ \
V

\.
\

1
><

w
\

v\

\
3

<

c
1 1

\
\
\
1
i

-Ch

CD

\\\
\\\

-c*

\M
\\\ \\
\

\
\
1

o m
Q-

o>
\
\

en

o
IVI

\
en

\
\

o m H m O
<=) <T>

\ \

\
1

\ \ \

XI
C/) CT>

\
\ \

o>

<=>

\ \
\ 1

\
\

o
c

\
\

\ \
\

\
\

\
\

o
CD

Oo

CO o>

\
1

\
\

\
CO en
ID
\

\
\

10-25

Y(p)

VARIABILITY ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

>

c>

c=

cz>

c^

<=>

C:>

<z

c3

c3

c3

<=

C3

<=:>

_
CI>

cs

/: // /^

A
1
'

4'//I \ \^ \o ///
o

J\ Y

iW^ \^^
v^
\
O

7
/
D

\
> C

v_
>

NN
9
1

^0
\

\
+

c '>

\*
o +
\
1

ro
<:=>

/
\
\

\
a

\
-

&

1
1

'

ro
<=>

y
\ \
\

OJ
<=>

=1

c
c

\ \

\
\

&

\ \
V
"

m
Oo C5

\ \ \ \
\ v\ \

\'

m o
C/5

<

-b'

c=>

\\
^\\
I

\ ^
\
c-

\ \
V

\
o O

\ \
\

\
1
It

4::*

o m
Q.

O
cn
<-:>

\>
\
\

\
\
\

\\

\
y

\
y

\
"O
H

\
p

O m H m
XI
C/)

cn CD
1

1 1

\
\

05
<=3

SD to to U3

cr> <=>

___j

c=>

DATA

"

FROM

<=>
Y(IO)

Yd)
Y(O.I)
Y(O.OI)

oo
<=>

, ,

Y(90)

Y(99)

Y(99.9)

Y
(99.99)

UNITED

oo C3

KINGDOM

!^
<=:>

CO cn

10-26

>

Y(p)- VARIABILITY

ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

A # 4 ^
\o
o
i

.+

/^
( \
(

-^ 'y**

<
<;

/
I

/ 1
\

r^^
r

%
<s

><*> +
+
V

-^
"-,

o > H t>
3)

"^

K>

\
X

+\ \

K ^ K ^
(10)
(0.1)
(0.01)

OM

,
Y(90) Y(99)

,
Y(99,9)

UNIl

A
\^
\

Y
(99.99)

FED

\
X

\ \

/
\
' 1

\
\
\

+
KINGDO

\
I

\ \ On
m m C3 o en H < m -& C3 u O CO H >
-z. r->

,\
^

\ \
\ \ \
\
'

\ \

\i v\ \ A^
\\
\
\

\ \ \ \ \ \
\
X
\

ii

\
1

-n

01
<=>

ll
)

\
\ \
1

m
<B

Q.
fD
1

\
\

(>
-t^

C^
c=>
-z.

\ \
\
I

\
\

7^

C/1

5 ^ m H m M
C/)

\ \
1

\
\

>

\ \
CT>
1
1

<=>

\
I

\
\
\

05
\
\

\
\
\

1
\ \
>

\
\ \

\
1

\
\

\
]
\

1
J

\
\
1

\ \

1
1

\
1 1
1

\
\

6 b

1
\ 1
1 1 1
1

\
\

1 1

\
\

-S-

to to io S" 10 u) U3 IC to

10-27

Y(p)

VARIABILITY ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

cn

o c
Ol

'^ /
o
ro o>
/

4? 71
1

^\^v.v ^ :'<:^v:^ ^

/ //
/ /
/ (
\

///
1
1

1
1

\
1

^^ >
\
>
<

1
c

^ ^ N \^ X
3*

^
+

\
+

\
V

"

\
t
4-

>
+

/
ro
<=>

1 1 J_
i

^
D +

m
\
\

\
o
M
>

<:

<=>

\
1

m o
H < m o
o
Z!

-n -n

\
Cn C5

\
N

-Ck

C3

W' \\
\\
\
1

\\

\.
\
\

\
\
1

\ \

\
V

m > H m o
!-

O
f7)
1

\
1
1

CO

^ z
Q.
fl>

\ \
\
1

\ \ \
1,

.Cik,

\V

<=>

_ O
oi

\
\ \

\
\

cn

o
cn cn
<:>
1

\
\

\
\
V \
\

\
c

> H m o <m 33 n > z u "

\
CD

H m ^

o
m x
s=

-0

\
\

5 m H m
CO

05 C^

\
\

\
\

\
\

o^ C3

\
\

\ \
\

~p b

\
1

> -n > z o o CO z o ^ > z o M

_^ c^
ID IX ID ID Co (X <^

o
o O
ID

V
DATA

^ cn
O O o o s
1

-o
FROM

J
c=>

"

"O
Y(IO)

Y(

1 )
,

Y(O.I)
Y(O.OI)

, ,

oo

o
CD
<=>

Y(90)

Y(99)

Y(99.9)

Y
(99.99)

UNITED

N
KINGDOM

<x>

o
CO cn

10-28

Y{p)

VARIABILITY ABOUT THE LONG-TERM MEDIAN,

IN

DECIBELS

on

o>

4 // /,
J

m^

L 1- \

o
>

// /

y/
1
'

\ >^" "^r^^ <^ \ *\ \ N^N

\

--^

/, //

\ X ^ \>
V

'^.

"fc^

"^^

^s
"s

ro
<=>

/
/

/
/
1

\
\
\

\
o
"

>
33

rs:>

o
I
1

<=>

\
\

\
1

t>

m
Tl

'\

\
OJ
<=>

\
\
,

/ 1
\

\
\
]

\
\
\

m -n -n m OJ o s
H < m o
-p

VA
\

V \
\

\ \
\

1
o

1
I
I

m
> H m o
\
\

\\ \
^S)

\ \

\ \
K
1
1

\ \

c:>

(J)

o m

<n

.^ CD

^
\ \
\
'

\ \
1

\ \
1

Q.
CD
CJi

\
\
1

o
\
C_n

\
\

\ \ \

o m m
(J)

<=>

\
\

\
\

'>

cr>

m m o s < m T) ^ o CO $ rn m ?* ^ CD -n > > z '^ o

=

\
<=>

\
\
'

^
C5

03
i

\
\
t

> o
\

?=j

"

o > -H >

\
\
1

o
\
cA

o_

~7

g S O
oo

_,
Yd)
Y(IO)
Y(O.I)
Y(O.OI)

o 2

\
\ \
]

1 1

Y(90)

Y(99)

Y(99.9)

2 Y m

O 5 O O
tM

en

\
(99.99)

o
1

o o o
2
1

\
\
\
1

CO
c=>

CO en

\
10-29

11.

REFERENCES

The references given below include only selected papers referred to in the text of this report. A comprehensive survey of work in the field of tropospheric propagation, and an extensive bibliography will be found in the following report:
Shkarofsky, I. P. (March 1958), Tropospheric scatter propagation. Res. Rpt. No. 7-200-1, RCA Victor Co., Ltd. Res. Labs, Montreal, Canada.

Four recent bibliographies are:

Abbott, R. L. (Nov. I960),

Bibliography of tropospheric radio wave scattering,

NBS

Tech.

Note No. 80.

and E. R. Westwater (Dec. 1961), Bibliography of Abbott, R. L. sions by atmospheric gases. Private Communication.
,

microwave thermal emis-

Nupen, Wilhelin. (1964), Bibliography on propagation of radio waves through the troposphere, NBS Tech. Note No. 304.
Dougherty, H. T. (Aug. 1964), Bibliography of fading on microwave line-of-sight tropospheric propagation paths and associated subjects, NBS Tech. Note No. 302,

Anderson, L.

and E. E. Gossard (Oct. 1953a), The effect of the oceanic duct on microJ. wave propagation, Am. Geophys. Union Trans. 34, No. 5, 695-700.
,

Anderson, L.
effect on

J.

UHF, Proc. IRE

and E. E. Gossard (Jan. 1953b), 136-139. 1, 41^, No.

Prediction of the nocturnal duct and

its

Arons, L. D. (Oct. 19 56), An analysis of radio-wave scattering in the diffraction region, Cornell University E. E. Report 312.

Artman,

and J. P. Gordon (Dec. 1954), Absorption of microwaves by oxygen in the J. O. millimeter wavelength region, Phys. Rev. 96, No. 5, 1237-1245.
,

Bachynski, M. P. (1959), No. 2, 308-335.

Microwave propagation over rough surfaces,

RCA

Review

20,

Bachynski, M. P. (July-Aug. I960), Propagation at oblique incidence over cylindrical obstacles, J. Res. NBS 64D (Radio Prop. ), No. 4, 311-315.
Bachynski, M. P. (March 1963), Scale model investigations of electromagnetic wave propagation over natural obstacles, RCA Review 24, No. 1, 105-144.

Barghausen, A. F.
1963),

F. O. Giraud, R. E. McGavin, S. Murahata, and R. W. Wilber (Jan. Equipment characteristics and their relation to system performance for tropospheric communication circuits, NBS Tech. Note 103.
,

Barsis, A. P., and M. E. Johnson (Nov. - Dec. 1962), Prolonged space -wave fade-outs in tropospheric propagation, J. Res. NBS 66D (Radio Prop. ), No. 6, 681-694. Barsis, A. P., and R. S. Kirby (Sept. - Oct. 1961), observed on a long knife-edge diffraction path, 437-448.

VHF
J.

and UHF signal characteristics Res. NBS 65D (Radio Prop. ), No.

5,

Barsis, A. P., K. A. Norton, P. L. Rice, and P. H. Elder (Aug. 1961), Performance predictions for single tropospheric communication links and for several links in tandem, NBS Tech. Note 102. (See also IRE Transactions on C ommunication Systems CS-10, No. 1, 2-22, March 1962).

11-1

Batchelor, G. K. (1947), Kolmogoroff's theory of locally isotropic turbulence, Proc. Camb. Phil. Soc. 43, 533-559.
Batchelor, G. K. (1953), Press).

The theory of homogeneous turbulence, (Cambridge University

Bean, B. R. (May 1954), Prolonged space-wave fadeouts at 1,046 Cheyenne Mountain propagation program, Proc, IRE 42, No. Bean, B. R. (1956), Some meteorological effects on scatteied Trans. Comm. Syst. CS4(1) 32-38.
, ,

Mc observed
5,

in

848-853.

VHF

radio waves,

IRE

Bean, B. R. (July-Aug. 1959), Climatology of ground-based radio ducts, (Radio Prop. ), No. 1, 29-34.

J.

Res.

NBS 63D ~

Bean, B. R. (1961), Concerning the bi-exponential nature of the tropospheric radio refractive index, Beitrage zur Physik der Atmosphare 34, No. 1/2, 81-91.
Beaji, B. R. , and R. L. Abbott (1957), Oxygen and water vapor absorption of radio the atmosphere, Geofisica Pura e Applicata - Milano 37, 127-134.

waves

in

Bean, B. R. and B. A. Cahoon (Nov. 1957), The use of surface observations to predict the total atmospheric bending of radio rays at small elevation angles, Proc. IRE 45, No. 11, 1545-1546.
,

Bean, B. R.

J.

D. Horn, and

A. M. Ozanich, Jr.

(Nov.

of the radio refractive index for the United States

I960), Climatic charts and data and the world, NBS Monograph No. 22.

Bean, B. R. J. D. Horn, and L. P. Riggs (Oct. 1962), Tech. Note 98.

,

Synoptic radio meteorology,

NBS

Bean, B. R.
index,

and G. D. Thayer (May 1959), Proc. IRE 47, No. 5, 740-755.

,

Models

of the

atmospheric radio refractive

of

Beard, C.
ocean,

I.

IRE Trans. Ant. Prop. AP-9

(September 1961, Coherent and incoherent scattering 470-483.

,

microwaves from

the

Beard, C.

I., I. Katz, and L. M. Spetner (April 19 56), Phenomenological vector model of microwave reflection from the ocean, IRE Trans. Ant. Prop. AP-4, No. 2, 162-167.

Beckmann,
Acta,

P. (1957),

A new

approach
,

to the

problem

of reflection

from

rough surface.

Tech. Ceskosl. Akad.

311-355; see also pp. 323-335, (1959).

Beckmann, P.

(I960), A generalized Rayleigh distribution and its application to tropospheric propagation, Electromagnetic Wave Propagation, (Symposium, Liege, 19 58), (Academic Press, London, 445-449).

Beckmann,

P. (1961a), The statistical distribution of the amplitude and phase of a multiply scattered field, Inst. Rad. Eng. andElec, Czechoslovak Akad. Sci. Paper No. 18, See also NBS Jour, Res. 66D (Radio Propagation), pp, 231-240, 1962.
, ,

Beckmann, P. (1961b), The depolarization

surfaces, Inst, Rad. Eng. and Elect,
,

of

electromagnetic waves scattered from rough Czechoslovak Akad, Sci,, Paper No. 19,
its

Beckmann, P, (September 1964), Rayleigh distribution and Res, 68D (Radio Science), No. 9, pp. 927-932.
,

generalization,

NBS

Jour,

Beckmann, P. and A. Spizzichino

The scattering of Electromagnetic waves from 1963) rough surfaces, International Series of Monographs on Electromagnetic Waves, Vol, (Pergamon Press, New York, N, Y.).
(
,

4,

11-2

Biot,

M. A. (Dec. 1957a), Some new aspects of the reflection of electromagnetic a rough surface, J. Appl. Phys. 28, No. 12, 1455-1463. M. A. (Nov. 1957b),
J.

waves on

Biot,

Acoust. .Soc.

Am. Z%

Reflection on a rough surface No. 11, 1193-1200.

from an acoustic

point source,

Booker, H. G. (1946), Elements of radio meteorology: How weather and climate cause \inorthodox radar vision beyond the geometrical horizon, J. Inst. Elec. Engrs. (London) 1, 69-78. 9_3. Pt. in-A, No. Booker, H. G. and J. T. de Bettencourt (Mar. 1955), tropospheric scattering using very narrow beams,
,

Theory of radio transmission by Proc. IRE 43, No. 3, 281-290.

Booker, H. G. and W. E. Gordon (Sept. 1950a), the troposphere, J. Geophys. Res. 55, No.
,

3,

Outline of a theory of radio scattering in 241-246; see also Proc. IRE 38, No. 4,

401, (April,
,

1950b).

Booker, H. G. and W. Walkinshaw (April 1946), The mode theory of tropospheric refraction and its relation to waveguides and diffraction. Report on Conference on Meteorological Factors in Radio Wave Propagation (The Phys. Soc. and the Royal Met. Soc.
,

London),

80-127.

Bray, W.

F. Hopkins, A. Kitchen, and J. A. Saxton (Jan. 1955), Review of longJ. distance radio-wave propagation above 30 Mc/s, Proc. lEE, Paper No. 1782R, Pt. B, 102 , 87-95.
,

Bremmer,
Co.
,

H. (1949),

Terrestrial radio waves; theory of propagation,

N. Y.
).

(Elsevier Publishing

Amsterdam and New York,

H. (Sept.
3,

Bremmer,
IS,

1957),

Distortion in tropospheric scatter,

Phillips

Telecomm. Rev.

No.
H.

137-154.
1959),

Bremmer,

(May

On

the theory of the fading properties of a fluctuating signal

im-

posed on a constant signal, NBS Circular 599.

Bugnolo, D. S. (Jiily 1958), Multiple scattering of electromagnetic radiation and the transport equation of diffusion, IRE Trans. Ant. Prop. AP-6, No. 3, 310.
Biollington, K. (Jan.

1950),

Radio propagation variations at

VHF

and UHF, Proc. IRE

38,

No.

1,

27-32.
1955),
10,

Bullington, K. (Oct.

Proc. IRE 43, No.

Characteristics of beyond-the-horizon radio transmission, 1175-1180.

Bussey, H. E. (July 1950), Microwave attenuation statistics estimated from rainfall and water vapor statistics, Proc. IRE 38, No. 7, 781-785.

CCIR
CCIR

(1955), Atlas of ground wave propagation curves for frequencies between 30 300 Mc/s, ITU, Geneva. (1959),

Mc/s and

Atlas of ground wave propagation curves for frequencies between 30 and 000 Mc/s (Vertical polarization only; prepared by the Radio Research Laboratories, Ministry of Postal Services, Tokyo, Japan, January 1958), ITU, Geneva.
10,

CCIR

(1963a),

the Xth

The concept of transmission loss in studies of radio systems, Dociiments of Plenary Assembly, ITU, Geneva, Vol. Ill, Recommendation 341, 29-31.

11-3

CCIR CCIR CCIR

(1963b), Transmission loss in studies of radio systems, Assembly, ITU, Geneva, Vol. Ill, Report 112, 84-89. (1963c),

Documents

of the

Xth Plenary-

Optimum use
Ill,

ITU, Geneva, Vol.

of the radio spectrum. Resolution 1, 111.

Documents

of the

Xth Plenary Assembly,

(1963d), Line frequencies or bands of interest to radioastronomy and related sciences, in the 30 - 300 Gc/s range arising from natural phenomena, Documents of the Xth

Plenary Assembly,

ITU, Geneva, Vol. IV, Report 223, 304-307.

CCIR CCIR CCIR CCIR

(1963e), Reference atmospheres, Documents of the Xth Plenary Assembly, Geneva, Vol. II, Report, 231, 74-75.

ITU,

(1963f), Estimation of tropospheric-wave transmission loss. Documents Plenary Assembly, ITU, Geneva, Vol. II, Report 244, 191-213.

of the

Xth

(1963g),

Documents
(1963h), 1000 Mc/s

Propagation curves for VHF/UHF broadcasting in the African Continent, of the Xth Plenary Assembly, ITU, Geneva, Vol. Ill, Report 240, 143-181.

and UHF propagation curves for the frequency range from 40 Mc/s to Broadcasting and mobile services. Documents of the Xth Plenary Assembly, ITU, Geneva, Vol. II, Recommendation 370, 24-36.
-

VHF

CCIR

Communication satellite systems -frequency sharing between communication systems and terrestrial services, Documents of the Xth Plenary Assembly, ITU, Geneva, Vol. IV, Report 209, 221-232.
(19631), satellites (1963J),

CCIR
CCIR

Plenary Assembly,

Influence of the atmosphere on wave propagation. Documents of the Xth ITU, Geneva, Vol. II, Report 233, 76-120.
of the

(1963k), Propagation data required for radio relay systems. Documents Plenary Assembly, ITU, Geneva, Vol. II, Report 242, 182-187.

Xth

CCIR

Fading of signals propagated by the ionosphere. Documents (1963i), Plenary Assembly, ITU, Geneva, Vol. II, Report 266, 327-334.

of the

Xth

CCIR (1963m),
CCIR

Terms and
I,

definitions,
2 39.

Dociiments of the Xth Plenary Assembly,

ITU,

Geneva, Vol.
(1964), the Xlth

Report 321,

Optimum use of the radio frequency spectrum. Document being prepared for Plenary Assembly, in accordance with Resolution 1 of the Xth Plenary Assembly, ITU, Geneva, Vol. IE, 111.

Chernov, L. A. (Jan. - Jiine 1955), Correlation of amplitude and phase fluctuations for wave translation in propagation in a medium with random irregularities, Akust. Zh. ~ 1, 89; Soviet Phys. - Acoust. No. 1-2, 94-101.
l_.

Christiansen, W. N. (1947), less Australia ] T_, No.

Rhombic antenna arrays,

4,

A. W. A. Tech. Rev.

[Amal. Wire-

361-383.

Clemow,

D. B.

munications,

and E. H. Bruce-Clayton (Jan. 1963), Brit. IRE J. 5, No. 1, 17-32.

Long range

VHF

air/ground com-

Cozzens, D. E. (June 1962), Nomograph for determining paraboloidal gain as a function of feed pattern and angular aperture. Microwave J. V, No. 6, 58-59.

Crawford, A. B. and D. C. Hogg (July 1956), millimeter wavelengths. Bell Syst. Tech.
,

Measurement
J.

of

atmospheric attenuation

at

35,

907-916.

11-4

Crawford, A. B. D. C. Hogg, and W. H. Kummer (Sept. 1959), Studies in tropospheric propagation beyond the horizon. Bell Syst. Tech. J. 28_ No. 5, 1067-1178.
,

Crichlow, W. Q. D. F. Smith, R. N. Morton, and W. R. Corliss (Aug. 1955), Worldwide radio noise levels expected in the frequency band 10 Kc to 100 Mc, NBS Circular 557.
,

Crysdale, J. H. (July 1958), Comparison of some experimental terrain diffraction losses with predictions based on Rice's theory for diffraction by a parabolic cylinder, IRE Trans. Ant. Prop. AP-6 No. 3, 293-295.
,

Crysdale,

J.

H.

J.

W. B. Day, W.

1957),

An experimental

dominating ridge, IRE

, .

S. Cook, M. E. Psutka, and P. E. Robillard (April investigation of the diffraction of electromagnetic waves by a Trans. Ant. Prop. AP-5 No. 2, 203-210.
,

and W L. Root (1958), An introduction to the theory or ranaom sigDavenport, W. B. Inc., New York, Chapters, nals and noise, McGraw-Hill Book Co.
,

dejager, C. (1952), The spectrum of turbulence in the earth's upper atmosphere, Liege 12, 223-252. Soc. Roy. des Sci.
,

Mem.

Dickson, F. H. J. J. Egli, J. W. Herbstreit, and G. S. Wickizer (Aug. 1953), Large reductions of VHF transmission loss and fading by the presence of a mountain obstacle in beyond-line- of- sight paths, Proc. IRE 41, No. 8, 967-969. See also subsequent correspondence by Crysdale and rebuttal by Dickson, et al., in Proc. IRE 43, No. 5, 627-628 (May 1955).
,

Doherty, L, H. (Sept. 1952), Geometrical optics and the field at a caustic with applications to radio wave propagation between aircraft, Cornell University School of Electrical Engineering Research Report EE-138.

Dolukhanov, M. P. (1957), Investigations into the propagation of radio waves over the earth's surface in the USSR., Radio Engr. and Electronics (USSR) 2, No. 11, 39-61.

Domb,

C. and M. H. L. Pryce (Sept. 1947), The calculation of field strengths over a spherical earth, lEE 94, Part III, No. 31, 325-339.
,

Dougherty, H. T., and L. J. Maloney, (Feb. 1964) The application of diffraction by convex surfaces to irregular terrain situations, J. Res. NBS 68D (Radio Science), No. 2, 239-250.
duCastel, F. (May 1957a), Different types of fluctuations of tropospheric fields and their physical interpretation, L'Onde Electrique 37, No. 362, 501-506.

duCastel, F. (Nov. 1957b), The use of ultra short waves for long distance telephone links in Africa (Results of Tests in the Cameroons), L'Onde Electrique 37, No. 368, 10251035. duCastel, F. (Nov. -Dec. I960), Experimental results from transhorizon tropospheric propagation, Ann des T^^comm. _1^, No. 11-12, 255-259. duCastel, F., and P. Misme (Nov. 1957), Elements of radio climatology, L'Onde Electrique 37, No. 368, 1045-1052.
duCastel, F,, P.

Misme, and J. Voge (March 1958), Reflection of an electromagnetic wave from an atmospheric layer with variable index of refraction, C. R. Acad., Sci. Fr.
246
,

No.

12,

1838-1840.

duCastel, F., P. Misme, A. Spizzichino, and J. Voge (1962), On the role of the process of reflection in radio wave propagation, J. Res. NBS 66D (Ra. -S. Yt^-.^, _Io. 3, 273-284.

On the climatology of groiind -based radio ducts and associated Dutton, E. J. (Jixne 1961), fading regions, NBS Tech. Note 96.
11-5

Dutton, E. J., and G. D. Thayer (Oct. 1961), Techniques for computing refraction of radio waves in the troposphere, NBS Tech. Note 97.

Fengler, G. (1964), Untersuchungen der elektromagnetischen wellenausbreitung im 500 MHZ-bereich iiber land tonter besonderer beriicksichtigung der meteorologischen, Berichte des Instituts fiir Radiometeorologie und Maritime Meteorologie an der Universitat Hamburg, Report No. 8.
Fengler, G.
,

J.
fiir

Jeske, and G. Stilke,

9.

Instituts

Radiometeorological papers II, Berichte des Radiometeorologie und Maritime Meteorologie an der Universitat Hamburg,

Report No.

Florman, E. F. , and J. J. Tary (Jan. 1962), Required signal-to-noise ratios, RF signal power, and bandwidth for multichannel radio commionications systems, NBS Tech.
Note 100.
Fok, V. A., L. A. Vainshtein, and M. G. Belkina ( 1958), Radiowave propagation in surface tropospheric ducts. Radio Eng. Electron. (USSR), 3, No. IZ, 1-27.
Friend, A. W. (June 1945), A summary and interpretation of ultra high frequency wave propagation data collected by the late Ross A. HiiLl, Proc, IRE 33, 358.
Friis, H. T. A. B. Crawford, and D. C. Hogg (May 1957), gation beyond the horizon. Bell Syst. Tech. J. 36, No.
,

A
3,

reflection theory for propa-

627-644.

Furutsu, K. (1956), On the multiple diffraction of electromagnetic waves by spherical mountains, J. Radio Res. Labs., Tokyo 3, 331.

Furutsu, K (1959), Labs., Tokyo

Wave propagation over an irregular

4,

terrain,

I,

11,

III,

J.

Radio Res.

135,

349(1957),

and

6,

71(1959).

Furutsu, K. (Jan. -Feb. 1963), On the theory of radio wave propagation over inhomogeneous earth, J. Res. NBS 67D (Radio Prop. ), No. 1, 39-62.

Grosskopf,

On the existing condition of research in the realm of tropoJ. (June 1956), spherically scattered radiation, Nachrtech. Z. 9, No. 6, 272-279.
J.

Grosskopf,

(Nov. 1958),

Some remarks on
11,

meter range,
Gvmn, K. L.
S.
,

Nachrtech. Z.

the analysis of fading in the meter and deciNo. 11, 577-586.

of precipitation

particles,

Qixart.

and T. W. R. East (Oct. 1954), The microwave properties 522-545. J. Roy. Meteorol. Soc. (London) 80,

Harper, A. E. (1941),

Rhombic antenna design,

(D.

van Nostrand Co.

Princeton, N.

J.

).

Hartman, W.

Path antenna gain and comments on "Properties of 400 J. (May 1963), long-distance tropospheric circuits," Proc. IEEE 51, No. 5, 847-848.
in

Mcps

Hartman, W. J., and R. E. Wilkerson (Nov. - Dec. 1959), Path antenna gain ential atmosphere, J. Res. NBS 63D (Radio Prop. ), No. 3, 273-286.
Hathaway,
S.

an expon-

D.

and H. W. Evans (Jan. 1959),

implications affecting relay system engineering.

Radio attenuation at 11 kMc and some Bell Syst. Tech. J. 38, No. 1, 73-97.

Haurwitz, B. and

J.

M. Austin
S.

1944]

Climatology, McGraw-Hill Co. Inc.,

New York.

Hay, H. G.

and R. uniform height,

,

Unwin (Dec. 1952), Tropospheric wave propagation in a duct of nonPhys. Soc. London Proc. 6^, No. 396b, 981-989.
11-6

television field strengths at ultra-high Head, H. T. (June I960), The influence of trees on 1016-1020. No. 6, 48, IRE Proc. frequencies,
statistical and isotropic turbulence. Proc. Heisenberg, W. (Dec. 1948), On the theory of 402-406. A 19 5, Roy. Soc. London Herbstreit, J. W., and P. L. Rice (Sept. 1959), Survey of Central Radio Propagation Laboratory research in tropospheric propagation, 1948-1956, NBS Tech. Note No. 26.

Hirai, Masaichi

(May 1961a), Multipath properties of tropospheric propagation of very short radio waves beyond the horizon. Jour. Radio Res. Lab., Japan 8, No, 37, 147-174.

Hirai, Masaichi (Sept, 1961b), Diversity effects in spaced- antenna reception of tropospheric scatter waves, Jour. Radio Res. Lab., Japan ^, 301-329,

Hitchcock, R. J., and P. A. C. Morris (July 1961), Wireless World, 37^7378.

The

HF

band:

Is a

new look required?

Hogg, D.
sky.

C,

and W, W, Mumford (March Microwave J. 3, 80-84.

I960),

The effective noise temperature

of the

Hogg, D.

C, and R. A. Semplak (Sept. 1961), The effect of rain and water vapor on sky noise at centimeter wavelengths. Bell Syst. Tech. J. 40, No. 5, 1331-1348.
,

Ikegami, F. (July 1959), Influence of an atmopsheric duct on nnicrowave fading, IRE Trans. Ant, Prop. AP-7 No. 3, 252-257.
Ikegami, F. (May-June 1964), Radiometeorological effects in propagation over the sea and islands, Rev. Elect. Commun. Lab. ToPyo, , 5:T-12, 5-3^4312-324.
,

International Telephone and Telegraph Corporation (1956), Reference data for radio engineers. Fourth Edition, (ITT, New York),

Janes, H. B.

and P. I, Wells (Oct, 1955), Some tropospheric scatter propagation measurements near the radio horizon, Proc. IRE 43, No. 10, 1336-1340.
,

Jasik, H. (1961), Antenna Engineering Handbook,

(McGraw

Hill)

Johnson, M. A. (1958), A review of tropospheric scatter propagation theory and tion to experiment, Proc. lEE 10 5B Suppl. 8, 165-176.
,

its

applica-

Josephson, B,, and A. Blomquist (April 1958), The influence of moisture in the ground, temperature and terrain on ground wave propagation in the VHF band, IRE Trans. Ant. Prop. AP-6 No. 2, 169-172.
,

Josephson, B., and G. Carlson (April 1958), Distance dependence, fading characteristics and pulse distortion of 3000 Mc trans-horizon signals, IRE Trans. Ant. Prop. AP-6 No. 2, 173-175.

Josephson, B., and F. Eklund (April 1958), Some microwave propagation experiences from a just-below-horizon path, IRE Trans. Ant. Prop. AP-6 No, 2, 176-178.
,

Jowett, J. K. S. (Jan. 1958), The measurement and prediction of VHF tropospheric field strengths at distances beyond the horizon, Proc. lEE 105B, Supp. 8, 91-96, and 122126, Paper No. 2500R.
Joy,

W. R. R.

(Jan. 1958a), The long-range propagation of radio waves at 10 length, Proc. lEE 105B Supp. 8, 153-157, Paper No. 2522R,
,

cm wave-

Joy,

W, R. R. (1958b), Radio propagation far beyond the horizon at about 3.2 cm wavelength, Proc. lEE 105B, Supp. 8, 1 58- 164 and 184- 188, Paper No. 2528R.
11-7

Kales, M. L, (May 1951), Elliptically polarized waves and antenna, Proc. IRE 39, No. 544-549.
Kalinin, A.
i

5,

I. (1957), Approximate methods of computing the field strength of ultra short waves with consideration of terrain relief. Radio Eng. 12, No. 4, 13-26, Radiotekhn.

Elektron.

J_2,

No.

4,

13-23.

Kalinin, lu. K (1958), Perturbation of plane radio wave by inhomogeneities of the earth's surface, Radiotech. and Elecktron ~ 3, 557-561, Translation in Radio Engineering and

Electronics 2. No.

4,

143-149.

Kerr, D. E. (1964), Propagation of short radio waves, MIT Radiation Laboratory Series (Boston Technical Publishers, Inc., Lexington, Mass.). 13,
Kirby, R. S.
,

ments over Pike's Peak

,

H. T. Dougherty, and P. L. McQuate (Oct. 1955), Obstacle gain measureat 60 to 1046 Mc/s, Proc. IRE 43, No. 10, 1467-1472.

Kirby, R. S. P. L. Rice, and L. J. Maloney (Oct. 1961), Characteristics of point-to-point tropospheric propagation and siting considerations, NBS Tech. Note No. 95.
Kitchen, F. A., and I, J. Richmond (March 1957), Some characteristics of long distance scatter transmissions (two parts), British Comm. and Electr. 4, No. 2, 74-78 (Feb. 1957); 4, No. 3, 146-148, (March 1957).

Kitchen, F. A., E. G. Richards, and I, J. Richmond (Jan. 1958), Some investigations of metre-wave radio propagation in the transhorizon region, Proc. lEE 105B, Supp. 8, 106-116, Paper No. 2509R.
Kitchen, F. A., W. R. R. Joy, andE. G. Richards (Aug. 1958), Influence of the semipermanent low-level ocean duct on centimetre wave scatter propagation beyond the horizon. Nature 182, No. 4632, 385-386.

Kolmogoroff, A. N. (1941), Dissipation of energy in locally isotropic turbulence, Comptes Rendus (Doklady) de I'Acadamie des Sciences de I'USSR 32_, No. 1, 16-18.
Krasil'nikov, V. A. (1949), The effect of variations of the coefficient of refraction in the atmosphere upon the propagation of ultra-short waves, Izvest. Akad. Nauk.S. S. S. R. Sev. Geograf i Geofiz 22, No. 1, 33- 57 (in Russian).

Kuhn, V. (Feb. and May 1958), Propagation investigation of the effect of various types of terrain in frequency bands I, II, and III, Tech. Comm. Lab. for Commercial Radio

andTelev.

BRF, DDR.

Lane,

J. A., and J. A. Saxton (July 1952), Dielectric dispersion in pure polar liquids at very high radio frequencies, Proc. Roy. Soc. A213, 400-408.

Laws,

J.

O.

tensity,

and D. A. Parsons (Apr. 1943), The relation of raindrop-size to inTrans. Amer. Geophys. Union 24, 452-460.
Proc. LEE

Lewin, L. (July 1962), Diversity reception and automatic phase correction, 109 Part B, No. 46, 295-304.
,

McGavin, R. E. (May
tive index,

1962),

survey

of the techniques for

measuring the radio refrac-

NBS

Tech. Note 99.

McGavin, R.E., and L.

Prop.), No.
,

J. Maloney (Sept. - Oct. 1959), Study at 1046 Mc/s of the reflection coefficient of irregular terrain at grazing angles, J. Res. NBS 63D (Radio
2.,

235-248.
Diffraction of ixltra-short radio waves.

McPetrie, J. S. and J. A. Saxton (Sept. 1942), Nature 150, 292.

11-8

McPetrie,

and L, H. Ford (1946), Some experiments on the propagation over land 9.2 cm wavelength, especially on the effect of obstacles, Proc, lEE 93, Pt. 3-A, Nos. 1-4, 531-538.
J,

S.

of radiation of

Megaw, E.

C. S. (Dec. 1950), Scattering of electromagnetic waves by atmospheric turbulence. Nature 166 1100-1104.
,

Megaw, E. C.
Soc.
_80,

S. (April 1954),

Interpretation of stellar scintillation. Quart.

J.

Roy. Met.

248-251.

Megaw, E. Pt. C

C. S. (Sept. 1957), Fundamental radio scatter propagation theory, Proc. lEE, 104, No. 6, 441-455, see also Monograph 236R, May 1957.

Merkulov, V. V. (1957), On the theory of propagation of electromagnetic waves in media with random inhomogeneities in the index of refraction, Soviet Physics: Tech. Phys. _2, 958-961, J. Electro- Tech. Phys. 27, No, 5, 1051.
Millington, G.

(May

1958),

Tropospheric scatter propagation, Electronic Eng.

30,

No. 363,

248-252.
Millington, G. , R. Hewitt, and F. S. Immirzi (Sept. 1962a), Double knife-edge diffraction in field strength predictions, Proc. lEE 109 , Part C, No. 16, 419-429. See also lEE Monograph No. 507E (Mar. 1962). Millington, G.
,

Proc. lEE 109

,

R. Hewitt, and F. S. Immirzi (Sept. 1962b), The Fresnel surface integral, Part C, 430-437. See also lEE Monograph No. 508E (Mar. 1962). ,

Millington, G. and G. A. Isted (July 1950), Ground wave propagation over an inhomogeneous, smooth earth, Part 2: Experimental evidence and practical implications, Proc, lEE, Part III r?, No, 48, 209.

Misme, P. (July 1958), The correlation between the electric field at a great distance and a new radiometeorological paranneter. Trans, IRE Trans, Ant, Prop, AP-6, No, 3,
289-292,

Misme, P, (March-April
cal calculations,

Ann. des

1960a), The equivalent gradient direct measurements and theoretiT^^comm. 15, Nos. 3-4, 92-99.

Misme, P. (Aug. Proc. IRE

1960b),
48, No.

Comments on "Models
8,

of the

atmospheric radio refractive index,"

1498-1501.

Misme, P. (Nov. -Dec.

Ann. des

1960c),
_1_5,

Some aspects

T^^comm.

of radiometeorology and radioclimatology, No. 11-12, 266-273.

Misme,

P.

(May-June 1961),

stability on transhorizon paths in the

The influence of the, equivalent gradient and atmospheric Sahara and the Congo, Ann. des Telecomm.

16,

Nos. 5-6, 110-116.

Moler, W. F. and D. B. Holden (Jan. - Feb. I960), Tropospheric scatter propagation and atmospheric circulations, J. Res. NBS 64D (Radio Prop. ), No. 1, 82-94.
,

Nakagami, M. (Oct.

1940), Study on the resultant amplitude of many vibrations whose phases and amplitudes are random, Nippon Elec. Comm. Eng. 22, 69-92.

National Bureau of Standards (1954), Applied Mathematics Series 32, and cosine integrals for arguments from 10 - 100.
National Bureau of Standards (Jime 1964), Mathematical Functions.

Table of the sine

Applied Mathematics Series

55,

Handbook

of

11-9

Neugebauer, H. E.

J.,

conical obstacles,
,

and M. P. Bachynski (July- Aug. I960), Diffraction by smooth J. Res. NBS 64D (Radio Prop. ), No. 4, 317-329.

Newton, R. G. and T. F. Rogers (Nov. 1953), Dependence of total microwave atmospheric absorption on propagation path elevation, Air Force Cambridge Report AFCRC Tech. 53-54A.

Nomura,

Y. A.

in an

and K. Takaku (Aug. 1955), On the propagation of electromagnetic waves inhomogeneous atmosphere, J. Phys. Soc. Japan 10, No. 8, 700-714.
,

Norton, K. A. (Dec. 1941), The calculation of ground -wave field intensity over a finitely conducting spherical earth. Proc. IRE 29, No. 12, 623-639. Norton, K. A. (1950),

Addendum to Reference E to the report of Ad Hoc Committee of the F. C. C. for the Evaluation of the Radio Propagation Factors Concerning the TV and Broadcasting Services in the Frequency Range Between 50 and 250 Mc/s. (See

FM

Norton and Fine reference below,

Norton, K. A. (Jan, 1953), No. 1, 146-152.

Transmission loss in radio propagation,

Proc. IRE 41,

Norton, K. A. (Jixly-Aug. 1959), System loss in radio wave propagation, (Radio Prop. ), No. 1, 53-73.

J.

Res.

NBS 63D

Norton, K. A. (July I960), Carrier frequency dependence of the basic transmission loss in tropospheric forward scatter propagation, J. Geophys. Res. 65, No. 7, 2029-2045.
Norton, K. A. (1962),
Efficient use of the radio spectrum,

NBS

Tech. Note 158.

Norton, K. A., and E. C. Barrows (1964), Observed vertical wavenumber spectra of refractivity near the ground, to be published. Norton, K. A. and H. Fine (Aug. 1, 1949), A study of methods for the efficient allocation of radio frequencies to broacasting services operating in the range above 50 Mc, Reference E to the Report of Ad Hoc Committee of the F. C. C. for the Evaluation of the Radio Propagation Factors Concerning the TV and FM Broadcasting Services in the Frequency Range Between 50 and 250 Mc/s.
,

Norton, K. A. and A. C. Omberg (Jan. Proc. IRE 35, No. 1, 4-24.

,

1947),

The

maximum

range of a radar set,

Norton, K. A., P. L. Rice, H. B. Janes, and A. P. Barsis (Oct. 1955), The rate of fading in propagation through a turbulent atmosphere, Proc. IRE 43, No. 10, 1341-1353. Norton, K. A. P. L. Rice, and L. E. Vogler (Oct. 1955), The use of angular distance in estimating transmission loss and fading range for propagation through a turbulent atmosphere over irregular terrain, Proc. IRE 43, No. 10, 1488-1526.
,

Norton, K. A.

H. Staras, and

M. Blum (Feb.

1952),

problem

of multiple radio interference to

FM

A statistical approach to the and television service, IRE Trans.

Ant. Prop.

AP-1, 43-49.

Norton, K. A., L. E. Vogler, W. V. Mansfield, and P. J. Short (Oct. 1955), The probability distribution of the amplitude of a constant vector plus a Rayleigh-distributed vector, Proc. IRE 43, No. 10, 1354-1361.

11-10

Obukhov, A. M. (1941), On the distribution of energy in the spectrum of turbulent flow, Bull. Acad. Sci. USSR Geo. and Geophys. Ser. 4-5, 453. See also Comptes Rendus (Doklady) L'Academie des Sciences de I'USSR 32, No. 1, 19 (1941).

Obukhov, A. M. (1953), On the effect of inhomogeneities of the atmosphere on sound and light propagation. Bull. Acad. Sci. USSR, Geog. and Geophys. Ser. 2, 155.
Onoe, M. M. Hirai, and S. Niwa (April 1958), Results of experiments of long distance overland propagation of ultra-short waves, J. Radio Res. Lab. (Tokyo) 5, No. 20, 79-94.
,

Onoe, M.

line of sight,

and K. Nishikori (Oct. 1957), Microwave propagation over the sea beyond the Radio Res. Lab. J. 4, No. 18, 395-406.
of the

Pearcey, T. (1956), Table

N. Y.).

Fresnel integral, (Cambridge Univ. Press,

New York,

Pekeris, C. L. (Feb, 1947), Note on scattering of radiation in an inhomogeneous medium, Phys. Rev. 7J.. No. 3, 268-269.
Perlat, A,, and J. Voge (Dec. 1953), Attenuation of centimeter and millimeter waves by the atmosphere, Ann. des T^^comm. 8, No. 12, 395-407.

Price, W. L. (July 1948), Radio shadow effects produced in the atmosphere by inversions, Proc. Phys. Soc. London^, No. 343, 59-78,
Randall, D. L. (1964), A summary of tropospheric radio duct meteorology at V.H.F. and UHF as observed on a trip around the world. World Conference on Radio Meteorology Boulder, Colorado, September 14-18.

Rayleigh, Lord (Aug. 1880), On the resultant of a large number of vibrations of the pitch and of arbitrary phase, Phil. Mag. 10, 73-78.
Riblet, H. J., and C. B. 63.
,

same

Barker (1948),

general divergence formula,

J.

Appl. Phys.

_19_,

Rice, P. L. and F. T. Daniel (Apr. 1955), Radio transmission loss vs. distance and antenna height at 100 Mc, Trans. IRE Ant. Prop. AP-3 , No. 2, 59-62. Rice, P. L. and J. W. Herbstreit (1964), Tropospheric propagation, (to be published in Vol. 20 of Advances in Electronics, Academic Press).
,

Rice, S. O. (Jan. 1945), 24, 46-156.

Mathematical analysis of random noise,

Bell.

System Tech.

J.

Rice,

S.

O. (1954),
J.

Diffraction of plane radio waves by a parabolic cylinder.

Bell System

Tech.

33,

417-504.

Rider, G. C. (1953), Some VHF experiments upon the diffraction effect of hills. The Marconi Rev, _16, No. 109, 96-106, 2nd quarter.

Rowden, R. A., L. F. Tagholm, and J. W. Stark (1958), A survey of tropospheric wave propagation measurements by the BBC, 1946-1957, Proc. lEE 105B Suppl. 8, 84-90 and 122-126, Paper No. 2517R.
,

Ryde,

J.

W. (1946), The attenuation and radar echoes produced at centimetre wavelengths by various meteorological phenomena. Conference on meteorological factors in radiowave propagation, Phys. Soc. (London) and Royal Meteorological Society, 169-188. W. and D. Ryde (1945), Attenuation of centimeter waves by rain, hail, fog, and clouds, General Electric Co., Wembley, England.

Ryde,

J.

U-U

Saxton, J, A. (Sept. 1951), The propagation of metre radio waves beyond the normal horizon,

Part

1,

Proc, lEE

_98_,

Part

III,

No. 55, 360-369.

Saxton, J. A., and J. A. Lane (May 1955), other obstacles. Wireless World 61, Saxton, J. A., J. A. Lane, R. ture of the troposphere,

VHF

and UHF reception-effects Z29-232.

of

trees and

W. Meadows, and P. A. Matthews (Feb. Proc. lEE 111 No. 2, 275-283.

,

1964),

Layer struc-

Schelkunoff, S. A. and H. T. Friis (1952), Antennas, theory and practice, Wiley and Sons,

New York
Schelling,

City.

J. C. , C. R. Burrows, and E. B. Ferrell (Mar. 1933), propagation, Proc. IRE Z]_, No. 3, 427-463.

Ultra-shortwave

Schiinemann, R. (Sept. 1957), Mechanism of ultra short wave propagation over great distances, Hochfreq. u. Electroak. 66, No. 2, 52-61.

Sherwood, E. M. and W. L. Ginzton (July 1955), Reflection coefficients at 10 cm., Proc. IRE 43, No. 7, 877-878.
,

of irregular terrain

Shkarofsky,

RCA
Siddiqui,
J,

P. (Mar. 1958), Tropospheric scatter propagation, Res. Rpt. No. 7-200-1, Victor Co., Ltd. Res. Labs., Montreal, Canada.
I.

M. M, (March-April 1962), Some problems connected with Rayleigh Res. NBS 66D (Radio Propagation), No. 2, pp. 167-174.

distributions,

Silvernnan, R. A. (Apr.

1957), Fading of radio waves scattered by dielectric turbulence, J. Appl. Phys. _28, No. 4, 506-511. Also New York Univ. Inst, of Math. Sci. , Electromagnetic Res. Division, Res. Rept. EM 101 (Jan. 1957).

Staras, H. (Oct. 1952), Scattering of electromagnetic energy in a randomly inhomogeneous

atmosphere,

J.

Appl. Phys.

_23,

No. 10, 1152-1156.

of

Staras, H. (Oct. 1955),

Forward scattering
1374-1380.

radio waves by anisotropic turbulence, Proc.

IRE

43, No.

10,

Staras, H. (April 1957), Antenna-to-medium coupling loss,

IRE Trans. Ant. Prop. AP-5,

No.

2,

228-231.

Starkey, B. J., W. R. Turner, S. R. Badcoe, and G. F. Kitchen (Jan. 1958), The effects of atmospheric discontinuity layers up to and including the tropopause on beyond-thehorizon propagation phenomena, Proc. lEE 105B, Suppl. 8, 97-105 and 122-126,

Paper No. 2486R.

Stokes, G. G. (1922), Mathematical and physical papers. Vol. in. On the composition and resolution of streams of polarized light from different sources, (Cambridge University Press, London), 233-258.

Straiton, A.

W., and C. W. Tolbert (May 1960), Anomalies in by atmospheric gases, Proc. IRE 48, No. 5, 898-903.

the absorption of radio

waves

Sutton,

O. G. (1955), Atmospheric turbulence (John Wiley and Co.).

Tao, K. (Jan. 1957), On the relationship between the scattering of radio waves and the statistical theory of turbulence, J, Radio Res. Lab. (Tokyo) 4, No. 15, 15-24.

T.A.S.O. (March

1959), Engineering aspects of television allocations, Report of the television allocations study organization.

11-12

Taylor, G. 196.

I.

(1922), Diffusion by continuous

movements, Proc. London Math. Soc.

II

20,

Thourel, L. (1960), The antennas, translated by H. de Laistre Banting (John Wiley and Sons, Inc., New York, N. Y. ).
Tolbert, C. W.

tion of millimeter radio

No.

2.

and A. W. Straiton (Apr 1957), Experimental measurement of the absorpwaves over extended ranges, IRE Trans. Ant. Prop. AP-5 239-241.
, . ,

Troitski, V. (May 19 56), The propagation of ultra-short horizon. Radio Technika 11, No. 5, 3-20.

waves

at great distances

beyond the

Troitski, V. N. (Jan. 1957a), About the influence of the form of the structure function of non- homogeneous dielectric permeability of air on long distance tropospheric propagation of ultra short waves, Radio Eng. 2, 34-37. Troitski, V. N. (1957b), Fading of ultra-short
10.

waves

in radio relay

systems.

Electrosviaz

Ugai, S. (May- June 1961), Characteristics of fading due to ducts and quantitative estimation of fading. Rev. Elect. Comm. Lab., Japan 9, No. 5-6, 319-360.

Ugai, S.

Microwave transmission across a S. Aoyagi, and S. Nakahara (May 1963), mountain by using diffraction gratings, Electronics and Communications in Japan 46,
,

No.

5,

7-17.

Unwin, R. S. (Nov. 1953), Ultra-short-wave field-strength in a ground-based radio duct, Nature 172 No. 4384, 856-857.
,

Van Vie ck,

No.
7,

J.

H. (Apr. 1947a), 413-424.

H. (Apr.
1947b),
71_,

The absorption

of

microwaves by oxygen,

Phys. Rev. 71,

Van VIeck,

J.

The absorption
7,

of

microwaves by uncondensed water

vapor,

Phys. Rev.
H. (1951),

No.

425-433.
of

Van Vie ck,

J.

Short Radio Waves,

Villars, F.
,

Theory of absorption by uncondensed gases, Propagation (McGraw-Hill Book Co. New York, N. Y. ), 646-664.
,

bvilent fluctiiations of the

and V. F. Weisskopf (Oct. 1955), On the scattering of radio waves by turatmosphere, Proc. IRE 43, No. 10, 1232-1239.

Voge,

J.

(Mar. 1953),

of

Commission

11,

The troposphere and wave propagation (Summary of Proceedings lOth General Assembly URSI (1952),) L'Onde Electronique 33,

No. 312, 136-150.

Voge,

Radioelectricity and the troposphere, parti, theories of propagation to J. (1955), long distances by means of atmospheric turbulence, L'Onde Electrique 35, 565-581.
J. (Nov. 1956), 1621-1622.

Voge,

Useful bandwidth in scatter transmission,

Proc. IRE 44, No.

11,

Voge,

J.

(Nov.

Telecomm.

Dec. I960), Theories of transhorizon tropospheric propagation, I5j, No. 11, 260-265.

Ann. des

Vogler, L. E. (July 1964),

region.

Radio

Sci.

J.

Calculation of groundwave attenuation in the far diffraction Res. NBS/USNC-URSI 68D, No. 7, 819-826.

11-13

Vogler, L. E.

and J. L. Noble (Sept. - Oct. 1963), Curves of ground proximity loss for dipole antennas (a digest), J. Res. NBS 67D (Radio Prop. ), No. 5, 567-568.
,

Vvedenskii, B. A., and A. G. Arenberg ( 1957), ultra-short waves. Radio Eng. 12, No.

1,

Long distance tropospheric propagation 3-13; Radio Eng. 12, No. 2, 10-25.

of

Vvedenskii, B. A., and A. V. Sokolov (19 57), Investigation of tropospheric propagation of meter, decimeter, and centimeter radio waves in the USSR, Radio Eng. and Elect. (USSR) 2, No. 11, 84-105.

Vysokovskii, D, M. (19 57a),

of ultra-short

waves

in the troposphere.

183-187.

Calculation of multiple scattering in the diffusion propagation Radio Eng. and Elect. (USSR) ~ 2, No. 6,

Vysokovskii, D. M. (1957b), Geometrical characteristics of the scattering of radio waves by turbulent inhomogeneities in the troposphere. Telecommunications (USSR) ~ 9,
11-20.

Vysokovskii, D. M. (1958), Diffused propagation of ultra-short waves in the troposphere with high-directivity antennas. Telecommunications (USSR) 5, 488-497.
Wait,
J.

R. (1958), On the theory of propagation of electromagnetic waves along a curved surface. Can. J. Phys. 36, No. 1, 9-17. R. (1959),

Wait,

J.

Electromagnetic radiation from cylindrical structures (Pergamon

N. Y.
).

Press,

New York,

Wait, J. R. (April 1959), Transmission of Power in Radio Propagation, Electronic and Radio Engineer, Vol. 36, Series No. 4, pp. 146-150.

Wait,

Electromagnetic waves in stratified media. International Series of J. R. (1962), Monographs on Electromagnetic Waves 3, (Pergamon Press, New York, N. Y.
)

Wait,

J.

R. (Nov.

1963),

J.

Res.
,

NBS 67D

Oblique propagation of gro\md waves across a coastline, part (Radio Prop. ), No. 6, 617-624.

I,

Wait,

J.

and A. M. Conda (Sept. - Oct. 1959), Diffraction of electromagnetic waves by R. smooth obstacles for grazing angles, J. Res. NBS 6 3D (Radio Prop.), No. 2, 181197.

Wait,

J.

and C. M. Jackson (Nov. 1963), Oblique propagation of ground waves across R. a coastline, part II, J. Res. NBS 67D (Radio Prop. ), No. 6, 625-630,
,

Wheelon, A. D. {June 1957), Relation of radio measurements to the spectrum of tropospheric dielectric fluctuations, J. Appl. Phys. 2, 684-693. Wheelon, A. D. (Sept. - Oct. 1959), Radio-wave scattering by tropospheric irregularities, J. Atmos. and Terr. J. Res. NBS 63D (Radio Prop. ), No. 2, 205-234; also, Phys. _15, Nos 3, 4, 185-205 (Oct. 1959).
Wilkerson, R. (1964),
Multiple knife-edge diffraction, (private communication).

Williamson, D. A., V. L. Fuller, A. G. Longi-y, and P. L. Rice (Mar. I960), A summary of VHF and UHF tropospheric transmission loss data and their long-term variability, NBS Tech. Note 43.

11-14

12.

LIST

OF SYMBOLS AND ABBREVIATIONS

precedes the Greek alphabet, and lower-case

In the following list the English alphabet

letters precede
for quantities
is

upper-case letters.

As

a general rule, upper-case letters have been used

is

expressed in decibels, for example p

transmitter power in watts, and

transmitter power in decibels above one watt.

Sometimes
for
is:

symbol may be used

in quite different contexts,

in

which case

it

is listed

each separate context.

Subscripts are used to modify the meaning of symbols.

The order

1.

Syinbol without a subscript.

h h
h
h(x)
ht.

2.

Symbol with

a subscript,

(letter subscripts in alphabetical

order followed by number subscripts in numerical order).

3.

Symbol

as a special function.

4.

Abbreviations.
is

Following each definition an equation number or section number

given to show the

term

in its

proper context.

Where

applicable,

reference
to the

is

made

to a figure.

Throughout the report, logarithms are

a

base

10

unless otherwise noted.

Effective earth's radius, allowing for average radio ray bending near the surface
of the earth,

(4.4) figure 4.

2.

a
e

An

equivalent earth's radius w^hich is the harmonic ^

mean

of the radii

a
t

and a
r

(7,10).

a a

The "effective absorbing area"

en

of

an antenna,
n

(2. 24),

The effective absorbing area

for the

discrete plane wave incident on an an-

tenna from a single source, (11.34),

a
ei
,

a
ez

The effective absorbing area

(11.86).

of the receiving

antenna for each of two waves,

a a

The fraction

of

energy absorbed along a ray path, or scattered out

a

of

it,

(II.

26).

pm

pn

The fraction

of energy,

above, for the

and n

multipath components
1

from

a single source,

where

and n take on integral values from

to

(11.39).

The radius

of a circular arc that is tangent to the

receiving antenna horizon ray

at the horizon,

and that merges smoothly with the corresponding arc through the
8. 7.

transmitting antenna horizon, (8,9) figure

a
s

Effective earth's radius factor corresponding to

D
s

(8.15).

a
s

Radiowave scattering cross-section

(IV. 13).

of a single scatterer or

group

of scatterers,

12-1

Radius

of a circular arc that is tangent to the transmitting

horizon ray at the

horizon, and that

merges smoothly with

(8. 9)

the corresponding arc through the re-

ceiving antenna horizon,

figure 8.7.
(IV. 14).

a
a
a

Radiowave scattering cross-section per unit volume,

V
vo
,

a
V2

a
V3

Radiowave scattering cross-section from refractivity turbulence, (IV. 21). Radiowave scattering cross-sections per unit volume for large, medium, and
small layers, (IV.
15)

VI

to (IV. 17).
(II.

a
a

The axial ratio

,

of the polarization ellipse of a plane -wave,

of the

15).

xn

a
XI

a
x2

Axial ratios of the polarization ellipse

first,

and second plane

wave from
a

a single source,

(11.3

5)

and

(11.8 5).

xr

The axial ratio

(11.17).

of the polarization ellipse

associated with the receiving pattern,

xrn

xri

xrz tern for the

(11.82).

Axial ratios of the polarization ellipse associated with the receiving patn
,

first,

and second plane wave from a single source,

(11.35)

and

a a

The actual earth's radius, usually taken

Radius
of the circular

to be 6370 kilometers,

(4.4).

arc that

is

tangent to the transmitting antenna horizon ray

at the horizon,

and that passes through a point h

kilometers below the trans-

mitting antenna, (8.8) figure 8.7.

Radius of the circular arc that

the horizon,

is

tangent to the receiving antenna horizon ray at

and that passes through a point h

kilometers below the receiving

antenna, (8.8) figure 8.7.

a
,

Positive or negative amplitudes of real and imaginary components of a complex

vector:
-* a = a

+ ia

a
,

= a

+ a

2
,

(II.

52).

a,

a
,

The real vector a

a

aa, where a

is a unit

vector.
of a

Real vectors defining real and imaginary components

complex vector: a

T
a
a^
a (-f)
e

+ ia"

(11.45).

A complex A complex
and
(2. 24).

vector:

a = a

(II.

45)
,

vector defined in terms of the unit vector system x

of a receiving

x
(-f)

(II.

62).

The effective absorbing area

antenna in the direction

(2. 22)

A A
Aa

An antenna

terminal, figure 6.3.

Attenuation relative to free space, expressed in decibels, defined as the basic

transmission loss relative

to that in free space,

(2.35).
to

See

The long-term median attenuation

by oxygen and water vapor,

of radio
3.

waves due

atmospheric absorption

section

12-2

A ar .A at

For transhorizon paths,

^

at

A ar

the

sum

of the

absorption from the ^

transmitter to the crossover of horizon rays and the absorption from the cross-

over

of

horizon rays

to the receiver,

section

3.

A A
A

Total absorption attenuation within a cloud,

(3. 13).

The hourly median attenuation relative

to free space,

annex

I.

Total absorption due to rainfall over a given path,

(3. 7).

A
A A A A 12
,

Attenuation relative to free space, defined as basic propagation loss relative to

that in free space,

(2.47).

See A.
full leaf,
(5. 18).

Rate

of attenuation

through woods in

Diffraction attenuation relative to free space at an angular distance

a

9 =

over

smooth

earth,

section

9.

2..

Antenna terminals, figure ^

6. 1.

A(v, 0)

Attenuation relative to free space as a function of the parameter v, (7.2) figure

7. 1.

A(v,

p)

Diffraction attenuation relative to free space for an isolated perfectly conducting

rounded obstacle,
A(0,
p)

(7.7),

figure 7.3.

The diffraction loss for

0=0
j

over an obstacle of radius

r,

(7. 7)

figure 7.4.

A(v
J

Attenuation relative to free space for each of several rays

as a function of the

parameter

v.

where
of

1, 2, 3, 4,

(III.

34).

A m (p)

The time availability

hourly median values

Am

Figures 1:21-1:26 show

,

A
o

(p)

plotted against the straight-line distance,

to 99. 99 percent. of

for values of

ranging

from 0.01

The dimensions
lar to

an atmospheric layer or feuillet in any direction perpendicu-

K,

(IV. 9).

b
b

Effective bandwidth of a receiver in cycles per second,

(V. 7).

The parameter
zation,

b,

a function of

ground constants, carrier frequency, and polari8. 2,

expressed in degrees, figure

and equations

(III. (III.

40) 40)

and

(III.

41).

b b

The parameter b The parameter b

for horizontal polarization defined by for vertical polarization,

b,
(III.

41).

B B B
s

Effective bandwidth,

expressed

in decibels

above one cycle per second, (V.8)

An

antenna terminal, figure 6.3.

B(K,
b)

The parameter
1, 2, t,

corresponding to the effective earth's radius

B(K, b)
^ '

a
s
,

(8.15).
(8. 13). '
^

B,

Values

of the parameter ^
(8. 2)

that

correspond ^

to values of

K
1,

2, t,

Defined by
o

as the product of several factors,

combined

for convenience in

considering diffraction.

12-3

Any
B(K,b)

point along the great circle path

from

to

figure

6. 3.

A A

parameter plotted

in figure 8. 3 as a function of
of

K and

0. 3

b,

(8.2).

Free space velocity

radio waves,

c =

299792.
-

km/sec.

parameter showing

the

phase change u

associated with the complex plane

to reflection

wave reflection
infinite

coefficient

R exp

[.Ktt - c)

corresponding
III. 1

from an

smooth plane surface,

of
c

(5.4) figures

through

III. 8.
(III. 13)

V^
c

Values
(III. 14)

for horizontal and vertical polarization,

III. 1

respectively,

and

figures

through

III. 8.

Polarization efficiency of the power transfer from transmitter to receiver,

P
(IV. 13).

2'

c 3

coefficients R^^ The phase changes associated with the complex reflection

R
C
Ci
C.

(III. 3 2).

e3

Difference in longitude between

and

B,

(6.

1)

and

(6. 2)

Cosine integral,

(III. 51). (III. 33),

Fresnel integral,

where j= 1,2,3,4.
to

C
C
,

A
C
,

parameter which relates K

Values
of

K(8497), (8.2).
to effective earth's radii

corresponding
o

a^, a^,

and

a^,

or'

os'

ot

(8.13).

C
Ol

,C
02

Values

of

corresponding
^

to effective earth's radii

a^

and a^, (8.13).

Qi
C(u), C(v)

Difference in the longitude of the points

and B'

(6. 6) to (6. 9)

Fresnel cosine integrals,

(IV. 8).
(8.
1)

(K, b)
,

parameter used
,

in calculating diffraction attenuation,

(K, b)

figure 8.4.

C (K

b),

C (K

b)

The parameter C
C^(K^)

corresponding

to

K^ and K^, also

written
C"

and C ^(K ^)
of

(8.11).
,

(K

The weighted average

values of C (K
of

b)

and ^^^(K^,

b)

(8.11).

Ci(r)

Cosine integral as a function

),

r,
of

(III. 51).

Ci(r

12
Ci(r

Cosine integral as a function

(III. 50).
\

CCIR

International Radio Consultative Committee.

CRPL
C. d

Central Radio Propagation Laboratory, National Bureau of Standards, U.S.A.

W.

Continuous wave.

Great circle propagation path distance, measured

circle path determined by two antenna locations,

at

sea level along the great

and

figure

6. 1,

Clearing depth in meters, defined as the distance from the edge of woods

to the

lower antenna along a propagation path,

d
e

(5. 19).
d,
f

path distance, a function of Effective propagation f f B f

10.
1,

mctere
,

and h

section

(10.3).
to its horizon,

d.

Lr

Great circle distance from the receiving antenna

figure 6.1. figure

6. 1.

Great circle distance from the transmitting antenna

to its horizon,

12-4

dq
d
r

differential amplitude reflection coefficient for a tropospheric layer,

(IV. 5),

Distance used In calculating ground reflections in knife edge diffraction;

defined by
(III. 29)
.

d
r

is

sr

Distance between the receiving ^ antenna horizon and the crossover of horizon

rays as measured at sea level, (6.20).

d

Distance between the transmitting antenna horizon and the crossover of horizon rays as measured at sea level,
(6. 20).

d'

sr

d'

If

St

or

or d

or
in

is negative,
6
.

d'
9,

ot

sr

or

d'

is

computed and substituted

for

d sr

st

reading figure

(6.23).

d d

so
si

factor used to normalize effective antenna heights in computing

fields are

d
e

(10. 2).

The theoretical distance where diffraction and scatter

equal over a smooth earth, (10.
1).

approximately

The greatest distance

o

for

which the attenuation relative

to free

space

is zero,

(5.10).

Distance from the transmitting, or the receiving antenna, to the crossover of

horizon rays, measured

d
,

at

sea level, figure

6. 1.

Great circle distance from one antenna

reflected ray, figure
5. 1.

of a pair to the point of reflection of a

d'

,d

Distances used in computing diffraction attenuation with ground reflec(III.

tions,

31) figure III. 9.

db

Decibels
all

= 10 log

(power ratio) or

20 log

(voltage ratio).

In this report,

logarithms are to the base 10 unless otherwise stated,

dbu

Decibels above one microvolt per meter,

Decibels above one watt.

dbw

D
D

Divergence coefficient, a factor used

to reflection

to allow^ for the

(5. 2). (2. 16).

divergence of energy due

from a convex surface,

Diameter

of a

parabolic reflector in meters,

D
D
e

Great circle distance between transmitting and receiving horizons,

ure 6.1.

(6. 17),

fig-

function of

used

in

computing diffraction
of the

loss,

(8. 16),

figure 8.8.
e

The positive or negative amplitude

of a

cross-polarized vector component

2.

complex polarization vector

e,

sections
of the

4 and

II. 2.

cr

The positive or negative amplitude

of a

cross-polarized vector component

(II.

cr

receiving antenna response pattern,

16).
e. e.
1

e.
1

The positive or negative amplitude

plex plane wave

of the real vector

(ir),

associated with a comare time -invariant

>^

(e
r

+ ie
1

exp
(II.

where

e r

and

and exp

(ir)

is a

time phasor,

8b)

12-5

The positive or negative amplitude

of a

of the principal polarization

component

complex polarization vector

e,

sections 2.4 and

II. 2.

pr

The positive
of a

or negative amplitude of the principal polarization

(II.

component

pr

receiving antenna response pattern,

16).

The positive or negative amplitude

with a complex plane
invariant and exp
(It)

of the real vector

component
e

associated
e
i

v^'ave
is a

\/2~(e
r

+ ie.) exp
1

(ir),

where

r
(II.

and

are time-

time phasor,

8a)

e
e

Equivalent free space field strength,

(II. 5),

(2.44).

Equivalent inverse distance field strength, (2.45).

,

The positive or negative real amplitudes

the

of real

and imaginary components of

complex polarization vector

of real

e,

(11.10).
e 9

e
9,

e
4)

The positive amplitudes ^ ^

<()

vectors
-wave,

and

e
<j)

associated with the

II. 1.

and

components

of a

complex plane

(II. 7)

figure

Real vectors associated with cross and principal polarization components

of a

uniform
e
c
,

elliptically polarized plane wave,

annex

II,

section

II. 2.

Directions of cross and principal polarization, chosen so that their vector

_ _
e
,

product

x e is a unit vector in the direction of propagation, (II. 14). c P Cross and principal polarization field components of a receiving antenna ree

sponse pattern, (11.16).

e
,

cr

pr

Directions of cross and principal polarization components of a receiving antenna response pattern,
(II. 18),
(II.

20).

e.

The real vector associated with

part of a complex plane wave

the

imaginary component
+ ie.) exp
r
1

of the
8b).

time -invariant

\J2

(e

(ir),

(II.

e r

The real vector associated with the real component

of a

of the time-invariant part

8a).
e

complex plane wave

\fT~ (e
of a

+ ie.) exp (ix),

(II.

Real vector components

complex polarization vector

which has been re(II.

solved into comoonents which are orthogonal in both space and time,
e 9
,

10).

e
<^

Real vectors associated with the

\J2

and
(ir),

cj)

components

of a

complex plane wave

(ir)

[e
9

exp

(ir

+ e
cj)

exp
II. 1

9
(II. 7)

(It )]
(p

exp

where only

the

phasor exp

depends

on time,
e

figure
e
cp

9
e
cp

e, e

A A A

unit vector unit vector

x^"

perpendicular to
9

e
cp

and r,
r

(II. 3 b)

figure
(II.

II. 1 II. 1

(r

x x o )/sin

perpendicular to
to indicate a

and x
'^

3a) figure

bar

is

used under the symbol

complex vector: e

= e

+ ie
c

=e

pr

+ie cr ,(2.

19).

The complex conjugate

,
I
I

of

e:

= e

ie

'

el e r letjle
'

The magnitudes
I,

of the

complex vectors

'

and

(II.

22).

'e'er'

lel.le pr'
p

The magnitudes
,

of the cross and principal polarization

com

ponents

e c

cr

and

pr

sections 2.4,

II. 2,

II.

3.

12-6

E
E E
E.R.P.
o

Field strength in dbu, (2.43).

The equivalent free space

field strength in dbu,

field,

(2.44).

The equivalent inverse distance

Effective radiated power,

(2.45).
^^pt^^i)

E. R. P. = P^ +

"

^'^^

'^^^
2.

E
f

ikw

Field strength in dbu per kilowatt effective radiated power, see section

Radio wave frequency in megahertz (megacycles per second).

Diffraction loss for each of several distinct rays over an isolated obstacle, where
J j
:=

1, 2, 3, 4,

(III.3 2-III.3 5).

f, f f

M Hz
,

^,

Radio wave frequency

-1^6
in

megahertz.
7).
(III.

op

Operating noise factor of a receiving system, (V.

f
,

12
,

f,, f 3 4

Diffraction loss for each of four distinct rays over an isolated obstacle,

32).

f(r
f(v.)

),

f(r

Functions of the normalized antenna heights

r
,

and
(III.

(III.

50).
10.

f(0

A A A

function identically equal to

f.

for

v = v.

33) figure

III.

factor used to reduce estimates of variability for antenna

(III.

beams elevated
.

above the horizon plane,

f ( v)

and 9, h b function used in computing path antenna gain, defined by (9. 13) figure
64) figure III. 22.

See

9,

9. 7.

F
F
F

The correction term F

allows for the reduction of scattering efficiency at

(9.
1)

great heights in the atmosphere,

.

and

(9. 7).
i

oi

Scattering efficiency correction term for the

(III. 63).

lobe of an antenna pattern.

op F(x

Operating noise factor of a receiving system expressed in decibels, (V.

),

8)

F(x

Functions used in computing diffraction attenuation,

8.6.

(8. 1)

and figures

8. 5

and

F(9d)

The attenuation function used

scatter paths,
(9. 1)

in calculating
1,

median basic transmission loss

III.

for

figures

9.

and

III.

11 to

14.
9
.

F(0

.d)

This function

is the

ei

same

as

F(9d)

with the effective angular distance

9,

ei

substituted for the angular distance,

annex

III,

(III.

57).

FM
g

Frequency modulation, Grade

of

service.

specified grade of service guarantees a corresponding

to the

degree of fidelity of the information delivered

section V.
g
5.

receiver output, annex V,

Maximum

free space directive gain,

flux density and
2

or directivity, the ratio of the available

for a loss-free antenna,

mean power
II.

/r|

section

2. 3,

annex

g ta

high gain antenna radiates

counted for by the main

III,

beam

watts per unit area in every direction not acb or by one of the side lobes of an antenna, annex

section

III. 6.

g bt

The directive gain

for a transmitting antenna,

annex

III,

section

III. 6.

12-7

gc
'cr
'P

The cross-polarization component

The cross-polarization component

of the directive gain,

(II.

24).
(II.

of the directive gain of a receiver,

(II.

19).

Principal polarization directive gain,

24).

Principal polarization directive gains for the receiving and transmitting antennas,
g
,

respectively, (11.26).
free space directive gains for the receiving and transmitting antennas

g
t

Maximum

respectively,
g

section 2.3,
g

^rn g ^tn
,

The directive gains ^

and
(11.34).

^r

and g

for the

^t

of a series of plane ^

waves,
-

(II. v

33)
/

ri

Directive gain factors defined for each antenna in the direction of the point of
rz

ground reflection, (5.1).

g
g g
,

The

maximum

value of the operating gain of a receiving system, (V.

for one antenna in the direction of the other,
of the transmitting

7).
5. 1.

The directive gain

g

section

The directive gain

and receiving antennas, each in the di(5. 1).

(II.

rection of the other, assuming matched antenna polarizations,

g
,

g
cp

Directive gains associated with the field components

e
cp

14).

g(f)

g(p,

f)

A A

frequency correction factor shown frequency factor used

to adjust

in figure III. 30,

(III.

66).

predicted long-term variability to allo'w for

frequency-related effects, (10.6) figure 10.3.

g(10, f), g(90,
f)

The frequency factor

g(p, f)

used

to adjust

(10)

and Y^(90)

pre-

dicted values, (10.6) figure 10.3.

g(r)

Directive gain in the direction r

Directive gain in the direction
g (r)

(II. 56)

g(_r)
g
(r),

(-r)

annex

II.

Cross polarization and principal polarization directive gains

r,
(11.61).

in the direction

g
a

(_r) r
(r

Free space directive gain

B

of the receiving

antenna in the direction

(-r)

(2.32).

^r^

(r
2

Directive gains associated with direct and ground-reflected rays, respectively,

^r^

(11.84).
of the

g (r)

Free space directive gain

also
g'(r)
,

transmitting antenna in the direction r

see

section
of a

2. 3.

g'

Power gain
minals
is

transmitting antenna when the power input to the antenna tersection 2.3.
r
,

p'

watts,

g'

(r)

Power gain
Grams.

of a

transmitting antenna in the direction

see subsection 2.3.

gm
g/m
G
G,

Grams
The

per meter.
free space directive gain relative to an isotropic radiator,
of of
g,
,

maximum

Decibel equivalent
Decibel equivalent

G, bt

hub
G
- 10 log g

(2. 14).

annex

III

section

III. 6.

for a transmitting antenna,

of a

annex

III

section

III, 6.

The hourly median operating signal gain

receiving system, (V.8).

12-8

G
P

Path antenna gain, the change in transmission loss or propagation loss

if

hypothetical loss-free isotropic antennas with no orientation, polarization, or

multipath coupling loss were used at the same locations at the actual antennas, (2,29)a

G G G

pf

Path antenna gain in free space,

(2. 3 2).
,

pm
PP
,

Long-term median value

Path antenna power gain,

of

G
p

(2.36),

(2. 29).

Free space gains

of the receiving

and transmitting antenna, respectively, in

decibels relative to an isotropic radiator, (2.37),

G G

ri

G
ti

Gains of the
(III. 57).

lobe of receiving and transmitting antennas,

respectively,

Maximum
system

value,

expressed
.

in decibels,

of the operating gain of the receiving

21).

for

frequencies in the receiver pass band, (V.

7. 1.

G(h)

Residual height gain function, figure

)

G(h

The function G(h)

)

for the transmitting or receiving antenna. for the transmitting and receiving antennas,

G(h ),G(h

The function
(7.5).

G(h)

respectively,

G(r)

Directive gain of an antenna in the direction

is

r.

The

maximum

value of G(r)

G,

section 2.3.
r

G
r

(r),

(-r)
r
(-r),

Directive gain, in decibels, of a receiving antenna in the directions

(2.32).

and

^r(^,2^
G' (r)
r

-10

1ogg^(r^^^).
gain,

Power

in decibels,

of a receiving antenna,

(2. 13).

G
t

(r)

Directive gain, in decibels, of a transmitting antenna,

(2. 13).

G'(r)

Power

gain,

in decibels,

of a transmitting antenna,

G'(r) =
t

G
t

(r)

L
et

(2. 13). 8. 6.

G(x

function used in computing diffraction attenuation,

in gigacycles per second.

(8. 1)

figures

8. 5

and

GHz
h

Radio frequency

Height above the surface of the ground as used in Height referred to sea level.
e

(3. 10),

(3. 12).

h
h

Height for elevated beams that

is

equivalent to h

for horizon rays,

(6. 10).

(III. 63)

h.

Equidistant heights of terrain above sea level,

(5. 15),

h
h

Height of the receiver horizon obstacle above sea level,

Lt

(6. 15).
(6. 15).

Height of the transmitter horizon obstacle above sea level,

Height of the intersection

of

horizon rays above a straight line between the ana,

tennas, determined using an effective earth's radius,

(9.3b) and figure

6. 1.

Height

of the

receiving or transmitting antenna above ground, assviming a

smooth earth.
to
I,

A smooth
or

earth

is

assumed

in the

curves

of figures I. 5

and

I.

26.

h
r

h
t

The height h ^

h
t

is

defined as the height of the receiving ^ ^ or transmitting ^

antenna above the average height of the central 80% of the terrain between the
antenna and
its

horizon, or above ground, whichever gives the larger value,

(6. 11).

12-9

re

h
te

Effective height of the receiving ground. o or transmitting o antenna above o

For

h
r

h
t

less than one kilometer

is used,
(6. 12).

re

h
r

h
te

h
t

For higher antennas a

correction Ah

rm
,

tm

Height of a knife edge above a reflecting plane on the receiver or transmitter

side of the knife edge,
(III.

37).

Height of the receiving antenna or transmitting antenna above sea level, figure 6.1, used in
(6. 11),

(6.15).

h
s

Elevation of the surface of the ground above

.

mean sea

level,

(4. 3).

The heights above sea level

transmitter and
its

of evenly
(6. 11).

spaced terrain elevations between the

horizon,

h h
t30

The height above sea level

of the

ground below the transmitting antenna,

h
t3

(6. 11).

The height ^

of the

horizon obstacle

= h

Lt

(6. 11).
^

Height of the crossover of horizon rays above a straight line between the transmitter and receiver horizon obstacles,
(9. 7)

figure

6. 1.

h
h'

h
h'

Heights of antenna terminals

Heights of antenna terminals
at the

and and

2
2

above the surface

of the earth,

figure

5. 1.

above a plane tangent to a smooth earth

(5.8).

bounce point

of a reflected ray,

h h

Average height above sea

level,

(5. 15).

Average height

of the transmitting antenna

its

above the central 80% of terrain

between the transmitter and

h
,

horizon,

(6. 11). (7. 6).

Normalized heights

of the transmitting in figures

III.

and receiving antennas, and

III.

h(r)

h(r

),

h(r

A A

function of function of
21.

r r

shown
or
r

20

21.
III.

defined by

(III. 50)

and shown on figures

20

and

III.

h(x)

straight line fitted by least squares to equidistant heights above sea level,

(5.15).
h(0), h(d)

Height above sea level of a smooth curve fitted

tennas,
(5.17).

to

terrain visible to both an-

and extrapolated

to the

transmitter at

h(0)

and the receiver

at

h(d),

h (x
i
i

series of equidistant heights above sea level of terrain visible to both antennas,
5.
1

section

H H oi
.

The frequency gain

function,

discussed in section
th

9. 2.

The frequency fre gain function

(III. 57).

for the

beam

intersection in a scattering plane,

(r

),

(r r^) )
2

T] The frequency gain function,

(9.
5)

as a function of

12
and
r

respec-

tively,

12-10

H H

(ri

<
s

1),

(ri

o
r\

=1)
s

Value of the frequency gain function,

than or equal to one, respectively,
t]

where

the

parameter

is less

(9. 6).

(r|

=0)

The frequency gain function when

tion of a constant

=0

-which corresponds to the

9. 5.

assump-

atmospheric refractive index, figure

Hz
i
i

Abbreviation for hertz = cycle per second.

= \f^,
(2. 19)

annex
s. s.

II.

Current

in r.

m.

amperes w^here

= 0,

1, 2.

I_, I., I

Current in r.mi.

amperes corresponding

to three

elementary dipoles in three

mutually perpendicular directions, (11.46).

j

Represents a series
(III. 3 5).

of subscripts

1, 2, 3, 4,

as used in equations

(III. 27)

to

k
k

Propagation constant,

k= k=

2it

/\

(II. 1).

Boltzmann's constant,
b

1.380 54 x 10

-23

joules per degree, (V.7).

kT

Johnson's noise power that would be available in the bandwidth b cycles per

second at a reference absolute temperature

(V.7)..

288.37

degrees Kelvin,

km
kw

Abbreviation for kilometer.

Abbreviation for kilowatt.

K K

A A

frequency-dependent coefficient,

(3.8).

parameter used

in

computing diffraction attenuation,

is a

function of the

effective earth's radius,

tion,

carrier frequency, ground constants, and polarizaIII. 4.

figure 8.

and annex
of the

The decibel ratio

root-sum-square

of

Rayleigh components

of a

received
2

signal relative to a constant or power-fading component,

annex V, section V.

and figure V.

K K K K K
K

h
o

The diffraction parameter K for horizontal polarization, annex

III.4.-

An

arbitrary constant in the systems equation, (V. 22).

The ratio K for an unwanted signal, annex V.

The diffraction parameter K for vertical polarization, annex

III. 4.

A
K
2

frequency and temperature-dependent attenuation coefficient for absorption

(3. 13)

within a cloud,
,

and table 3.1.

K
s

K
t

Values of the diffraction parameter ^

a,,
1

for

corresponding earth's

fa

radii
K(a), K(8497)

a
2

a
r

a
s

a, (8.8) to
t
^ '

(8. 13). '

^

The diffraction parameter K for an effective earth's radius

a =

a,

and for

8497 km.

K(f

frequency-dependent coefficient used in computing the rate of absorption by

(3.9a)

rain,

and figure 3.8.

K(N
s

function of the surface refractivity,

N
s

used in computing F(9d),

(III.

46)

12-11

(N
s

),

K (N
1

),

K
2

(N
s
,

Functions of the surface refractivity,

48).

N
s

used in computing

F(ed)
jf

(III.

Used as
pedance

a subscript to indicate a load,

of a load at a

for example,
v,

represents the im-

radio frequency

(2.4).
is

range of eddy sizes or layers; the radio wave scattered forward

"
i
,

most

af-

fected by a particular range of "eddy sizes,

or by layers of an average
(IV. 1).

thickness

i/2,

that are visible to both antennas,

The effective loss factor

for a receiving antenna,

or the reciprocal of the

power

receiving efficiency, (2.3),

It

erv

The effective loss factor

defined as the ratio p

for a receiving antenna at a frequency

hertz,

av

/p

av

(2.9)

The effective loss factor

power radiation

for a transmitting antenna,

(2.3).

or the reciprocal of its

efficiency,

S.

The effective loss factor

hertz,
(2. 10).
.

for a transmitting antenna at a radio frequency

mv
o

A mismatch

loss factor defined by (2. ^

^

7).
'

i
Li

Scale of turbulence, (IV. 19).

Transmission loss expressed

Basic transmission loss,

in decibels,

(2. 2).

Lb Lbd L bf
L
.

(2. 28)

and

(2. 29).

Basic transmission loss for a diffraction path, (7.3), (7.4).

Basic transmission loss in free space,
(2. 31).

L bm
bsr

Hourly median basic transmission loss,

Reference value
scatter,
(9. 1).

of

long-term median basic transmission loss due

to

forward

L L

Calculated value of transmission loss,

c

Polarization coupling loss,

(2. 25).

Lcr
L,
dr

Reference value

of

hourly median transmission loss wnen diffraction and

scatter losses are combined, (9. 14).

Reference value

of

hourly median transmission loss due to diffraction,

(9. 14).

L er
L erV
L>
,

Effective loss factor for a receiving antenna,

expressed
v

in decibels,

(2. 11).

The effective loss

(2.11).

factor,

er

at a

radio frequency

hertz,

et

L etv

The effective loss factor for a transmitting antenna, expressed

(2. 3)

in decibels,

and

(2. 11).

L.

An
,

"equivalent free-space transmission loss,

of the

" (2.34).

L
ft

The decibel ratio

resistance component of antenna input impedance to

fr

the free space antenna radiation resistance for the receiving

and transmitting

antennas, respectively, (2.39).

Loss

in antenna gain for the

scattering subvolume,

(III. 57).

12-12

1j

Loss
gP

in path antenna gain,

L,

defined as the difference bet'ween basic transmis,

sion loss

maximum
gain: ^

and path loss L, or as the difference bet^veen the sum of the b o gains of the transmitting and receiving antennas and the path antenna

L
gP

L.

p Transmission loss associated with the

t

=L^-L =G+G r -G b o

db,(2.3 7).
^'^ i
'

power contribution,

(III. 55)

and

(III. 57)

ix
it

Transmission Transmission

line
line

and matching network losses at the receiver. and matching network losses
loss,
at the transmitter,

L
Li

(V. ZO)

m
mo
o

Hourly median transmission

(V. 20).
of the time with a probability

The transmission loss exceeded (100-p) percent

Q, (V.43).

L L

Path loss, defined as transmission loss minus the sum of the

space gains
of the antennas:

maximum

free

=L-G -G r
t

(2. 27).

Propagation loss, (2.41).

P

L L

pb
rr

Basic propagation loss, (2.42).

The ratio

of the actual radiation

resistance of the receiving antenna to

(2.40).

its

ra-

diation resistance in free space,

L L

rt

The ratio

of the actual radiation

resistance of the transmitting antenna to

its

radiation resistance in free space, (2.40).

The system loss expressed

s

in decibels,

defined by

(2. 1).

System loss includes

ground and dielectric losses and antenna circuit losses.

L L L

sr

Reference value
to obtain the

of

median forward scatter transnaission

loss,

used with L,

dr

reference value
of

cr

(9. 14).

um

Median transmission loss

. .

an unwanted signal, annex V.4.

12
,

L,

...

series of hourly median values of transmission loss arranged in

to the largest value,

order from the smallest

L(p)

annex

III,

subsection
(III.

7. 2.

Transmission loss exceeded (100-p) percent

The long-term median value
1),
. . .

of the time,

68).

L(50)

of

transmission loss, section

10. 3.

L(O.Ol), L(0.

L(99. 99)

Transmission loss exceeded (100-p) percent

0.
1,
.
. .

of the

time

where p^O.Ol,

99. 99,
of

section 10.3.

L L

(50)
(p)

Long-term median value

basic transmission loss, section 10.3.

Time
I.

availability of hourly

median basic transmission

loss,

annex

I,

figures

7 to I. 17.

L,

bm

(50)

Long-term median value

of

L,

(2.36).

L.(p)

Instantaneous values of transmission loss not exceeded p percent of the time,

(V.5).

12-13

L.(0.

11
1),

(0.9)

Instantaneous values of transmission loss not exceeded 10 and 90 percent

(V.4).
of all

of the time,

(p)

Hourly median transmission loss not exceeded for p percent

hours or

exceeded for (100-p) percent

of all hours,

(V,25).

L L

m (50)

Long-term median transmission

Q)

loss,

(V.2).
of the

m (p,
mo (g)
n
(p)

Hourly median transmission loss not exceeded for p percent

probability Q.

time with a

Maximum
gion,

allowable transmission loss for a grade

of the

service, (V.27).
in a

L
L

Transmission loss not exceeded p percent

(10. 5).

time

given climatic re-

(50)

Predicted median long-term transmission loss for a given climatic region,

characterized by the subscript n^ (10.4).

L
L

um (p) um (50)

Hourly median transmission loss

cent of all hours, (V. 33).

of

an unwanted signal not exceeded for

per-

Long-term median transmission loss

for an unv/anted signal,

(V.39).

Lim.

Abbreviation for limit, as used for example on figure 8.4.

m m m
m,
h
,

A A

symbol used

to designate the slope of a straight line,

(5. 15).

subscript used to identify service limited by noise, annex V.

Average refractive index gradient,

dn/dz, across a layer, (IV.

h

5).

Parameters used
\

in

computing the magnitudes R,

(III. 10).

and R

of the

smooth plane
'^

earth reflection coefficient R,

Average refractive index gradient

(IV. 5).

for the region in

which a layer

is

imbedded,

mm.
mho.

Abbreviation for minimum.

unit of conductance,
III. 1,

the reciprocal of resistance which is

to III. 8.

measured

in

ohms,

annex

figures

III. 1

mm. mv/m

Abbreviation for millimeter.

Millivolts per meter.

M
M

Liquid water content of a cloud measured in grams per cubic meter,

(3. 13).

A term A term

defined by (IV.

7)

used

in the

power reflection coefficient q 2

a
,

(IV.

6)

defined by (IV. 22) used use in defining

vo

the scattering ^ cross-section

from refractivity turbulence.

12-14

MHz
M. U.F.
n

Radio frequency in megahertz.

Abbreviation of

maximum
or
t

usable frequency.

Refractive index of the atmosphere, section 4,

The ratio a lb
o

or
/

used

to

compute n, (9.12).
(4. 1).

Atmospheric refractive index

n
1
,

at the surface of the earth,

n
2

Refractive indices of adjacent layers of

homogeneous media,
(9. 12).
,

(IV. 3).

parameter used

in calculating path antenna gain,

= (n-1)

N N

Atmospheric refractivity defined as N

The number
to (IV. 17). of layers

10

section 4.
15)

per unit volume of a scattering cross-section, (IV.

N o Ns N
v

Surface refractivity reduced to sea level, (4.3).

The value

of

at the surface of the earth,

(4. 1).

The number

of scattering

subvolumes
11).

that

make an appreciable contribution

to the total available

power, (IV.

Time
is not

availability,

the percentage of time a given value of

transmission loss

exceeded, section 10.

function of the dielectric constant and grazing 'angle used in computing the
(III. 8)

plane wave reflection coefficient,

Radio frequency signal power that would be available from an equivalent lossfree receiving antenna,
(2. 2).

-L

The available power corresponding

tropic antennas,
(II.

to

propagation between hypothetical iso-

40)
i

P_.

Contribution to the total available power from the

(in. 55)

scattering subvolume,

and (IV.

11).

Radio frequency signal power available

tenna,
(2.
1)

at the

terminals of the receiving an-

P y

Available power at the terminals of an equivalent loss-free receiving antenna

at a radio

frequency

v,

(2.9).

P
J,

Available power at the terminals of the actual receiving antenna at a radio fre-

quency
p.

(2.6).

"Instantaneous" radio frequency signal power available at the terminals of an

equivalent loss -free antenna, defined as the average power for a single cycle
of the

radio frequency, annex V.

to the

^ir' ^f r V

Power delivered

receiving antenna load, at a radio frequency

v,

(2. 5).

12-15

^n' ^ftv

Power delivered by

the transmitter to the transmission line,

at the receiver,

(2. 5).

The median wanted signal power available

annex V,

Median value

of the total noise

power

in watts,

(V. 7).

Operating sensitivity, the median wanted signal power,

isfactory service in the presence of noise, annex V.

required for sat-

Fixed value

of

transmitter power output, expressed in watts, (V. 26).

station,

Power, in watts, radiated from a wanted

Total power radiated

(V.34).

from

the transmitting antenna in a given

band

of radio

frequencies, (2.2).

Total power radiated at a frequency

tv

v,

(2. 10).

Pi
Pi tv

Radio frequency power input

to the

terminals of the transmitting antenna,

v,

(2. 1).

Total pow^er delivered to the transmitting antenna at a frequency

(2. 10).

Power radiated from an unwanted

station,

(V. 34).

Instantaneous power of an unwanted signal available to a receiving system, annex V.

Available power per unit scattering volume, (IV, 11).
Pvi

Available power per unit scattering volume for the

(IV. 12)

scattering subvolume,

Median unwanted signal power, annex V.

PjCq)

Value of "instantaneous" available power exceeded for

short period, (V.6).

100 q

percent

of a

See

p..

mr

(g)

The value

of

required to provide service of grade

(II.

g,

annex V.

Unit connplex polarization vector for the incident wave,

21).
(II. 3 5).

Unit complex polarization vector for the

incident plane wave,

Unit complex polarization vector associated with a receiving pattern,

(II. 18).

The complex polarization vector p ~^ associated with a receiving pattern and

the
P-. p^ ri Tz
'

th

incident wave, (11.3

5).

The complex receiving antenna polarization vectors p

paths between transmitter and receiver, (11.83).

for each of two ray

P(r), p^( -r)Unit

complex polarization vectors

p r
,

for the transnnitter,

(-r)
,

in the direction

(r)

and for the receiver,

P

in the direction

(2. 18),

P^l

Polarization efficiency for transfer of energy in free space at a single radio

frequency,
(2. 22)

and

(II.

29)

P P

The north or south pole in figure 6.3.

The available power from a loss-free receiving antenna which

equivalent to the actual receiving antenna,
(2. 2).

is

otherwise

The radio frequency signal pow^er available

antenna,
(2.
1)

at the

terminals of the receiving

12-16

ab

receiving Available power at the terminals of a hypothetical loss-free isotropic

antenna, assuming no orientation, polarization,
or multipath coupling loss be-

tween transmitting and receiving antennas,

(2. 28).

P
i

= 10 log p.,
i

the instantaneous

power

of a

wanted radio signal expressed

in

decibels, (V.

1).

Power, in
ir

dbw,

delivered to the receiving antenna load,

(2. 5).

Power, in
ir

dbw,
of

delivered by the transmitter to the transmission

P.

line,

(2. 5),

The component

which

is not

affected by the usually rapid phase interfer-

ence fading, most often identified as the short-term median of the available

power
'^

P.
1

(V.

1).

The hourly median value

output: ^

in
,

dbw

of the total noise

power delivered

to a

receiver

mn

= 10 log p

rnn

(V.8).
of fading

Operating sensitivity, assuming a specified type

specified type of noise, annex V.

wanted signal and a

fixed transmitter output power,

expressed in dbw, (V.26).

Total power radiated from a wanted station, expressed in dbw, (V. 34).

The

total

power radiated from

the transmitting antenna in

dbw:

P^

- 10 log p^,

(2.2).

^1

Radio frequency power input

dbw,
(2.
1).

to the

terminals of the transmitting antenna, in

Pcwer radiated from an unwanted

station,

(V.33).
at a receiving

The median unwanted signal power available

unwanted station radiating p
P.(P)
-watts,

antenna from an

annex V.

Percentage
annex V.

of

time

that a given value of instantaneous

power

is

exceeded,

P.(q)

The percentage,
P.
of

100 q,
P.(q)
is

of a

short period of time or the probability q

if

that

will exceed
P.
1

known

the phase interference fading distribution

relative to the short-term

median value P

is

known, annex V.

it

Transmitter output power which will provide

p

at least a

grade g

service for

percent

of the time,

(V. 25).

P
P

m (P)
m (50)
mo (P)
mr (g)

The hourly median wanted signal power

hours, annex V.

exceeded for p percent

of all

The long-term median

of all

hourly median values

P.
,

usually identified

also as the long-term median of

(V. 2).

P P

Observed values

of

m (p).
of a receiving

The operating sensitivity

value of

system, defined as the

of

minimum
in the

which will provide a required grade

service

g,

pre-

sence of noise alone, (V.9).

12-17

p um (p) P

The hourly median power P

of the time,

um
um

expected

to be available at least

percent

annex V.

um (50)

The hourly median power

of the time,

expected

to be available at least 50

percent

annex V.
q
,

100 q is the percentage of a short period of time or

p.
1

is the

probability that

will

exceed

p.(q)
1

for a given

median value
Y.(q)
,

m
p

which

is the

same

as the

probability that

Y.

will exceed

annex V.

A parameter
(III. 14).

used

in calculating a plane

wave reflection

coefficient,

(III. 7)

to

The ratio q

= r
2

s r
1

used

to

compute
2

The power reflection coefficient,

AH , (9.5). for a tropospheric layer is approxiinfinitely

mated by

(IV.

6)

The plane wave Fresnel reflection coefficient for an

boundary, (IV.
3).
8.

extended plane

Q
Q(P)

Service probability, discussed in subsection V.

Service probability corresponding to the time availability p, annex V.

)

Q(z

mo

Service probability expressed as a function of the standard normal deviate

z

mo

(V.44).
of the direct

r r r
r

The length in free space

Radius
of curvature,

ray path between antennas, figure

5. 1.

(7.9).
2.

Resistance of an antenna, section

Magnitude

of the

vector

r = r r
r, 9,

in the direction
(\>,

r(9,

cj)) ,

and a coordinate

of

the polar coordinate

system

annex

II.

Effective distance for absorption by oxygen in the atmosphere, (3.4) figures

3.2 to 3.4.
er

Effective rain-bearing distance, (3. 11) and

(3. 12)

figures

3. 10 to 3. 13.

Effective distance for absorption by water vapor in the atmosphere, (3.4).

figures 3.2 to 3.4.
"fr' "ft

Antenna radiation resistance

in free space for the receiving

and transmiitting

antennas, respectively, (2.38).

Resistance
Jtv

of a load,

(2.4).

Ratio between the hourly median wanted signal power and the hourly median
operating noise power, annex V.

specified value of

which must be exceeded for at least a specified per1Ir-

centage of time to provide satisfactory service in the absence of unwanted

signals other than noise,

annex V.

IZ-li

Length
o

of a direct
5. 1

ray between antennas over an effective earth of radius

a,

figure
r r
,

r
t

Antenna radiation resistance

spectively, (Z.38).

of the

receiving and transmitting antennas, re-

r' , r' t r

Resistance component
mitting antennas,

of

antenna input impedance for the receiving and trans-

respectively, (2.38).

Ratio between the hourly median wanted signal power and the hourly median

unwanted signal power available

r

at the receiver,

annex V, page V-1.

ur

specified value of r which must be exceeded for at least a specified peru centage of time to provide satisfactory service in the presence of a single
signal,
of

unwanted
r
r' V

annex V.

Resistance

an equivalent loss-free antenna, (2.4).

its

Resistance of an actual antenna in

,

actual environment, (2.4).

12
r
,

Parameters used
by (9.4).

in

computing

the

frequency gain function

Ho

and defined

r r

r
,

Distances w^hose
r
,

sum

is the

path length of a reflected ray, figure

5. 1.
(III.

Distances to and from the bounce point of reflected rays,

III. 9.

28) fig-

ure
r
,

Straight line distances

from transmitting and receiving antennas

x.

to a point

on the ground a distance

r r
^

from

the transmitting antenna,

r =

figure 6.4.

The vector distance between two antennas,

rr,

(11.47).
II,

A
r^

unit vector directed

away from an antenna, annex

subsection

II. 1.

r^,

Direction of the most important propagation path from the transmitter to the
receiver, or from the receiver to the transmitter.

^'^e'
r
1

cartesian unit vector coordinate system, annex

to noise ratio of

II.

mr (g)
ur
(

The minimum acceptable signal

a given grade
g

which will provide service

of

in the
r

absence
,

unwanted signals other than noise, annex V.

g)

The protection
g
,

ratio,

required to provide a specified grade of service,

annex

section V. 4.

r.

s.

Abbreviation of root-mean-square.
Location
of the

R R

receiving antenna, figure

5. 1.

The magnitude
a plane

of the theoretical coefficient

exp[-i(iT -c)]

for reflection of

wave from a smooth plane surface

(5.1).

of a given conductivity

and dielec-

tric constant,

12-19

Re Rh

An

"effective" ground reflection coefficient, (5.

1).
(III. 12)

Plane earth reflection coefficient R for horizontal polarization,

figures
III. 1

and

R m. R mr Rr R rs Ru

Rm

= 10 log r

available at the receiver output, (V.9).

m
,

to III. 8.

decibels, the

median wanted signal

to

median noise ratio

The decibel equivalent

of

mr

mr

=10

log r

mr

annex V.

Rainfall rate in millimeters per hour,

(3. 10).

Surface rainfall rate,

(3. 10).

Ru

= 10 log r

the ratio

between

the hourly

median wanted signal power and

at the receiving

the hourly

median unwanted signal power available

antenna

terminals, (V.15).

R ui
.

The ratio between the instantaneous wanted signal power and the instantaneous unwanted signal power at the receiving antenna terminals, (V.
10).

R ur R
V

ur ur ur Plane earth plane wave reflection coefficient R for vertical polarization,

(III. 12)

The decibel equivalent

of

=10

log r

figures

Til. 1

to

III. 8.

^'

'^o

Vector distances from transmitter and receiver, respectively,

to a point

R^

R.

R^

Unit vectors from the centers of radiation of the receiving and transmitting
antennas, respectively, (IV.
1)

R oi
.

point

from which power

of

is

coherently scattered or reflected, (IV. 11).

R(0,5)

R (50) R (p) m R mr (g)

R
R R
u u
(p)

mm
3. 14.

Cumulative distribution

instantaneous path average rainfall rates, figure

function of
of
of

L,

dr

-L cr

(9. 14)

figure 9. 16.
50

The value The value

R
R

exceeded exceeded
of

at least

percent of the time, (V. 29).

at least

p percent of the time, (V. 24).

The minimum value

that will provide a desired grade of service in

the presence of noise alone,

(V.9).
at least

(50)
(g)

A A

specified value of

R
R

exceeded

percent of the time, (V. 36).

specified value of

exceeded

at least 50

percent of the time, (V. 36).

to

ur

Median wanted signal

grade
g

to

median unwanted signal ratio required

provide a

service, annex

V
to

section 4.

^ur*^'

^'

^^^ required ratio

R
5).

provide service of grade

for at least

per-

cent of the time, (V.3

^uro^^^
s

The required value

of

for non-fading
s = a

wanted and unwanted signals, (V.14).

Path asymmetry factor,

/p

(6. 19).

12-20

Total
c
s

mean power

flux density,

(II. 2 5).

Mean power
The fraction
e
s

flux density associated with cross-polarization

components,

(II.

23).

of the total flux density that contributes to the available

s

power,
e
/P
,

(II.

43a)

Path asymmetry factor for beams elevated above the horizon,

e
(III.

= a e

64)

Mean power
Free space
o
s

flux density associated with left-handed polarization,

field strength in watts per

(11.

28).

square kilometer, (2.43).

Mean power
P
(II.

flux density associated with principal polarization

components,

23).

Mean
r

po-wer flux density associated with right-handed polarization,

s
,

(II.

28).

<s >
e
s(r)
s (r), s
(r)

The

statistical "expected value" of

(II.

43b).
(2. 23).

Total P
tion

mean power

flux density at the receiving antenna,

Mean power

flux densities associated with the cross

components

^
e

and principal polariza-

_^

of

in the direction
33). (IV. 8).

r,

(II.

23).

S.

Fresnel integral,

(III.

S(u), S(v) Si(r)

t

Fresnel sine integrals,

Sine integral as a function of

r,

(III. 51).
(II. 1),

Time

at the transmitter,
uf

in seconds,

T
T
o

Location

transmitting antenna, figure

5. 1.

Reference absolute temperature,

288.37 degrees Kelvin,

T(r)

Temperature

in the troposphere in degrees Kelvin.

T
s

(K)

Effective sky noise temperature in degrees Kelvin.

T. A. S. O.

Abbreviation

of

Television Allocations Study Organization.

subscript used to indicate signal from an unwanted or interfering station,

annex V. u
U(vp)

A parameter

defined by (TV.
in

9).

parameter used

computing diffraction over a rounded obstacle,

(III. 26)

and figure 7.5.

U H F
v

Abbreviation of ultra high frequency,

A parameter A parameter
rr.!

used in computing diffraction over an isolated obstacle,

defined by (IV.
'

(7. 1).

V
V.

9)

The

-th
1

scattering subvolume, (IV. 11).

of
s.
j

V.

The parameter v for each

paths over an isolated obstacle,

(III. 27).

Complex open-circuit r.m.

components,
(11.3 2).
s.

signal voltage for coherently phased multipath

The open-circuit r.m.

frequency
v,

voltage for an equivalent loss-free antenna at a

(2.8).
s.

v'

The actual open-circuit r.m.

quency
V
,

voltage at the antenna terminals at a fre-

(2.5).

12-21

V(50, d
e

paraineter used with the calculated long-term reference value,

1

predict median long-term transmission loss, figure 10.

or equations (10.4)

to

and

(III.

67).
)

Y n (50,

d
e

The parameter V(50, d

for a given climatic region characterized by the

subscript n, (10.4) figure 10.1.

VHF
w
X X X
,

Abbreviation

of

very high frequency.

7).

Half the width of a first Fresnel zone, (IV.

A A
X

specified value, the discussion preceding

(2. 14).

variable designating distance from an antenna, figure 6.4.

III.

Points at which a first Fresnel ellipse cuts the great circle plane.
III.

18 to

23.
of a load,

X,

x'

Reactance

an actual lossy antenna, and an equivalent loss-free

antenna, respectively, (2.4).

X.

The

distance from the transmitter along a great circle path, figure 6.4.

m
,

One
x
,

of

three mutually perpendicular directions,

to

= 0,

1, 2,

annex
8. 5

II.

Xq

Parameters used

compute diffraction

loss,

(8. 2)

figures

and

8.6.-

Points chosen to exclude terrain adjacent to either antenna which is not

visible to the other in computing a curve
fit,

(5. 15).
(II. 2)

X X

Axes

of a

cartesian unit vector coordinate system,

of distances

figure

II. 1.

The average
Initial

and x

(5. 15b).

X
y.

bearing from antenna terminal A, measured from true north, figure

6.3.

Terrain elevations, modified

Modified terrain elevation,
Initial

to

account for the curvature of the earth,

(6. 10).

y(x)

y(x) - h(x)

-x

/(2a),

(5. 16).

Y
Y

bearing from antenna terminal

B, measured from true north, figure

6.3.

symbol used

to

describe the characteristics of long-term fading, (V.

1)

and (V.3).
Y.

The phase-interference fading

of a

received signal, (V.

1)

and (V.3).
16). 10).

Y Y
Y'
.

Long-term fading connponent

of

unwanted signal fading, (V.

of the fading of

The phase interference component

Bearing from any point B'

an unwanted signal, (V.

along the great circle path

or of
rn

AB

figure 6.3.
6)

Y{p)

Long-term variability
and (V.4).

of

in

terms

of

hourly medians, (10.

Y.(q)

Phase interference fading evaluated for

characteristic of
p = 100 q
,

the particular

phase interference fading

of the time,

a wanted signal exceeded p percent

where

(V.

5).

12-22

Y.(q, K)
1

Cumulative distribution function for the phase interference fading

of a

wanted

signal,

(V. 12),
of

Y
i

(q,

Cumulative distribution function for the phase interference fading

an un-

wanted

signal,

(V. 12).

Y
Y
n

(p)
(p)
- p)

Variability of hourly

median transmission

loss,

(V. 29).

Variability of the operating noise factor,

(p)

= F

(p)

(50),

(V.32).

Y (100

Value

of

(p)

exceeded (100
(V.31).

- p)

percent

of the

hours or not exceeded p

percent

of the hours,

Y Y

o o

(p)
(p,

Basic estimate of variability in a continental temperate climate, figure 10.2.

d
e
)

Basic estimate of variability as a function of effective distance, (10.

6)

fig-

ure 10.2.

(p)

Variability of the ratio of wanted to unwanted signal, (V.38). Variability of an unwanted signal, (V.39).

u (p) Y (100-p)
z
z
SLv

Value

of

(p)

exceeded 100
(

- p)

percent
6)
.

of the time,

(V.38).

Thickness

of a

tropospheric layer, (IV.

(2.4).

Impedance

of a load,

z z

mo
o

A
A A

standard normal deviate, (V.43).

of a

The thickness
P

tropospheric layer, (IV.

50).

6).

z z

standard normal deviate, (V.

uc

standard normal deviate corresponding to the total variance

53).

2
cr

uc

(p)

of

an

estimate of the service criterion, (V.

z
z'
z'
''

Impedance

of of

an equivalent loss-free antenna, (2.4). an actual lossy antenna, (2.4).

of
z'
,

Impedance

The conjugate

following

(2. 5).

Z Z
Z.

Great circle path length between antenna terminals

and

B, figure u
,

6. 3.

The difference between the two random variables Y and Y

(V. 16).

The difference between the phase interference fading components

Y.

and

Y
Z'

ui

.,

(V. 11).
B'
,

Great circle path distance bet-ween an antenna and an arbitrary point

figure
6. 3

(p)

Approximate cumulative distribution function

Y
-

for the

random variable Z =

(V. 17).
of

(50)
<,

The median value

)

(p)

by definition equal

to

zero,

Z.(q,

The cumulative distribution function

leigh distributed,
(V. 13).
of

for the special case

where

Y.

is

Ray-

Z. (q,

>,

The approximate value

u
)

Z.(q,

<,

<)

see (V.

12)

and Table

V..2.
,

Z. (q, K, la

Approximate cumulative distribution function

(V. 12).

of the

random variable Z
1

12-23

The parameter a

is

defined in equation (3.9b) and plotted as a function of fre-

quency on figure 3.9.

a
e
,

The angles between the "bottoms"

e

of transmitting or receiving
(III.

antenna beams

or side lobes and a line joining the antennas,

ei

61).

Ki

Angles a
e

and

p e

for the

lobe of an antenna pattern,

is

eo 'Peo

When beams

are elevated sufficiently that there

- a e

no bending of the ray due to

;

atmospheric refraction a ^
considered a
a o
a a
,

eo

^e

- Q

^eo

(III. 60)
(III.

when ray ' bending ^ must be AP

and p
e
,

are computed using

61).
,

The angles a
o
.

oo
,

oo

modified by the corrections Aa

of
j

(6. 19).
(III.

.,

P
,

The angles a
o

OJ

OJ

made by each

rays,

over an isolated obstacle,

a line
(6. 18)

36).

oo

oo

The angles between a transmitter or receiver horizon ray and

tween the antenna locations on an earth
of effective radius,
a,

drawn befigure
6. 1.

a
OI

a
02

The angles a
o2

and p

for each of four rays over an isolated obstacle,

(III.

36).

(fGHz)
^oo

The function a

in (3.9b) as a function of frequency in

GHz,

figure 3.9.

Differential absorption in decibels per kilometer for oxygen under standard conditions of temperature and pressure,
(3.4).

^r

Rate of absorption by rain, (3.8).

Surface value of the rate of absorption by rain,
(3. 11).

^rs
y 'wo

Differential absorption in decibels per kilometer for water vapor under standard

conditions of temperature and pressure and for a surface value of absolute hu-

midity of
Y(r)

lOg/cc, (3.4).

Differential atmospheric absorption in

db/km
r
,

for a path length

(3.1).

v
r

(r)
(h),

Differ ntial rain absorption along a path

y

(3. 7).

v
o

(h)

Differential absorption in

db/km

for oxygen and water vapor,

respectively,

as a function of height,
r(r)
6

h,

(3.3).
of path distance

Absorption coefficient as a function

r,

(3.Z)

and (3.6).

parameter used

in

computing the

first

Fresnel zone

in a reflecting plane,

(III. 18).

6 6

The effective half-power semi-beamwidth

e

of

an antenna,

(2. 15)

and annex

III.

The effective half-power semi-beamwidth

rected out of the great circle plane, annex

of

an antenna that

is

elevated or di-

III. 6.

The semi-beamwidth
5

of

an equivalent

beam

pattern with a square cross-section,

5n/!t7~4,

o
5
,

annex

III. 6.

6
t

The effective half-power semi-beamwidth ^

antennas, respectively, (9.11) and
(9. 12).

for the receiving 6 and transmitting 6

rwo

two

Azimuthal eq equivalent semi-beamwidths with square cross-section,

figure
III.

(III. 58)

23.
(III. 58)

angle equivalent semi-beamwidths with square cross-section, Vertical ang]

figure
III.

23.

12-24

6 6 6

w
wo
z

Azimuthal semi-beamwidth,

(Z. 15).
III. 6.

Azimuthal equivalent semi-beamwidth with square cross-section, annex

Vertical angle semi-beamwidth,
(2. 15).
III. 6.

zo
,

Vertical angle equivalent semi-beamwidth, annex

A(3

Aa

Correction terms applied to compute a

o

(6. 19)

figure 6.9.
(5. 19). (6. 12)

Depression
c

of field strength

below smooth earth values,

Ah
A
j

correction term used to compute the effective height for high antennas,
6. 7.

figure

The

value of Ar

where Ar

= r
1

+ r^
2
its

(III.

27)

and

(III.

29).

An
Ar

The deviation

of refractive

index from

expected value, (IV. 20),

r
,

The path length difference bet-ween a direct ray,

and a reflected ray,

Ar A

= r
1

r.,

- r

(5.4),

(5.9)

and

(7. 1).

Auxiliary functions used to check the magnitude of error in the graphical determina-

"r

-2
tion of diffraction attenuation,
(8. 5)

figures

8. 5

and 8.6.

AH o AN
A

correction term applied to the frequency gain function,

(9.

5)

and figure 9.4.

The refractivity gradient

from

the surface value,

to the value of

at a

height of one kilometer above the surface, (4.2).

,

A A

Ray path differences between

a direct ray and a ray path over a single iso(III.

lated obstacle with ground reflections,

28) figure (III. 9).

A,

Ir

2r

Ray path difference between straight and ground reflected rays on A 3r 4r either side of an isolated obstacle, (III. 31, (III. 3 7) figure (III. 31).
,

Aa (N
o
s

),

A3 (N
o
s

The correction terms Aa

AS

for values of
o

N
s

other than 301,

Aa

GO
(301),

(6.21) figure 6. 10.

(301)

AP

The correction terms Aa

AP O

for

N =301,
S

(6. 21)

read from

figure 6.9.

Ah(h
r

N
s

),

Ah(h
t

N
s

The correction Ah as a function

h
r

of

N
s

and

of

receiver and trans-

mitter heights

and h
t

(6. 12)

figure 6.7.
(IV. 20).

<An>
<(An) >

The expected value

The variance

of refractive index,

of fluctuations in refractive index,

(IV. 19).

Ratio of the dielectric constant of the earth's surface to the dielectric constant of
air,

figures 8,

and 8.2, annex

III.

4. 64)

A
,>
^^.1

small increment as defined by

of the

(11.

and used in

(II. 65),

(II. 72)

to (II. 75).

Angle between the axis

main beam and

the axis of the first side lobe of

an

antenna pattern, figure 111,22.

12-25

twl

tw2

Azimuth angles
to the

of the first

and second lobes

III.

of a

transmitting antenna relative

main beam

axis,

figure

23.
of a

tz

,6
1

Elevation angles of the first

tz2
to the

and second lobes

III.

transmitting antenna relative

main beam

axis,

figure

23.
(III. 60),

t,

The angle that a scattering plane makes with the great circle plane,
and figure
III.

(III. 61),

22.

s
n

se

A A

function of h function of h

antenna beams directed above the horizon or a'way from the great circle plane,
(III.
r|

OS es
and and

N N

used in computing F used in computing F

oo oioi
and

(9.3)

and figure 9.2.

and

for scattering

from

64).
r\

Characteristic impedance of free space,

o

= 4itc. 10

-7

120it

ohms,

(II. 5).

The angular distance,

9,

is the

angle between radio horizon rays in the great

(6. 19).

circle plane defined by the antenna locations,

9
9,

A
b

polar coordinate,

(II.

56).

Angle

of elevation of the
(III.

lower half power point

9,

of

an antenna

beam above

the

horizontal,
9
9,
,

62).

See

and

f ( 9,

9
.

Values
.

of

9
9,

for the receiving and transmitting antennas, respectively,

(III.

61).

bri
e

9,

Values

of

bti

for the

beam

intersection,

(III. 59).

The angle between radio rays elevated above the horizon and/ or away from the
great circle plane,
(III.
i

64).

The angle
ei

9
e

at the

intersection of radio rays elevated above the horizon

(III. 57).

and/ or away from the great circle plane,

9 9
ei

"

Horizon elevation angles

^'
, , . .

at the receiver

and transmitter, respectively,

. .
.

(6. 15).

,9

ez

en
III.

The angle
22.

9
e

for the first,

second,

th

intersection of radio rays,

figure
9

Angle
(5. 12).

of elevation of a direct

ray relative to the horizontal at the lower antenna,

See

9,

and

f ( 9,

9,

hr

ht

Angle

of elevation of a knife
(III.

edge relative to the horizontal at the receiving or

transmitting antenna,
9.

38).
j

Angle between direct and/ or reflected ray over a knife-edge, where

as shown in figure
III. 9.

1, 2, 3,

Angles defined
jr
9,,
J

in

(III. 29),

where

1, 2, 3, 4,

-which are added to

to

determine

9. =
J

9 + 9, jr =

Ir

,9

2r

,9

3r

,9

4r

Values

of

9.

for

1, 2, 3,

4,

(III. 29)

jr

Angle

of elevation

above the horizontal, figures 3.2

to 3.4.

12-26

oo

Angle between radio horizon rays, assuming straight rays above an earth
fective radius,
a,

of ef-

figure
of a

6. 1.

or

The angular elevation

ot
(6. 16),

horizon ray at the receiver or transmitter horizon,

figure

6. 1.

9,9,9,9
Jc

The angle between rays from the transmitting and receiving antennas over
III. 9.

an isolated obstacle with ground reflections, figure

A wave number
X.

direction defined by
length,
(9. 12)

(IV. 1).

Free space radio wave

The ratio
'

used for example

and figure
r\
,

in (2. 16).

|j.

6/5

used in

9. 8.

parameter
(9. 12)

that is half the value of

used in computing loss

in

antenna gain,

(9. 11),
V V
i
,

and figure

9. 7.
(2. 4)

Radio frequency in hertz,

V

to

(2. 12).

Limits

of integration (2. 11)

and

(2. 12)

chosen

to include essentially all of the

wanted signal modulation side bands.

n
p
p

constant,

tt

= 3.14159264.

Correlation coefficient bet-ween two random variables,

Index of curvature for the crest curvature of a rounded obstacle in the great
circle path direction,
(7. 8).

p..
p
'^i

The correlation betw^een variations due

to

sources
,

and

j,

(10.8).

a
r

The correlation betw^een variations Y and Y

a
,

(10.9). (10.9).
of

p
'^i

The correlation between variations Y and Y

The norinalized correlation or covariance between path-to-path variations

P
p p

m (50)

and P

um

(50)

(V.45).

tn

The long-term correlation between P

The long-term correlation between

'^tu
(T

m P m

and F
and

op

(V.27).
,

um

(V.34).

Surface conductivity in mhos per meter, figures

(p)

8. 1

and 8.2, annex

2
cr

III. 4.

cr

The standard deviation corresponding

The path-to-path variance
of

to the

variance

(p).

2
(T

(p)

observed from predicted p-percentiles

of

of trans-

mission loss for a large number

values for all
2
cr

randomly different paths with a given

set of

parameters used

in the prediction process,

(V.36) figure V.4.

(50)

The path-to-path variance

of the difference

between observed and predicted longThe corresponding standard deviation

term median values

is
cr
(T

of
6.

transmission loss.

(50)

annex V.

The root-mean-square deviation

to a
2

of

great circle path terrain elevations relative

(5. 1).

smooth curve

fitted to the terrain,

cr

op
2
cr

(p)

Total variance of any estimate of the service criterion for service limited only

by external noise, (V.43).

uc
(p)

The corresponding standard deviation

is

cr

(p).

P Total variance of any estimate of the service criterion for service limited only

by interference from a single unwanted source, (V.45).

The corresponding

standard deviation

is

cr

uc

(p).

12-27

ur

Variance

of the

estimate

ur

(p,

g)

(V.45).
20
of

A symbol
'

to

represent the summation

values of h
1

terms, as in
i

(5. 15)

where

1=0

.S^ h.
1

means

the
T T

sum

of all

from

to

= 20.

The amount a radio ray bends

Delayed
timie of the phasor,

in the
(ir)

atmosphere,
,

(III.

62).
is the

exp

where
c

t = k(ct-r) is the r

time of recep-

tion at free-space

radio-wave velocities,

free space velocity of radio-

waves,
(II. 1).

is the

time at the radio source, and

is the

length of the radio ray,

T
T

a
,

Time element defined by

T az

(II.

The time element

ai

a corresponding to direct and ground-reflected

(11.79).
is a

70) as

= IOtt

cos

9.

waves

at

the receiving antenna,

T.
1

time-independent phase which

function of

r,

(II. 9),

(11.31).

T.

in

The time-independent phase

II. 6.

for the

component

of

an incident wave, annex

T.

T.

The time-independent phase

Initial

for two

components
of

of

an incident wave, (11.85).

m
,

phase
1, 2,

of the

current supported by one

elementary dipoles, where

m
T
,

= 0,

(11.46).
of the

Initial

phases

currents supported by three elementary dipoles, (11.46).

is a

Timie-independent phase which

function of the ray path, including allowances

for path length differences and diffraction or reflection phase shifts, (11.31).
X
^

The phase function

pE

pn'

pi

on an antenna from
T
r
T

first, and second plane for the n P a single source, (11.32) and (11.8 5).
t
,

th

wave incident

Antenna phase response for the receiving antenna,

rn
,

(II.

16). first,

T ri

T rz

The antenna phase response,

t r

for the
(11.32)

and second plane

wave incident on
T
T
,

the receiving antenna,

and (11.81).
(II.

Antenna phase response for a transmitting antenna,

T
ti
,

16).

T
t2

tn

The antenna phase response

(11.3 2)

th
t
t

for the

first,

and second plane wave,

and

(11,8 5).

T,
,

Phases associated with the electrical

d,

field

components

e,,
9,

(II. 7)

t(9,

and travels d b kilometers through an atmosphere characterized by a surface refractivity N

)

Bending of a radio ray that takes

off at

an initial angle

(III.

61).

<})

One

of the polar coordinates, of

r,

9,

(j),

(II. 56)

and figure

II. 1.

<\){v,0)

Component
figure
7. 1,

phase lag due

(III.

to diffraction

over an idealized knife edge,

(7. 13)

and
of

30).
to diffraction
7. 5

tJ3(vp)

Component

phase lag due

(7. 13)

over an isolated perfectly-conducting

30).

rounded obstacle,
4>(0, p)

figure

and

(III.

The component

of the

phase lag

of the diffracted field

over an isolated perfectly(III.

conducting rounded obstacle for

v =

0,

(7. 13)

figure 7.4 and

30)

12-28

$
4

A
.

Latitudes of antenna terminals

and B,

(6. 1)

to (6.9) figure 6.3.

Latitude of an arbitrary point along the great circle path

from

to B,

(6. 7).

The phase lag

of the diffracted field for the

(III.

ray over an isolated perfectlyj

conducting rounded obstacle

$(v,
p)

30a),

where

1, 2, 3,

4.

The

total

phase lag

of the diffracted field

(7. 13).

over an isolated rounded obstacle with

reflections
$(v,
0)

from

terrain,

The

total

phase lag
(7.13).

of the diffracted field

over an ideal knife edge with ground

reflections,
$.(v,
p)

The phase lag

ray,

of the diffracted
,

ray over an isolated rounded obstacle for the

.(v, p)
J

s ^

(III.

30). th
j

$ $

.(v, 0)

The phase lag over an ideal knife edge for the

,

ray,
1, 2, 3,

(III.

30).
32)
of a

S
^

$
J

1
ijj

The phase lag $

of a

.(v, p)
J

for values of

4,

(III.

The grazing angle

earth,
(5.
1)

ray reflected from a point on the surface

or grazing angle at a feuillet, annex IV.
5. 1,

smooth

figure

5. 1,

di

m
P
,
>li

Minimum

grazing angle, section

di

The acute angle between principal polarization vectors

and
P

pr

(2. 26).

th

The acute angle,

P2

di

for each of two waves,

of the

(11.8 5).

pi
4j
,

4"

The angle between the plane

lower half-power point

(III.

of

an antenna

beam

and the receiver or transmitter horizon plane,

4i .,

60).
(III. 59).

4^

The angle
Angle
edge,

i\i

or
r

\\i

for the
t

lobe of an antenna pattern,

of a reflected

ri

ti
lJj

of reflection at the
(III. 3 6)

ground

ray that passes over a knife-

figure

III. 9.

n
r2
,

The half-power beamwidth,

Q,
t

= 25,

(9. 10)

and figure

III.

22.

The half-power beamwidths

spectively,
(9.10).

of the

receiving and transmitting => o antennas, re-

Q,

ro

rl

n
to

Half-power beamwidths corresponding

to

25
o

26,
1

for the receiving

tl

and transnnitting antenna patterns, respectively, figure

III.

22.

12-29
GPO
849-642

^4w