Journal of Theoretical and Applied Information Technology

20th August 2013. Vol. 54 No.2

© 2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

TRANSCEIVER DESIGN AND PERFORMANCE ANALYSIS

OF FREE-SPACE OPTICAL COMMUNICATION
S.SUGUMARAN1, A.JABEENA2,T.TULASI RAM KUMAR3,K.DILEEP KUMAR4

School of electronics engineering, Vellore institute of technology,

Vellore 632014, Tamil Nadu, India
E-mail: 1ssugumaran@vit.ac.in , 2ajabeena@vit.ac.in , 3tulasiram.talluri44@gmail.com
, 4dil.vit@gmail.com

ABSTRACT

In free-space optical communication links, atmospheric turbulence causes fluctuations in both the intensity
and the phase of received light signal, impairing link performance. In this paper, we describe several
communication techniques to mitigate turbulence-induced intensity fluctuations, that is signal fading. These
techniques are applicable in the regime in which the receiver aperture is smaller than the correlation length
of the fading and the observation interval is shorter than the correlation time of fading. We assume that the
receiver has no knowledge of the instantaneous fading state. When the receiver knows only the marginal
statistics of the fading, a symbol-by-symbol Ml detector can be used to improve the detection performance.
If the receiver has knowledge of the joint temporal statistics of fading, maximum- likelihood sequence
detection (MLSD) can be employed, yielding a further performance improvement, but at the cost very high
complexity
Keywords-Atmospheric Turbulence, Free-Space Optical Communication, MLSD, Spatial Diversity
Reception.

1. INTRODUCTION the receiver aperture can be made larger than the

related correlation length then turbulence-
Free space optical communication has attracted induced fading can be reduced substantially by
considerable attention recently for a variety of aperture averaging [13]. FSO is an optical
applications [1]-[4]. Because of the complexity communication technology that uses light
associated with phase or frequency modulation, propagation in free space to transmit data for
current free-space optical communication telecommunications or computer networking.
systems typically use intensity modulation with Free space means air, outer space, vacuum, or
direct detection (IM/DD). Atmospheric something similar. This contrasts with using
turbulence can degrade performance of FSO solids such as optical fiber cable or an optical
links, particularly over ranges of the order of transmission line. The technology is useful
1km or longer than that. In homogeneities in the where the physical connections are impractical
temperature and pressure of the atmosphere lead because of high costs or other considerations.
to variations of the refractive index along the
transmission path. These indexes in 2. MODELING OF ATHMOSPHERE
homogeneities can deteriorate the quality of TURBULENCE:
received image and can cause fluctuations in
both the intensity and phase of received signal. Atmospheric turbulence can be physically
These fluctuations can lead to an increase in the described by Kolmogorov theory [6]-[9].the
link error probability, limiting the performance energy is large eddies is redistributed without
of communication systems. Atmospheric loss to eddies of decreasing size until finally
turbulence has been studied extensively and dissipated by viscosity. The size of turbulence
various theoretical models have been proposed to eddies normally ranges from a few more
describe turbulence- induced image degradation millimetres to a few meters, denoted as the inner
and intensity fluctuations which are signal fading scale lo and the outer scale, respectively. We can
[7]-[13]. Two useful parameters describing express the refractive index as
turbulence-induced fading are the correlation
length of intensity fluctuations and, the 𝑛(𝑟̅, 𝑡) = 𝑛0 + 𝑛1 (𝑟̅ , 𝑡) (1)
correlation time of intensity fluctuations. When

356
 Journal of Theoretical and Applied Information Technology
20th August 2013. Vol. 54 No.2
© 2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

Where no is the average index and n1 is the Since the random distribution is Gaussian
fluctuation component induced by spatial distributed under the assumption of weak
variations of temperature and pressure in the turbulence, we can use the rytov method to
atmosphere. The correlation function of n1 is
derive the normalized log amplitude covariance
defined as
⌈𝑛1 (𝑟�1 , 𝑡𝑡 ; 𝑟�2 , 𝑡2 ) = 𝐸[𝑛1 (𝑟�1 , 𝑡1 ). (𝑟�2 , 𝑡2 )] function for two positions in receiving plane
(2)
perpendicular to the direction of propagation.

B. Spatial and Temporal Coherence of the Since the random distribution is Gaussian-
Optical Signals through Turbulence distributed under the assumption of weak
turbulence, we can use the rytov method to
To describe spatial coherence of optical waves, derive the normalized log-amplitude covariance
so called mutual coherence function is widely function for two positions in a receiving plane
used [7] perpendicular to the direction of propagation.

�(𝑃1 , 𝑡1 ; 𝑃2 , 𝑡2 )] = E[𝑢(𝑃1 , 𝑡1 ). 𝑢∗ (𝑃2 , 𝑡2 )]� (3) 𝑏𝑋 (𝑑12 ) =

𝐵𝑋 (𝑃1 ,𝑃2 )
(8)
𝐵𝑋 (𝑃1 ,𝑃1 )

Where u (P,t) is the complex optical field, setting Where 𝑑12 is the distance between 𝑃1 and 𝑃2 . We
in 𝑡1 = 𝑡2 in (4). We obtain the spatial define the correlation length of intensity
MCF ⌈(𝑟�1 , 𝑟�2 ) . The rytov method is frequently fluctuations, 𝑑0 , such that 𝑏𝑋 (𝑑0 ) = 𝑒 −2 . When
used to expand the field u(𝑟̅ ): the propagation path length L satisfies the
condition 𝑙0 < √λL < 𝐿0 , where λ is the
𝑢(𝑟̅) = 𝐴0 (𝑟̅). 𝑒𝑥𝑝[𝑖𝜑0 (𝑟̅ )] = 𝑢0 (𝑟̅). 𝑒𝑥𝑝(𝜑1 ) wavelength and 𝑙0 and 𝐿0 are inner and outer
(4) length scales, respectively, 𝑑0 can be
approximated by [13]
Where 𝑢0 (𝑟̅ ) is the field amplitude without air
turbulence:
𝑑0 ≈ √λL
𝑢0 (𝑟̅ ) = 𝐴0 (𝑟̅). 𝑒𝑥𝑝[𝑖𝜑0 (𝑟̅ )]
(5)
C. Probability Distributions of Turbulence-
Exponent of the perturbation factor is: Induced Intensity Fading:
𝐴(𝑟̅ )
𝜑1 = log � � + 𝑖[𝜑(𝑟̅ ) − 𝜑0 (𝑟̅)] = 𝑋 + 𝑖𝑆 As discussed previously, when the propagation
𝐴0 (𝑟̅ )
(6)
distance is long, log-amplitude fluctuations can
become significant. In this section, we will
Where X is the log amplitude function and S is derive the statistical properties of the log-
amplitude fluctuations, which we refer to as
phase function. We assume X and S be
“intensity fading” or simply “fading.” The
homogenous, marginal distribution of fading is derived while
the joint spatial and temporal distribution of
Isotropic and independent Gaussian random fading are derived.
variables. This assumption is valid for long
propagation distances through turbulence. 1) Marginal Distribution of Fading: In this
section, we derive the marginal distribution of
In order to characterise turbulence fading at a single point in space at a single
induced fluctuations of the log amplitude X, we instant in time. The marginal distribution is used
use the log amplitude covariance function: in symbol-by-symbol ML detection. For
propagation distances less than a few kilometres,
𝐵𝑥 (𝑃1 ; 𝑃2 ) = 𝐸[𝑋(𝑃1 )𝑋(𝑃2 )] − variations of the log-amplitude are typically
much smaller than variations of the phase. Over
𝐸[𝑋(𝑃1 )], [𝑋(𝑃2 )]. (7)
longer propagation distances, where turbulence
becomes more severe, the variation of the log-

357
 Journal of Theoretical and Applied Information Technology
20th August 2013. Vol. 54 No.2
© 2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

amplitude can become comparable to that of the

phase. Based on the atmosphere turbulence 𝜎𝑋2 |𝑠𝑝ℎ𝑒𝑟𝑖𝑐𝑎𝑙 =
model adopted here and assuming weak 5
2𝜋𝑓 7/6 𝐿 𝑥 6
turbulence, we can obtain the approximate 0.56(
𝑣
) ∫0 𝐶𝑛2 (𝑥) �𝐿 � (𝐿 − 𝑥)5/6 𝑑𝑥
analytic expression for the covariance of the log- (9)
amplitude fluctuation of plane and spherical
waves [9]:
2𝜋𝑓 𝐿
𝜎𝑋2 |𝑝𝑙𝑎𝑛𝑒 = 0.56( )7/6 ∫0 𝐶𝑛2 (𝑥)(𝐿 −
𝑣
𝑥)5/6 𝑑𝑥

4. SPATIAL DIVERSITY RECEPTION:

2) Joint Spatial and Temporal Distributions of
Fading: Spatial diversity reception, which has been well-
In this section, we derive the joint spatial and studied for application at radio and microwave
temporal distributions of fading. The joint spatial frequencies, which has the potential to mitigate
distribution describes the fading at multiple the degradation caused by atmospheric
points in space at a single instant of time and is turbulence [10]-[12] and [14]-[15]. Spatial
used in evaluating the performance of spatial diversity reception in FSO communication has
diversity reception. The joint temporal been proposed and studied in [14] and [15]. One
distribution describes the fading at a single point of the scientists named Ibrahim [14] has been
in space at multiple instants of time. This studied the performance of spatial-diversity
distribution is the basis for the MLSD We optical reception on turbulence channels,
assume that the log-amplitude at receivers is assuming that turbulence-induced fading is
described by a joint Gaussian distribution. The uncorrelated at each of the optical receivers. In
auto-covariance matrix of the log-amplitude at order for this assumption to hold true, then the
receivers in a plane transverse to the direction of spacing between receivers should exceed the
propagation is given by fading correlation length in the plane of
𝜎𝑋2 ⋯ 𝜎𝑋2 𝑏𝑋 (𝑑1𝑛 ) receivers, it may be difficult to satisfy this
𝐶𝑋 = � ⋮ ⋱ ⋮ � assumptions in practical, for various reasons.
𝜎𝑋2 𝑏𝑋 (𝑑𝑛1 ) ⋯ 𝜎𝑋2 Available space may not be permit sufficient
𝑛×𝑛
receiver spacing. In power-limited links, which
often employ well-collimated beams, then the
receiver spacing required for uncorrelated fading
3. ML DETECTION OF ON-OFF may exceed the beam diameter.
KEYING IN TURBULENCE
CHANNELS: 5. PROPOSED TRANSMITS DIVERSITY
SCHEME:
In this paper, we consider intensity
modulation/direct detection (IM/DD) links using The use of optical arrays, similar to the use of
on-off keying (OOK). In most practical systems, antenna-array technology for microwave system,
the receiver signal-to-noise ratio (SNR) is limited is considered as the means of combatting fading.
by shot noise caused by ambient light much Specifically, we adopt a multiple- input-signal-
stronger than the desired signal and/or by output (MISO) array based on laser sources,
thermal noise in the electronics following the assumed to be intensity-modulated only and all
photo detector. In this case, the noise can usually pointed towards distant photo detector assumed
be modelled to high accuracy as additive, white to be ideal receiver of noncoherent (direct-
Gaussian noise that is statistically independent of detection). The sources and the detector are
the desired signal. Let denote the bit interval of physically situated so that all transmitters are
the OOK system and assume that the receiver simultaneously observed by the receiver. The
integrates the received photocurrent for an fading experienced between source-detector pairs
interval during each bit interval. At the end of the is assumed to be statistically independent. As we
integration interval, the resulting electrical signal presented in [16], assuming transmission
can be expressed as: matrices based on real orthogonal; designs for
𝑟𝑒 = 𝑛(𝐼𝑠 + 𝐼𝑏 ) + 𝑛 full rate and full diversity space-time block
(10) codes, our MISO system model implies the

358
 Journal of Theoretical and Applied Information Technology
20th August 2013. Vol. 54 No.2
© 2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

diversity of order, so given that the conditional

bit-error rate (BER) is given as
𝑑 2
� � 2
𝑃𝑏 (𝐸|{𝐼𝑗 }1<𝑗<𝐿 ) = 𝑄 �� 𝐿 ��∑𝐿𝑗=1 𝐼𝑗 � �
2𝑁0
𝑑 2
� �
𝑃𝑏 (𝐸|{𝐼𝑗 }1<𝑗<𝐿 ) = 𝑄 �� ∑𝐿 𝐼 2 �
𝐿
2𝑁0 𝑗=1 𝑗 (12)

(11) 6. RESULTS:

Where Q(.) is the Gaussian- Q function. Here, The ML detection schemes over turbulence
the division by is considered so as to maintain channels are studied based on the statistical
the average optical power in the air at a constant distributions of turbulence-induced fading. If the
instantaneous fading state is unknown but the
level of being transmitted by each laser then an
marginal fading statistics are known, we can
average optical power of a source alternative to apply
this transmit diversity scheme is repetition
coding where the same signal is simultaneously
transmitted from different laser sources so then ML symbol-by-symbol detection to improve
the conditional BER is given by detection efficiency. If the temporal correlation
of fading is known, i.e., the joint temporal
distribution of turbulence-induced fading, we can
apply MLSD, leading to a further improvement
in detection performance.

Fig (b)

Fig (A) Figure (a) and (b) shows that Bit-error

probability of dual-branch receiver versus
average electrical signal-to-noise ratio using
maximum-likelihood detection (lines with
circles) and equal-gain combining (lines without
symbols) for different values of 𝜌𝑑 , the
normalized correlation between the two
receivers. The line with squares represents the
bit-error probability using a single receiver. The

359
 Journal of Theoretical and Applied Information Technology
20th August 2013. Vol. 54 No.2
© 2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

turbulence induced 𝜎𝑋 =0.1 in fig (a) and 𝜎𝑋 =0.25 [5] Bernard Sklar (July 1997). "Rayleigh Fading
in (b) this fading has mean E[X] = 0. Channels in Mobile Digital Communication
Systems Part I: Characterization". IEEE
7. CONCLUSION: Communications Magazine 35 (7): 90–100.
doi:10.1109/35.601747+ISSN+0163-6804.
The ML detection schemes over turbulence
[6] M. Born and E. Wolf, Principles of Optics,
channels are studied based on statistical
6th ed. city??, Canada:
distributions of turbulence-induced fading. If the Pergamon, 1980.
instantaneous fading state is unknown but we [7] J. W. Goodman, Statistical Optics. New
know the marginal fading statistics, we can apply York: Wiley, 1985.
ML symbol-by-symbol detection to improve [8] A. Ishimaru, Wave Propagation and
detection efficiency. If the temporal correlation Scattering in Random
of fading is known that is, the joint temporal Media. New York: Academic, 1978, vol. 1–2.
[9] S. Karp, R. Gagliardi, S. E. Moran, and L. B.
distribution of turbulence-induced fading, we can
Stotts, Optical Channels.
also apply MLSD, leading to a further New York: Plenum, 1988.
improvement in performance of detection. [10] G. L. Stuber, Principles of Mobile
Spatial diversity reception can also help to Communication. New York: Kluwer
mitigate turbulence-induced fading. When the Academic, 1996.
spacing between receivers is not much greater [11] J. G. Proakis, Digital Communication, 3rd
than the fading correlation length, diversity gain ed. New York: McGraw- Hill, 1995.
of this is reduced by correlation, but ML [12] M. Srinivasan and V. Vilnrotter,
“Avalanche photodiode arrays for optical
detection can be used to overcome some of this
Communication receivers,” NASA TMO
loss. We have shown the same in then dual- Progress Rep. 42-144,
receiver case, ML diversity reception performs 2001.
the conventional EGC method. [13] L. C. Andrews and R. L. Phillips, Laser
Beam Propagation through Random Media:
SPIE Optical Engineering Press, 1998.
REFERENCES: [14] M. M. Ibrahim and A. M. Ibrahim,
“Performance analysis of optical receivers
[1] Clint Turner (October 3, 2007). "A 173-mile with space diversity reception,” Proc. IEE—
2-way all-electronic optical contact". Commun., vol. 143,
Modulated light web site. Retrieved June 28, no. 6, pp. 369–372, December 1996.
[15] X. Zhu and J. M. Kahn, “Maximum-
2011.
likelihood spatial-diversity reception on
[2] Wilson, K. "Recent Development in High- correlated turbulent free-space optical
Data Rate Optical Communications at JPL". channels,” presented at the IEEE Conf. on
Jet Propulsion Laboratory. NASA Technical Global Commun. San Francisco, CA, Nov.–
Reports Server. Retrieved 4 October 2011. Dec. 27–1, 2000.
[3] W. Elmenreich, N. Marchenko, H. Adam, C. [16] I. S. Gradshteyn and I. M. Ryzhik, Table of
Hofbauer, G. Brandner, C. Bettstetter, and Integrals, Series and Products,7th ed. New
York: Academic, 2007
M. Huemer (2008). "Building blocks of
cooperative relaying in wireless systems". e
& i, Springer 125 (10): 353–
359. doi:10.1007/s00502-008-0571-7.
[4] John G. Proakis (1995). Digital
Communications (3rd ed.). Singapore:
McGraw-Hill Book Co. pp. 767–768.
ISBN 0-07-113814-5.

360