Free-Space Laser Communications: Fundamentals, System Design, Analysis and Applications

Free-Space Laser Communications:
Fundamentals, System Design, Analysis

and Applications
Dr. Arun K. Majumdar

a.majumdar@IEEE.org
105 W. Mojave Rose Ave.
Ridgecrest, California 93555,
USA
Lecture Series:1
Brno University of Technology, Brno,
Czech Republic
December 1-6, 2009
Copyright © 2009 Arun K. Majumdar 1

Course Outline
1. Introduction
• Definition of free-space laser communications
• Why optical communications? Optical / RF comparison
• Basic block diagram
• Applications overview
2. Major sub-systems for laser communications systems and Link

Analysis
• Laser Transmitter
• Modulation methods
• Transmitting optics

Course Outline
• Optical Receiver
– Photo-detectors
– Pre-amplifier
– Optics, Fiber Optics
• Acquisition, Pointing, and Tracking
3. Optical Signal Detection

• Direct Detection: Detection statistics
• SNR Bit-Error-Rate (BER) probability
• Coherent Detection
4. Atmospheric Channel Effects

• Attenuation
• Beam Wonder
• Turbulence (Scintillation/ Fading)
• Turbid (rain, fog, snow)
• Cloud-free line of sight

Course Outline
• Received Power
• Link Margin
• Data Rate
• Reliability

Course Outline
5. Basic Free-Space Laser Communications System
- Wavelength Selection
- Free-Space Lasercom Subsystems

Course Outline
6. Free-Space Laser Communications Systems Performance
• Metrics for evaluating the performance
• SNR and BER in presence of atmospheric turbulence
• Probability of Fade
• Examples
– Terrestrial (Horizontal Link)
– Uplink
– Downlink

Course Outline
7. Mitigating Turbulence Effects
• Multiple Transmitters
• Adaptive Optics
8. Animation Show
9. Summary: Improvement of Lasercom Performance
REFERENCES

Objectives
At the end of the course participants will be able to:
• Understand basic operational principles of free-space
laser communications
• Describe lasercom systems using fundamental design
concepts
• Describe atmospheric propagation effects on lasercom
performance
• Quantitatively evaluate degradation in system
performance as a function of various atmospheric
parameters
• Perform link budget analysis and calculate Bit Error Rate
(BER)

WHAT IS THE BIG PICTURE OF FREE-SPACE LASER
COMMUNICATIONS?
Air-to-Air
• Air-to-Ground
•
Ground-to-Air
•
Ground-to-Ground
•

Why Optical Communications?
• The main reason is the potential increase in
information and power that can be
transmitted
• Note: For a circular lens antenna of
diameter d, transmitting an electromagnetic
wave of wavelength λ, the antenna
transmitter gain:
• Gain, Ga=16/ӨT2
• ӨT = transmitting divergent angle ≈ λ/d, so
that Ga = 16 d 2/ λ2
Example: 6 in lens antenna at 6x10^14 Hz has
122 dB Gain, compared to an improvement
over an RF antenna of 210 ft (~ 64 m)
generating gain of 60 dB !
Optical and RF comparison
Antenna Gain
Comparison for
Optical and RF

Major sub-systems for laser
communications systems and Link
Analysis
– Laser Transmitter
– Transmitter Optics
– Beam Propagation
– Optical Receiver
– Receiver Optics
– Acquisition, Pointing and Tracking

Modulation Method
• Figure. Selected Modulation Formats

Optical Receivers
The purpose of the receiver is:
(i) To convert the optical signal to electrical
(ii) Recover data
DIRECT DETECTION
Figure. Typical direct detection digital optical receiver

Coherent Detection
For detecting weak signal, coherent detection scheme is
applied where the signal is mixed with a single-frequency
strong local oscillator signal. The mixing process converts
the weak signal to an intermediate frequency (IF) in the RF
for improved detection and processing.

Optical Receivers
Receiver performance
The Signal-to-Noise-Ratio for an optical receiver containing a
p-i-n diode preceded by an EDFA of the receiver can be
calculated as:
SNR =Ip2 / (σ2T + σ2s+ σ2sig-sp + σ2sp-sp )
The Bit-Error-Rate (BER), is the probability of incorrect bit
identification by the decision circuit of the receiver. With
equal occurrence probabilities of logical “1” s and “0”s , and
Gaussian noise, the BER is given by:
BER = (1/4)· [erfc{(I1 –ID) / σ121/2 } + erfc{(ID –I ) / σ021/2 }]
0
Where I1 and I are the average signal currents at the input of the
0
decision circuit for a “1” and “0”, respectively.

σ1 and σ0 are the rms noise currents for a “1” and “0”. ID is
the threshold current value of the decision circuit. An
adequate choice of ID is: ID = (σ0 I1 + σ1 I ) / (σ1+ σ0)
0
Thus, BER = (1/2) erfc(Q/21/2 ), where Q = (I1- I ) / (σ1+ σ0)

0

Free-Space Laser Communication:
the Atmospheric Channel
Laser power
reduction due
Transceiver A to atmospheric Transceiver B
channel effects
Potential atmospheric effects:

 Physical obstructions – birds, bugs, tree limbs, other
 Absorption – primarily due to water vapor and carbon dioxide
 Scattering – dust particles, water droplets (fog, rain, snow)
 Building sway – wind, differential heating/cooling, ground motion
 Scintillation – atmospheric turbulence

Various atmospheric effects relevant to free-
space laser communications

The Atmospheric Channel: Absorption
• Absorption depends on water vapor and

carbon dioxide content of the atmospheric
channel, which in turn depends on
humidity and altitude
• Transmission “windows” occur at visible

wavelengths and in the ranges 1.5-1.8 µ m,
3-4 µ m, and 8-14 µ m.

The Atmospheric Channel:
Scattering
• caused when wavelength collides with scattering particle

• no loss of energy, only directional redistribution Atoms & molecules
• physical size of particle determines type of scattering:
Aerosols & droplets
particle << λ → Rayleigh scattering (symmetric)
particle ≈ λ → Mie scattering (forward direction)
particle > λ → extreme forward scattering
I(z)  z 
Transmittance (scattering + absorption):τ = = exp − ∫ γ dz
Io  0 
No smoke Weak smoke Heavy smoke

BER 10-8 BER 10-4 BER 10-3
Communication Transmitter
Transmitter (155Mb/s)
The res
the wav
naked e
atmosph
the follo
Atmospheric Turbulence Effects on Propagation
gradient
Humiditygradient
Temperature
Fluctuations of the refractive index are locally homogeneous and

isotropic: Dn (r ) =  n(r) − n(0) 2 = Cn2 r 2 / 3 , lo < r < Lo
 

Turbulence-Induced
Refractive Index Fluctuations
December 15, 2002 December 16, 2002 December 17, 2002
February 8, 2003 February 12, 2003 February 13, 2003

Atmospheric Models
2
2 ν   h   h   h 
Cn (h) = 0.00594  (10− 5 h)10 exp − −16
 + 2.7 × 10 exp −  + A exp − 
 
27  1000   1500   100 
2  h   h   h 
Cn (h) = 8.1 6× 1 0− 5 4h1 0 e x p −  + 3.0 2× 1 0−1 7
e x 
p −  + 1.9 0× 1 0−1 5
e x −
p 
 1 0 0 0
  1 5 0 0
  1 0 0


•
Copyright © 2009 Arun K. Majumdar
SL
25
CLEAR1 Model
10.34 < h ≤ 30
2
log10 (C n ) = A + Bh + Ch 2 + D exp{−0.5[(h − E ) / F ] 2 }
where A= -17.0577, B= -0.0449, C= -0.0005

D= 0.6181, E= 15.5617, F= 3.4666

Propagation of a Gaussian Laser Beam in Free Space
Receiver beam size:

1
w(z) = w (rˆ 2 + ẑ 2 ) 2
o
Receiver radius of curvature:

z (rˆ 2 + ẑ 2 )
R (z) =
rˆ(1 − rˆ) − ẑ 2
Transmitter focusing parameter:
R -z
rˆ(z) = o
Ro
Normalized diffractive distance:
z
ẑ = ẑ d = kwo2 / 2
ẑ d

Goal: Maximization of Intensity on Receiver
F o c u se d B e am
free space
we ak
Average Peak Power Density

turb ule nc e
Average Power Density
turbulence
s tro ng
turb ule nc e
-2 -1 0 1 2 0 1000 2000 3000 4000 5000

Beam Profile on T arget (m ) Pro pa ga tio n D is tan ce (m)

Normalized variance of irradiance (I)
fluctuations:
2
I2 − I
σI =
2
2
I
For weak scintillation regime, the irradiance

variance is proportional to the Rytov
variance for a plane wave,
σ1 2 =1.23C n 2 k 7 / 6 L11 / 6
The three-dimensional power spectrum of

refractive index fluctuations is the original
Kolmogorov spectrum:
φn (κ) = 0.033Cn 2κ−11/ 3 ,
1/L0<< κ << 1/ l0
where κ =2π/turbulence size

where k is the wave number.
The modified von Karman spectrum is: (taking into account of inner
and outer scales)
exp( −κ 2 / κm )
2
φn (κ) = 0.033 Cn 2
, 0 << κ < ∞
(κ2 +κ0 )11 / 6
2
κml0 = 5.92 and κ0 = 1/L0
Figure shows the power spectrum of refractive index fluctuations for various turbulence
models:
For Weak turbulence regime:
For plane wave: σI 2 ( L) =σ1 2 =1.23C n 2 k 7 / 6 L11 / 6 ,

What is Lens Aperture Averaging?
Aperture-Averaging Factor A: describes the percent decrease in intensity

fluctuations due to having a receiver that is larger than a point.
Example: Log-Irradiance Variance = 1.0 A = 0.75 (σ ′I = Aσ I ) 2 2
Aperture-Averaged Log-Irradiance = (1.0)(.75) =0.75

25% reduction in scintillation
Fluctuations in intensity are “averaged” over receiving aperture of
diameter D:
Aperture Averaging Model*:

 -D2 x 2  ρ 2 2    − 1
2
16
1

A = ∫ x dx exp  2+ o − ρ φ
  cos ( x ) − x 1 − x 2  1/ 2 

o
π 2
 w 2 ẑ 2 2

w (z)  
  
0  o 
ρ
o 
*Ricklin and Davidson, JOSA A 20(5), 856, 2003.

Behavior of the Aperture Averaging Factor A
• Aperture averaging can significantly reduce intensity scintillations

• Scintillations increase with path length
• For smaller aperture sizes in stronger turbulence, scintillations can be
severe
• Doubling the receiver aperture size decreases scintillations by about a
factor of two
• Doubling the wavelength roughly doubles the aperture size required to
“average” scintillations
• Degree of beam divergence does not play a significant role
1.0 1.0
C n2 = 10e-14
C n2 = 5x10e-14 C n2 = 10e-14
0.8 0.8 C n2 = 5x10e-14
L = 2000 m
Aperture Averaging Factor A
λ = 1.55µ
Aperture Averaging Factor A
L = 2000 m
0.6 0.6 λ = 0.785µ
0.4 0.4
0.2 0.2
0.0 0.0
0.00 0.05 0.10 0.15 0.20 0.00 0.02 0.04 0.06 0.08 0.10
Lens D iameter (m)Copyright © 2009 Arun K. Majumdar Lens D iameter (m) 32
Coherence-Induced “Artificial” Aperture Averaging
Aperture Averaging: Fluctuations in intensity

are “averaged” over the receiving aperture of
diameter D
“Artificial” Aperture Averaging: reduce the beam

coherence length rather than increase the receiving
lens diameter

Aperture-Averaged Log-Intensity Variance
D ivergent Beam Divergent Beam

10 0.050
λ = 0.785µ m
ζs=1 z = 2000 m
λ = 0.785µ m
Aperture-averaged Log-intensity Variance

0.045 wo = 2.5 cm
8 z = 2000 m
wo = 2.5 cm Cn2 = 1x10-14 m-2/3
Α = .00 5 , 5 ζ s = 1
Log-intensity Variance
Cn2 -14
= 1x10 m -2/3 D = 10 cm
6 0.040
ζ s = 20
Α = 0 s =. 0 5
20 1 ,
0.035
4
Α = 0 s =. 0 4
50 9 ,
= ζ s = 50
0.030
2 Α = 0 s =. 0 4 5
1000 ,
ζ s = 1000
0.025
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
0 5 10 15 20 25 30 35 40 45 50 55 60 65
ρ radial distance from center of receiver aperture (cm
ρ radial distance from beam center (cm )
log-intensity variance averaged

log-intensity variance showing
over 10 cm diameter aperture
off-axis fluctuations (point
receiver) (σ ln′ 2 Z = Aσ ln2 Z )

Optical Communication Link
• Figure 1 illustrates the major subsystems in a complete free-space
laser communications system.
Data Transmitter Channel Receiver Data
Source Free-Space Detection

Laser Absorption Direct Detection
Scattering Optical Preamplified
Turbulence Heterodyne
Modulator Background
radiance Demodulation
Internal Incoherent/Coherent
External Optical/Electrical
Bit Rate
Detector
Coding p-i-n PD
APD
Amplifier
Decoding
Bit-Error Rate (BER)
Basic Free-Space Laser Communications System

Wavelength selection criteria
Choice of the transmitting laser wavelength will depend upon:
- Atmospheric propagation characteristics
- Optical background noise
- Technologies developed for lasers, detectors, and spectral
filters
(wind velocity of 30 m/s, and a 45º zenith angle for propagation using
Hufnagel approximation were assumed)
Free-Space Laser Communications Link
Analysis
Consider a transmitter antenna with gain GT transmitting a total
power PT Watts for a communication range, L.

Free-Space Laser Communication Link Equation,
Link Margin and Data Rate
• Received Power
Link equation combines attenuation and geometrical aspects to calculate the received optical
power as a function of range, telescope aperture sizes and atmospheric transmissions.
The link equation can be used to generate power detection curves as a function of range.
Figure shows the calculated received power as a function of range for the case of a 10 Mbit/s
bandwidth, using a LED operating at 0.85- μm wavelength, 40 mW power, 13-cm receiver,
atmospheric transmission r3eceiver4 optical efficiency of 0.2, transmitter divergence angle of
1 degree =0.0175 radians, and NEP (noise equivalent power) of the Si detector of 300 nW for
daytime operation.
(Ref. Dennis Killinger, “Free space optics for laser communication
through air,” Optics & Photonics News, October 2002)
Light Haze: low attenuation (10-4/m or 0.2 dB/Km)
Clouds similar to modertae fog- Modertae attaenuation ( 10-2/m or 20
dB/Km)

• Link Margin
Link margin describes how much margin a given system has at a given
range to compensate for scattering, absorption and turbulence losses.
The link margin is defined as: M = (Received Power Available)/
(Required Received Power)
Required Received power for a given data rate and receiver sensitivity is:
Preq = Nb. r.(hc/λ) where Nb is the receiver sensitivity (Photons/Bit), r is the data
rate, h = Planck’s constant, c = velocity of light
The Margin, M is then given by:
M = PT/[r.(hc/λ) ].(dR2/θT2L2)τatm τ TτR. (1/ Nb)
• Data Rate
The data rate is given by: r = (PT τatm τ TτR. .A)[π(θT/2)2L2.Ep. Nb. ] where Ep
is the laser photon energy=hc/ λ.
Example: For a 10 cm telescope, diffraction limited divergence = 14
μrad, transmitter peak power =200 mW, transmitter efficiency =o.5,
receiver efficiency = 0.5, and using an avalanche photo-detector with
sensitivity of 60 photons/bit for 10-8 BER , the Figure shows the data rate as
a function of range, L.

• Ref. Scott Bloom, Eric Korevaar, John Schuster and Hienz Willebrand, “Understanding the
performance of free-spaceoptics,” JON (OSA), Vol.2, No.6, 178-200 ((2003).
• Ref. E. Korevaar, S. Bloom, K. Slatnick, V. Chan, I.Chen, M.Rivers, C. Foster, K. Choi

and C.S. Liu, “Status of SDIO/IS&T Lasercom Testbed Program,” SPIE. Vol.1866 (1993).

Table 1. Link Analysis Example of a Satellite-to-Ground Laser Communication System
Parameter Value/Factor dB
Wavelength (λ) 0.635 micrometer

Range (L) 4.83 x 105 meter
Data Rate 3 Gbps
Receiver Diameter (D) 1.4 meter
Transmitter Divergence 2.07 x 10-4 radians
Angle (θT)
Transmitter Antenna 3.73 x 108 +85.72
Gain (GT = 16/ (θT)2)
Transmitter Optical Loss 0.1 -10.0
Space Loss ( S = (λ/4πL)2 ) 1.09 x 10-26 -259.61

Receiver Antenna
Gain ( GR = (πD/λ)2 ) 47.974 x 1012 +136.81
Receiver Optical Loss 0.1 -10.0
SYSTEM LOSS -57.08
Atmospheric Turbulence -11.30

Margin
Clear Air Transmission -2.08
Loss
TOTAL LINK LOSS -70.46

LINK MARGIN -6.00
DESIGN LOSS -76.46
Required Received Signal

at 3 Gbps 9.36 x 10-8 Watt -70.29 (=10 log10 9.36x10-8)
Required Laser Power at 3
Gbps = Required received 4.14 Watt (= 10 6.17/10 ) -70.29+76.46 = 6.17
signal – Design Loss

Example 2. Link Budget for 10 Gbps Laser Communication between
Satellite and Ground Station
Parameter/Item Downlink Uplink

(satellite -to- ground) (ground-to-satellite)
Wavelength 1.55 micrometer 1.55 micrometer
Laser Power 1 Watt 1 Watt
Transmitting Antenna (efficiency= 20 cm 100 cm
50%)
Antenna Gain 109.15 dB 123.13 dB
Range 38,000 km 38,000 km
Free-Space Loss -289.77 dB -289.77 dB
Receiving Antenna (efficiency=50%) 100 cm 20 cm
Antenna Gain 123.13 dB 109.15 dB
Atmospheric Loss , etc.(Absorption –10.1 dB -9.6 dB
loss: 3.0 dB, Strehl ratio due to the
atmospheric turbulence: 0.27 dB,
coupling loss for wavefront
sensing:0.5 db)
Receiving Power
Sensitivity -37.59 dBm -34.09 dB
REQUIRED POWER 70 photons/bit 70 photons/bit
MARGIN 40.47 dBm 40.47 dBm
2.9 dB 4.6 dB

EXAMPLE 3. Very Short Range through Low Visibility Atmospheric Laser Communication Link
Factor
Parameter/Item
Atmospheric Loss -200 dB/Km
Wavelength 785 nm
Range 200 meter
Data Rate 1250 Mbit/s
Peak Laser Power 1.mW
Transmit Aperture 5 cm
Transmit Divergence (at 1/e2 point) 0.5 mradian
Receiver Aperture 7.5 cm
Receiver Sensitivity 800 nWatt

Peak Laser Transmit Power -14.44 dBW
Extinction ratio degradation -0.2 dB
Pointing Loss -1 dB
Geometric Range Loss -2.50 dB
Atmospheric Loss -40 dB
Atmospheric Scintillation Fade -1 dB
Receive Optics Attenuation -1.4 dB

Bandpass Filter Loss -0.7 dB
RECEIVED PEAK POWER AT DETECTOR -61.24 dB
REQUIRED PEAK POWER AT DETECTOR -60.97 dB
LINK MARGIN AT RANGE -0.27 dB

RELIABILITY OF LASER COMMUNICATION
LINKS
• Consider the link power budget. It includes all average losses of optical power
P [dBm], which arise between the laser source and the receiving photo-
detector.
• Pt [dBm] = transmitter power, Prec [dBm] = received power, P0 [dBm]
= receiver sensitivity and Lp [dBm] = propagation loss. LM is an initial link
parameter that serves to express the reliability of the lasercom system.
• LM = Pt - Lp - P0
• The link availability is a percentage of time Tav[%], when the data
transmission bit error rate is less than its defined value. The link availability
can be expressed as by a probability that additional optical power losses LA
[dB] caused by atmospheric effects are less than link margin LM. The
attenuation of radiation in the atmosphere has a dominant share among all
losses.
• The link availability can be expressed by means of a probability density
p(α A) of an attenuation coefficient α A [dB/km] from the following
equation: αA
Tav = 100% ⋅ ∫ p(αA) ⋅ d (αA)
0
•
•
•
• where α A is the limiting attenuation coefficient value, which is given by
• α A = [LM(D)/D].1000, D being the range.

One of the possible ways to determine the distribution of p(αA) is based on long-time
monitoring of of a received signal level of a real measuring link. Another way consists in
utilizing data that was collected in the past. Visibility V[km] is the quantity to be monitored
and it serves to determine the attenuation coefficient.
Statistical distribution functions F(A< α) can be created, which represents statistical link
models. The values of the above integral can be determined from these functions for given
limiting attenuation coefficients. An example of statistical link model is shown in the following
figure.
Note that for two limiting attenuation coefficient values αA= 21 dB/km, and αA= 8 dB/km, the
corresponding link availabilities are Tav = 93% and Tav = 91% .

PROBABILITY DENSITY FUNCTIONS OF IRRADIANCE
FLUCTUATIONS
Scintillation can lead to power losses at the receiver: eventually can cause
fading of the received signal below a prescribed threshold value. Therefore
we need to know the form of the PDF to ev aluate lasercom system
performance.
Some of the PDFs:
Lognormal distribution:
   I  1 2  
2
 ln 
  + σ I ( r , L)  

1    < I ( r , L) >  2   , I>0 (nonnegative
p( I ) = exp − 
Iσ I (r , L) 2π 2σ I (r , L)
2
 
 
 
irradiance)
2α
K Distribution: p( I ) = (αI ) (α−1) / 2 K α−1 (2 αI ), I>0
Γ(α)
(1 + r )e −r
Lognormal -Rician Distribution: p( I ) =
2πσ z
 1 2 2
∞
 (1 + r )rI   (1 + r ) I (ln z + σz ) 
2 dz
× ∫ I 0 2  exp − −  2 , I>0
2σ z
2
0  z   z z
 
∞
Gamma -Gamma Distribution: p ( I ) = ∫ p y ( I x) p x ( x)dx
0
=
2(αβ) (α+β ) / 2 (α+β ) / 2 −1
Γ(α)Γ( β )
I ( )
K α−β 2 αβI , I > 0

The Probability of Error, Bit Error Rate (BER)
pI(s) = probability distribution of

irradiance
Is= instantaneous signal current with
mean value
<Ps> = mean signal value
<SNR> is the mean SNR in presence
of turbulence

DIRECT DETECTION
Signal to be detected is always mixed with noise, such as background and

detector (shot) noise. The signal in presence of noise can be detected using
thresholding technique. The signal is present if the output of the receiver
exceeds that threshold value. If noise alone exceeds that threshold, it is
interpreted as signal, which is termed as “false alarms”. If the noise and the
signal together does not exceed the threshold (even if the signal is present)
we call this “Missed detection”. The following figure depicts this concept.

CASE: NO TURBULENCE
Output current from the detector: i=iS +iN
iS ( the signal current) is given by

η
eP
iS = S
hν
2 2η
e 2 BPS
iN =2eBiS =
hν
B=bandwidth
iN
2
= σN
2
i
SNRNO TURB. =S
σN
η PS
=
2 hν B
If we take into account of the background noise, PB, we can write a more
general expression for SNR as follows:
PS
SNR NO TURB. =
2hν B

 η ( PS +PB )
 
CASE: WITH TURBULENCE
Note that iS is fluctuating and is a random variable. The mean signal

current is
ηe PS
iS =
hν
2
σSN 2 = iS 2 − iS + iN
2
2ηe 2 B PS
2
ηe 2
= ∆PS +
hν hν
2
where ∆PS
2
= PS 2
− PS
iS SNR NO TURB.
SNRTURB = = where σI 2 ( D) =A σI 2 (0) ,
σSN PS 0
+σI ( D) SNR NO TURB.
2
PS
A being the aperture averaging coefficient

Bit Error-Rate (BER) Performance
Some Basics of BER
The bit-error-rate (BER) is the probability of incorrect bit

identification by the decision circuit.
A typical requirement for optical receivers is BER < 10 -9 (i.e., less
than one error in one billion bits). The receiver sensitivity is the
minimum average received optical power required to achieve BER
= 10-9
Let us calculate the BER for an “Ideal Receiver”- light signal

with power P and B is the bit rate.
- # Photons/sec = P/hν.
- Ave # Photons/bit interval = P/(hνB)
- Poisson probability, p(n)= e -α αn / n ! where α = P/(hνB)

- P[01] = p(0) = e-α
- BER = p(1) P[01] +p(0) P[10] = ½ e- P/(hνB)
- For BER = 10-12, we need an average of 27 photons per bit
The figure shows the time fluctuating digital signal and probability
distribution centered at average signal levels I 1 and I0 (point of decision:
time wise, t = td, and signal wise I = I D )

An error occurs if I<ID for a “1” bit, or if I>ID for a “0” bit.
We can calculate the BER as follows:
BER = p(1) P[01] +p(0) P[10]
Probability of Prob. of detecting a “0”

transmitting a “1” given that a “1” was sent
(usually=1/2)
Assuming a Gaussian PDFs with variance σ0,1

we find,
ID ∞
BER = p(1) ∫ p1 ( I )dI + p(0) ∫ p0 ( I )dI
−∞ ID
1 I −I D I −I 0
= (erfc( 1 ) +erfc( D )
4 σ1 2 σ0 2
assume p(0) = p(1) =1/2
In the above equation erfc denotes the complimentary error function:
∞
2
π∫
erfc( x) = esp( −y 2
) dy
x
We can also find the optimal decision threshold that minimized the BER
from:
d(BER)/dID = 0 , and is given by: p1(ID) = p0(ID)

i.e., where the pdf for “1”s intersect the pdf for “0”s. This is a
transcendental equation for ID that has to be solved numerically. By
choosing ID so that P[1 0] = P[0 1], gives a very good approximate value
foir the optimum decision level as
σ
ID =
σ
0 I1 +1I 0
σ σ
0 +1
Q-Value and Receiver Sensitivity
It is then useful to define the Q-value as a measure
I1 −
I0
Q= and the BER is then related to the Q as
σ σ
0 + 1
1 Q exp(−
Q 2 / 2)
BER = erfc ≈
2 2 Q 2π
Q is the optical SNR. Therefore we can also write,
1 SNR
BER = erfc
2 2
Once I0, I1, σ

0 and σ
1 are found, the BER can be found from the Q.

Modulation used in a digital communication system is “binary transmission” by a
sequence of bits denoted by the symbols “1” and “0”. The performance measure in
digital communications is provided by “probability of error”, the bit error-rate (BER).
The most basic form of pulsed modulation in binary direct detection receiver is on-off
keying (OOK). The object is to determine the presence of signal in a noisy environment.
If a “0” is mistaken by “1” , the probability is denoted by Pr(1 0), while a “1” may be
mistaken by a “0” with probability Pr(01). The overall probability of error Pr(E) is:
Pr(E) = p0 Pr(1 0) + p1 Pr(01) ,
p0 is the transmission probability of a binary “0”, p1 is the

transmission probability of a binary “1”.
For OOK transmission, assuming Gaussian distribution for noise alone and
signal plus noise,
1
∞
1  i 
∫e
/ 2σ N 2
erfc T 
2
Pr(1 0) = −i
di =
 2σ 
2π σ N iT
2  N 
1 i −i 
Pr(01)= erfc S T 
2  2σ 
 N 
NO TURBULENCE: BERNO TURB. = Pr(E) =
1  i  1  SNRNO TURB. 
erfc S  = erfc
 2  
2  2σ N   2 2 
1
∞  SNR S 
 ds
2 ∫0
WITH TURBULENCE: BERTURB. = Pr(E) = p ( s ) erfc
I
2 2 i 
 S 

Effect of Atmospheric Turbulence on Bit Error Rate
• Atmospheric turbulence significantly impacts BER

• Even with aperture averaging, reduction in BER is several orders of
magnitude
• As atmospheric turbulence strength and path lengths increase, so
does the BER
L = 2000 m
L = 1000 m 0
10
100 D = 4 cm, Cn2 = 10e-14
D = 4 cm, Cn2 = 10e-14 D = 8 cm, Cn2 = 10e-14
D = 8 cm, Cn2 = 10e-14 10-1 D = 4 cm, Cn2 = 5x10e-14
10-1 D = 4 cm, Cn2 = 5x10e-14 D = 8 cm, Cn2 = 5x10e-14
D = 8 cm, Cn2 = 5x10e-14
10-2
10-2
Bit Error Rate
10-3
Bit Error Rate
10-3
no turbulence
no turbulence
10-4
10-4
10-5
10-5
10-6
10-6
10-7
10-7
10-8
10-8
-50.0 -45.5 -41.0 -36.5 -32.0 -27.5 -23.0 -18.5 -14.0
-50.0 -45.5 -41.0 -36.5 -32.0 -27.5 -23.0 -18.5 -14.0
Receiver Power (dBm)
Receiver Power (dBm)

What to do? Adaptive Optics
Partial Coherence: Poor Man’s Adaptive Optics
Collimated Beam
coherent beam
10-1 partially coherent beam
Cn2 = 1.2x10-14 m-2/3

-2
10 λ = 0.785µ m
z = 2000 m
10-3 wo = 2.5 cm
Bit Error Rate
D = 10 cm
-4
10
free
10-5 space
10-6
10-7 Cn2 = 1x10-15 m-2/3
10-8
-65 -60 -55 -50 -45 -40 -35 -30 -25
Popt (dBm)
Weak turbulence:
PCB reduces BER by 3 orders of magnitude
Moderate turbulence:
PCB reduces BER by only 1 order of
magnitude
The figure shows a plot of BER as a function of SNR for
different signal fluctuations, defined by β0 2 =0.50C n 2 k 7 / 6 L11 / 6
(for weak fluctuation, β0 2 =0.1, and for moderate to strong
fluctuations, β0 2 =4).
•*Laser Beam Scintillation with Applications, L.C. Andrews, R.L. Phillips, and C.Y.
Hopen (SPIE Press, Bellingham, 2001).

PROBABILITY OF FADE
The probability that the output current of the detector will drop
below a prescribed threshold iT is defined by
∞ iT iT
Pr (i ≤ iT ) = ∫∫p
0 0
S+ N (i s ) p I ( s ) dids ≈ ∫p
0
I (i )di
 I (0, L) 
Fade threshold parameter: FT = 10 log 10   dB

 IT 
Case 1. Terrestrial Laser Communication Link
The figure shows probability of fade as a function of threshold

level, D=0 defines a point receiver.
The Following figures show the probability of fade for various
path lengths ,
Cn2 = 10-13 m-2/3 , wavelength, λ = 1.55 µ m. • Laser Beam Scintillation
with Applications, L.C.
Andrews, R.L. Phillips,
and C.Y. Hopen (SPIE
Press, Bellingham,
2001).

Case 2. Uplink Slant Path Laser Communication
Link
2
Note that the atmospheric model for C n is to be
taken from Hufnagel -Valley (H -V) model,
described earlier. This model shows the variation of
Cn2 as a function of height taking into account of the
zen ith angle. The probability of fade for an uplink
spherical wave to a geo -stationary satellite under
various atmospheric conditions is shown in the
following figure.
Case 3. Downlink Slant Path Laser

Communication Link
The plane wave model can be u sed to calculate the

irradiance variance and then probability of fade.
The figure shows the probability of fade for a
downlink path from a satellite in geo -stationary
orbit.
Probability of Fade for Uplink and Downlink
•* Laser Beam Scintillation with Applications, L.C. Andrews, R.L. Phillips, and C.Y. Hopen (SPIE
Press, Bellingham, 2001).

Mitigating Turbulence Effects
Multiple Transmitters Approach
(Courtesy Jaime Anguita: Ref. Jai Anguita, Mark A. Neifeld and Bane Vasic, “Multi-Beam Space-Time
Coded Communication Systems for Optical Atmospheric Channels,” Proc. SPIE, Free-Space Laser
Communications VI, Vol. 6304, Paper # 50, 2006)
Aperture averaging and multiple beams is effective in reducing

scintillation, improving performance
Adaptive Optics approach can be incorporated to mitigate

turbulence effects for achieving free space laser communications
REFERENCES
• 1. Free-Space Laser Communications: Principles and Advances, A. K.
Majumdar and J. C. Ricklin, Eds. (Springer, 2008)
• 1a. A.K. Majumdar and J.C. Ricklin, “Effects of the atmospheric channel
on free-space laser communications”, Proc. of SPIE Vol. 5892, 2005.
• 2. J. C. Ricklin and F. M. Davidson, “Atmospheric optical communication
with a Gaussian Schell beam,” J. Opt. Soc. Am. A 20(5), 856-866 (2003).
• 3. J. C. Ricklin and F. M. Davidson, “Atmospheric turbulence effects on
a partially coherent Gaussian beam: implications for free-space laser
communication,” J. Opt. Soc. Am. A 19(9), 1794-1803 (2002).
• 4. W. B. Miller, J. C. Ricklin and L. C. Andrews, “Log-amplitude
variance and wave structure function: a new perspective for Gaussian
beams,” J. Opt. Soc. Am. A 10(4), 661-672 (1993).
• 5. L. C. Andrews, W. B. Miller and J. C. Ricklin, “Geometrical
representation of Gaussian beams propagating through complex optical
systems,” Appl. Opt. 32(30), 5918-5929 (1993).
• 6. Laser Beam Propagation Through Random Media, L. C. Andrews and
R. L. Phillips (SPIE Press, Bellingham, 1998).
• 7. Laser Beam Scintillation with Applications, L.C. Andrews, R.L.
Phillips, and C.Y. Hopen (SPIE Press, Bellingham, 2001).
• 8. Optical Communications, R.M. Gagliardi and S. Karp (R.E. Krieger
Publishing Company, 1988).
• 9. Optical Channels, S. Karp, R.M. Gagliardi, S.E. Moran and L. B.
Stotts ( Plenum Press, New York, 1988).

REFERENCES
• 10. I.I. Kim, H.Hakakha, P. Adhikari, E. Korevaar and A.K.
Majumdar, “Scintillation reduction using multiple transmitters” in
Free-Space Laser Communication Technologies IX, Proc. SPIE,
2990,102-113 (1997).
• 11. A.K. Majumdar, “Optical communication between aircraft in
low-visibility atmosphere using diode lasers,” Appl. Opt. 24,
3659-3665 (1985).
• 12. A.K. Majumdar and W.C. Brown, “Atmospheric turbulence
effects on the performance of multi-gigabit downlink PPM laser
communications,” SPIE Vol.1218 Free-Space Laser
Communication Technologies II , 568-584 (1990).

Free-Space Laser Communications: Fundamentals, System Design, Analysis and Applications

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Free-Space Laser Communications: Fundamentals, System Design, Analysis and Applications

Încărcat de

Drepturi de autor:

Formate disponibile

Free-Space Laser Communications:

Fundamentals, System Design, Analysis

Dr. Arun K. Majumdar

Copyright © 2009 Arun K. Majumdar 1

2. Major sub-systems for laser communications systems and Link

Copyright © 2009 Arun K. Majumdar 2

3. Optical Signal Detection

4. Atmospheric Channel Effects

Copyright © 2009 Arun K. Majumdar 3

Copyright © 2009 Arun K. Majumdar 4

Copyright © 2009 Arun K. Majumdar 5

Copyright © 2009 Arun K. Majumdar 6

9. Summary: Improvement of Lasercom Performance

Copyright © 2009 Arun K. Majumdar 7

Copyright © 2009 Arun K. Majumdar 8

Copyright © 2009 Arun K. Majumdar 9

Copyright © 2009 Arun K. Majumdar 11

Copyright © 2009 Arun K. Majumdar 12

• Figure. Selected Modulation Formats

Figure. Typical direct detection digital optical receiver

Copyright © 2009 Arun K. Majumdar 14

Copyright © 2009 Arun K. Majumdar 15

decision circuit for a “1” and “0”, respectively.

Thus, BER = (1/2) erfc(Q/21/2 ), where Q = (I1- I ) / (σ1+ σ0)

Copyright © 2009 Arun K. Majumdar 16

Potential atmospheric effects:

Copyright © 2009 Arun K. Majumdar 17

Copyright © 2009 Arun K. Majumdar 18

• Absorption depends on water vapor and

• Transmission “windows” occur at visible

Copyright © 2009 Arun K. Majumdar 19

• caused when wavelength collides with scattering particle

No smoke Weak smoke Heavy smoke

Fluctuations of the refractive index are locally homogeneous and

Copyright © 2009 Arun K. Majumdar 22

December 15, 2002 December 16, 2002 December 17, 2002

February 8, 2003 February 12, 2003 February 13, 2003

Copyright © 2009 Arun K. Majumdar 23

Copyright © 2009 Arun K. Majumdar 24

where A= -17.0577, B= -0.0449, C= -0.0005

Copyright © 2009 Arun K. Majumdar 26

Receiver beam size:

Receiver radius of curvature:

Copyright © 2009 Arun K. Majumdar 27

Average Peak Power Density

-2 -1 0 1 2 0 1000 2000 3000 4000 5000

Copyright © 2009 Arun K. Majumdar 28

For weak scintillation regime, the irradiance

The three-dimensional power spectrum of

φn (κ) = 0.033Cn 2κ−11/ 3 ,

where κ =2π/turbulence size

Copyright © 2009 Arun K. Majumdar 29

κml0 = 5.92 and κ0 = 1/L0

Copyright © 2009 Arun K. Majumdar 30

Aperture-Averaging Factor A: describes the percent decrease in intensity

Aperture-Averaged Log-Irradiance = (1.0)(.75) =0.75

Aperture Averaging Model*:

*Ricklin and Davidson, JOSA A 20(5), 856, 2003.

• Aperture averaging can significantly reduce intensity scintillations

Aperture Averaging: Fluctuations in intensity

“Artificial” Aperture Averaging: reduce the beam

Copyright © 2009 Arun K. Majumdar 33

D ivergent Beam Divergent Beam