JOURNAL OF LIGHTWAVE TECHNOLOGY

Free-Space Optical Communication With Spatial

Modulation and Coherent Detection Over H-K
Atmospheric Turbulence Channels
Kostas P. Peppas, Senior Member, IEEE, and P. Takis Mathiopoulos, Senior Member, IEEE

Index TermsAtmospheric turbulence, average bit error probability, coherent detection, free space optical communication systems, H-K distribution, multiple-input multiple-output (MIMO)
systems, optical spatial modulation.

I. INTRODUCTION

REE-SPACE optical (FSO) communication systems have

recently attracted great attention within the research community as well as for commercial use. FSO systems can provide ultra-high data rates (at the order of multiple gigabit per
second), immunity to electromagnetic interference, excellent
security and large unlicensed bandwidth, i.e., hundred or even
thousand times higher than radio-frequency (RF) systems, along
with low installation and operational cost [1].
The challenge in employing such systems is that FSO links are
highly vulnerable due to the detrimental effects of attenuation
under adverse weather conditions (e.g., fog), pointing errors
and atmospheric turbulence [1]. One method to improve the
reliability of the FSO link is to employ spatial diversity, i.e.
multiple-lasers and multiple-apertures to create a multiple-input
multiple-output (MIMO) optical channel. Because of its low
complexity, spatial diversity is a particularly attractive fading

mitigation technique and performance enhancements have been

extensively studied in many past research works in the field of
FSO communications (e.g., see [2][5]).
In order to evaluate the impact of atmospheric turbulence on
the performance of OSM, accurate models for the fading distribution are necessary. For example, the lognormal distribution
is often used to model weak turbulence conditions whereas the
negative exponential and the K-distribution are used to model
strong turbulence conditions [6]. Other more general statistical models have also been proposed to model scintillation over
all turbulence conditions, such as, for example, the Gamma
Gamma [7], the lognormal-Rice (or Beckmann) [8] the homodyned K distribution (H-K) [9] and the I-K [10][12] distributions. The validity of these models is based on the argument
that scintillation is a doubly stochastic random process modelling both small and large scale turbulence effects. Besides,
they agree well with measurement data and simulations for a
wide range of turbulence conditions.
In this paper, the H-K distribution is adopted to model
turbulence-induced fading [9]. The main reason for this choice
is the fact that this distribution is based on a very general scattering model which is valid over a wide range of atmospheric
conditions. It is also noted that the H-K distribution generalizes
existing models such as the K-distribution. The H-K distribution
models the field of the optical wave as the sum of a deterministic component and a random component, the intensity of which
follows the Rice (Nakagami-n) distribution. The average intensity of the random portion of the field is treated as a fluctuating
quantity [10]. It is important to underline that, to the best of our
knowledge, in the open technical literature there have been no
papers published analyzing and evaluating the performance of
FSO systems over such channels, because of the complicated
mathematical form of the required for the analysis probability
density functions (PDF).
Depending on their detection, FSO systems can be classified
into two main categories, namely non-coherent (direct detection) and coherent (heterodyne detection) systems [13][17].
Non-coherent FSO systems employing intensity modulation
and direct detection (IM/DD) have the optical power output
of a source varied in accordance with some characteristic of
the modulating signal. A direct detection receiver only responds
to the intensity of the received field and is oblivious to the
received signal phase. Coherent FSO systems encode the information bits directly onto the electric field of the optical beam.
At the receiver, a local oscillator (LO) is employed to extract
the information encoded on the optical carrier electric field. On

IE
E
Pr E
oo
f

AbstractThe use of optical spatial modulation (OSM) is proposed as a simple low-complexity means of achieving spatial diversity in coherent free-space optical (FSO) communication systems.
In doing so, this paper presents a generic analytical framework for
obtaining the average bit error probability (ABEP) of uncoded and
coded OSM with coherent detection in the presence of turbulenceinduced fading. Although the framework is general enough to accommodate any type of channel models based on turbulence scattering, the focus of the analysis is on the H-K distribution as such
model is very general and valid over a wide range of atmospheric
propagation conditions. Using this framework, it is shown that
OSM can offer comparable performance with conventional coherent FSO schemes employing spatial diversity at the transmitter
or the receiver only, while outperforming the latter in terms of
spectral efficiency and hardware complexity. Furthermore, various numerical performance evaluation results are also presented
and compared with equivalent results obtained by Monte Carlo
simulations which verify the accuracy of the derived analytical
expressions.

Manuscript received March 23, 2015; revised June 21, 2015; accepted July
14, 2015.
K. P. Peppas is with the Department of Informatics and Telecommunications,
University of Peloponnese, Tripoli 22100, Greece (e-mail: peppas@uop.gr).
P. T. Mathiopoulos is with the Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Athens 15784,
Greece (e-mail: mathio@di.uoa.gr).
Digital Object Identifier 10.1109/JLT.2015.2465385

0733-8724 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications standards/publications/rights/index.html for more information.

JOURNAL OF LIGHTWAVE TECHNOLOGY

TABLE I
LIST OF MATHEMATICAL NOTATIONS
j 2 = 1 denotes the imaginary unit
|z | denotes the magnitude of the complex number z
{z } denotes the real part of the complex number z
{z } denotes the real part of the complex number z
f (x )
f (x) = o[g (x)] as x x 0 if lim x x 0 g ( x ) = 0
 2F denotes the square Frobenius norm
() T denotes the matrix transpose
denotes convolution
E denotes expectation
f X () denotes the PDF of the RV X
F X () denotes the cumulative distribution function of the RV X
MX () denotes the MGF of the RV X
I a () is the modified Bessel function of the first kind and order a [26, eq. (8.431)]
K a () is the modified Bessel function of the second kind and order a [26, eq. (8.432)]

x 1
dt is the Gamma function [26, eq. (8.310/1)]
(x) = 0 exp(t)t


Q (x) = 21 x exp(t 2 /2)dt is the Gauss Q-function

W p , q () is the Whittaker function [26, eq. (9.220)]
Pr{} denotes the probability operator
denotes estimated value at the receiver side

specifically and within this new analytical framework, the main

novel research contributions of the paper are as follows:
1) New analytical expressions for the ABEP of coherent OSM
under turbulence conditions modeled by the H-K distribution are derived. When the transmitter is equipped with
two apertures the resulting analytical expressions are exact, whereas for an arbitrary number of transmit apertures
tight upperbounds can be obtained.
2) ABEP bounds for coded OSM systems are derived and
the performance enhancements when channel coding is
employed are presented and analyzed.
The ABEP of OSM is also compared to that of conventional
FSO schemes with transmit or receive diversity only, i.e., when
maximal ratio combining (MRC), selection combining (SC) or
Alamouti-type space-time block codes (STBC) are employed.
It is finally noted that the theoretical analysis is substantiated by
comparing the theoretical and equivalent simulated performance
evaluation results obtained by means of Monte Carlo techniques.
The paper is organized as follows. After this introduction,
Section II outlines the system and channel models. In Section III
analytical expressions for the ABEP of uncoded OSM systems
are presented. Asymptotic ABEP expressions are also derived,
wherefrom the diversity gain of coherent OSM can be readily
deduced. The performance of coded OSM systems is discussed
in Section IV. In Section V an overview of the conventional diversity techniques for coherent OSM is presented. In Section VI
the various performance evaluation results and their interpretations as well as comparisons are presented. Finally, concluding
remarks can be found in Section VII. Notations: A comprehensive list of all mathematical notations used in this paper can be
found in Table I.

IE
E
Pr E
oo
f

the one hand, coherent FSO systems can provide significant

performance enhancements due to spatial temporal selectivity
and heterodyne gain in comparison to direct detection systems.
Moreover, they are more versatile as any kind of amplitude,
frequency, or phase modulation can be employed. On the other
hand, the implementation of coherent receivers is more complex
as the LO field should be spatially and temporally coherent with
the received field. Throughout this work, coherent FSO systems
are only considered.
Recently, the so-called optical spatial modulation (OSM)
has emerged as a power- and bandwidth-efficient single-carrier
transmission technique for optical wireless communication systems [18][20]. This spatial diversity scheme, initially proposed
in [21] and further investigated in [22] and[23], employs a simple modulation mechanism that foresees to activate just one out
of several MIMO transmitters at any time instant and to use the
index of the activated transmitter as an additional dimension for
conveying implicit information. It has been shown that OSM
can increase the data rate by base two logarithm of the number
of transmit units [18]. Also, OSM can increase the data rate by
a factor of 2 and 4, as compared to onoff keying and pulse position modulation, respectively [18], [19]. It is underlined that
such performance gains are obtained with a significant reduction
in receiver complexity.
Because of the above mentioned advantages of OSM over
other more conventional transmission schemes and given the
wide applicability of FSO, it is important to investigate the
potential performance enhancements obtained by incorporating
OSM in FSO systems. However, in general, this research topic
has not been thoroughly investigated. In fact, only recently,
there have been papers published in the open technical literature dealing with performance analysis studies of FSO systems
employing spatial modulation and operating in the presence of
atmospheric turbulence, e.g., see [24] and [25]. Specifically, in
[24], the combination of subcarrier IM and spatial modulation
with receiver diversity was proposed to enhance the performance of IM/DD FSO systems. In [25], another IM/DD based
system FSO system which combines antenna shift keying with
joint pulse position and amplitude modulations was considered.
For this system, which was denoted as spatial pulse position
and amplitude modulation (SPPAM), the atmospheric turbulence channel was modeled as log-normal or GammaGamma
distributions and was evaluated, in terms of bounds, for uncoded and coded signals. Average bit error probability (ABEP)
performance evaluation results have shown that SPPAM offers a
compromise between spectral and power efficiencies as well as
a certain degree of robustness against atmospheric turbulence.
Despite these two papers which deal with non-coherent detection schemes, the potential enhancements of OSM on improving the performance of FSO systems with coherent detection
still remains an open research topic which, to the best of our
knowledge, has not been addressed so far in the open technical
literature.
Motivated by the above, in this paper we present for the first
time a generic analytical framework which can be used to accurately obtain the performance of outdoor OSM with coherent
detection in the presence of turbulence-induced fading. More

II. SYSTEM AND CHANNEL MODEL

In this section, a detailed description of the OSM FSO system model, i.e., transmitter, channel and receiver is provided.
Moreover, the H-K distribution is introduced and analytical

PEPPAS AND MATHIOPOULOS: FREE-SPACE OPTICAL COMMUNICATION WITH SPATIAL MODULATION AND COHERENT DETECTION

expressions for its parameters in terms of equivalent physical parameters of the turbulence phenomenon, such as the refractiveindex structure parameter, optical wave number, and propagation path length, are derived.
A. Preliminaries

B. Receiver Structure

Expanding (2) and ignoring the double-frequency terms that are

filtered out by the bandpass filter, the resulting photocurrent can
be expressed as
2 2
hm ,n + RPLO
in (t) = RPt Em

+2R Pt PLO Em hm ,n cos(IF t m ,n b m )

= iDC (t) + iAC (t).

(3)

2 2
In (3), iDC (t) = RPt Em
hm ,n + RPLO is the dc component
generated
by
the
signal
and
LO fields, respectively, iAC (t) =

2R Pt PLO cos(IF t m ,n b m ) is the ac component in

the received photocurrent which, unlike for direct detection,
contains information about the frequency and phase of the received signal. It is assumed that for coherent detection the IF
IF is nonzero, so that the signal power can be expressed as
2 2
hm ,n .
Ps = 2R2 Pt PLO Em
As in [13][17], we also consider that PLO  Ps and thus, the
dc photocurrent can be approximated as iDC (t) RPLO . The
photodetection process is impaired by shot noise with variance
2
= 2qe RPLO Be where Be is the electrical bandwidth of
shot,L
the photodetector. It is also noted that because of the large value
of RPLO the photocurrent due to thermal noise and the dark
current can be ignored [16].
Following [15] and[17], the sufficient statistics at the nth
coherent receiver can be expressed as

IE
E
Pr E
oo
f

Let us consider an M N MIMO FSO system with M transmit units (lasers) and N coherent receivers. It is assumed that
the receiving apertures are separated by more than a coherence
wavelength to ensure the independency of fading channels. The
basic principle of OSM modulation is as follows [18], [20]:
1) The transmitter encodes blocks of log2 (M ) data bits into
the index of a single transmit unit. Such a block of bits
is hereafter referred to as message and is denoted by
bm , m = 1, 2, . . . , M . It is assumed that the M messages are transmitted by the encoder with equal probability and that the related transmitted signal is denoted by
m = Em exp(jb ). During each time slot, only one
E
m
transmitter , where  = 1, 2, . . . , M is active for data
transmission. The information bits are modulated on the
electric field of an optical signal beam through an external
modulator. During each particular time slot, the remaining
transmit lasers are kept silent, i.e., they do not transmit.
2) At the receiver, the incoming optical field is mixed with a
LO field and the combined wave is first converted by the
photodetector to an electrical one. A bandpass filter is then
employed to extract the intermediate frequency (IF) component of the total output current. Finally, a N -hypothesis
detection problem is solved to retrieve the active transmit
unit index, which results in the estimation of the unique
sequence of bits emitted by the transmitter.

The received electric field at the aperture plane of the nth

receiver after mixing with a LO beam, can be expressed as [16],
[17]

2Pt Z0 Em hm ,n cos(0 t + m ,n + b m )
en (t) =

+ 2PLO Z0 cos(LO t).
(1)

In the above equation, Pt is the transmit laser power, Z0 is the

free space impedance, hm ,n and m ,n denote the magnitude and
the phase of the complex channel coefficient between the mth
transmit and the nth receive aperture, respectively. Furthermore,
PLO denotes the power of the LO, LO = 0 + IF where 0
and IF are the carrier and the intermediate radian frequencies,
respectively.
The output current of the nth photodetector can be mathematically expressed as [16], [17]
R
in (t) =
[en (t)]2
Z0

yn =

hm ,n Em exp[j(m ,n + b m )] + zn

where = RPt /(qe Be ) is the average signal-to-noise ratio

(SNR) and zn is the noise at the nth receiver. Assuming that
the LO power is large and the receiver noise is dominated by
the LO related noise terms, the additive white Gaussian noise
model can be employed as an accurate approximation of the
Poisson photon-counting detection model [16], [17]. Thus, zn
can be modeled as a zero-mean unit variance complex Gaussian
random variable (RV) [17].
Similar to [27], it is assumed that the receiver has knowledge
of the actual fading gains and that the fading process remains
constant over the bit interval and changes from bit to bit in an
independent manner. At the receiver, the optimal spatial modulation detector estimates the active transmitter index, , at a
given time slot according to [28]
 = argmaxpy (y|x, H)


= argmin
(2)

where R = qe /(h0 ) is the responsivity of the photodetector

with qe = 1.6 1019 Cb is the charge of an electron, h =
6.6 1034 J s is the Planck constant, is the photodetector
efficiency, and 0 = 0 /(2) is the optical center frequency.

(4)



 h x 2F 2 yT h x

(5)

where
- x is an M -dimensional vector with elements corresponding
to the electrical field Em exp(jb m ) transmitted over the
optical MIMO channel;

JOURNAL OF LIGHTWAVE TECHNOLOGY

- H(t) is an N M optical MIMO channel defined as

H(t) = [h1 , h2 , . . . , hM ]

h1 1 (t) exp(j1 1 )

h (t) exp(j )
21
21


..

hN 1 (t) exp(jN 1 )

...
...

..

...

h1 M (t) exp(j1 M )

h2 M (t) exp(j2 M )

..

hN M (t) exp(jN M )
(6)

- z is the N -dimensional noise vector;

- py (y|x, H) is the PDF of y conditioned on the transmitted
vector x and the channel H.

form. Specifically, for Aij = 0 (9) reduces to the K-distribution

whereas for = 1, (9) reduces to a special case of the I-K
distribution [10, Eq. (10)].
The th normalized moment of Iij is given by [10, Eq. (23)]
as
 

E{Iij }
(ij + k) (ij ij )
!
=
(1 + )
k
E{Iij }
ij
(ij )
!
ij
k =0
(10)
where ij  A2ij /b0 i j is the coherence parameter, defined as
the power ratio of mean intensities of the constant-amplitude
component and random component of the field in (7) [10], [11].
Using (10), the scintillation index can be readily calculated as
I2i j 

C. Channel Model

(11)

Under the assumption of plane wave propagation, I2i j can be

directly related to atmospheric conditions as [11, Eq. (7), Eq.
(9)]
 2
1 (1 + 0.512 ), 1  1
2
I i j
(12)
4/5
1 + 0.86/1 , 1  1

IE
E
Pr E
oo
f

A discrete scattering channel model is considered, where the

radiation field of an optical wave at a particular point is assumed
to be composed of a number of scattered components that have
propagated through different paths. Under the Ricean assump ij (t) between the
tion [10], the complex channel path gains h
ith transmitter and the jth photodetector can be expressed as
ij (t) = hij (t) exp(jt) where is the radian frequency of the
h
optical wave and

E{I2ij }
ij + 2ij ij + 2
1=
.
E{Iij }2
ij (1 + ij )2

hij (t) = {hij (t)} + j{hij (t)}

= Aij exp[jij (t)] + Rij (t) exp[jij (t)]

(7)

where the term Aij exp[jij (t)] is a deterministic component

and Rij (t) exp[jij (t)] is a circular complex Gaussian RV.
Hence, the amplitude Rij is Rayleigh distributed with scale
2
= bij /2 [10, Eq. (13)] and the phase ij is uniparameter ij
formly distributed over [0, 2). Under the assumption of a doubly stochastic scintillation model [10], the effect of random
fluctuations in the turbulence parameters is modeled by allowing random variations in the parameter bij of the Rayleigh component. Following [10], it is further assumed that bij follows a
Gamma distribution with PDF given by


 i j i j 1

ij
b
b
exp
fb i j (b) =
(8)
b0
(ij )
b0 i j
where is the shaping parameter which represents the effective
number of scatters and b0 i j = E{bij }. Then, the PDF of the
irradiance Iij = |hij (t)|2 , fI i j (I), can be expressed as [10, Eq.
(8)]


ij /b0 i j i j
fI i j (I) =
(ij )



ij b I + Aij 2

b i j 2 exp

b0 i j
b
0


2Aij I
db
(9)
I0
b
which is actually the integral representation of the H-K distribution [10]. It is noted that, with the exception of the special
cases Aij = 0 or = 1, fI i j (I) cannot be expressed in closed

11/6

where 12 = 1.23Cn2 i j k 7/6 Lij

is the Rytov variance, k =
2/ is the optical wave number with being the wavelength, Lij is the link distance and Cn i j denotes the index of
refraction structure parameter. For FSO links which operate
near the ground plane, Cn2 i j 1.7 1014 m2/3 and 8.4
1015 m2/3 for the daytime and night time periods, respectively [29]. Moreover, 1  1 and 1  1 correspond to weak
and strong turbulence conditions, respectively.
Using (12), the parameters of the H-K distribution, and ,
can be directly related to physical parameters of the turbulence
by following a similar line of arguments as in [11], where similar
results have been derived for the I-K distribution. In particular,
on the one hand, weak turbulence conditions are characterized
in the H-K distribution by large values of ij . In this case the
scintillation index given by (11) can be approximated as
I2i j

2
,
ij

with ij  1.

(13)

On the other hand assuming strong turbulence conditions, in

which case ij goes towards zero, (11) can be approximated as
I2i j 1 +

2
,
ij

with ij  1.

(14)

By comparing (13) and (14) with the first and second branches
of (12), respectively, ij and ij can be obtained as
4/5

ij = 2.331 i j
ij =

2
.
12i j (1 + 0.512i j )

(15)
(16)

To the best of our knowledge, the relationship of ij and ij with

1 i j given by (15) and (16) has not been previously published.

PEPPAS AND MATHIOPOULOS: FREE-SPACE OPTICAL COMMUNICATION WITH SPATIAL MODULATION AND COHERENT DETECTION

III. PERFORMANCE ANALYSIS OF UNCODED OSM

In this section, by employing the well-known moment generating function (MGF)-based approach for the performance
analysis of digital communications over fading channels [30],
analytical expressions for the ABEP of uncoded OSM systems
will be derived. In addition, expressions for the diversity and
coding gains of OSM systems are also presented, thus providing useful insight as to how these parameters affect the overall
system performance.
A. Preliminaries

n has uncorrelated components too and its squared envelope,

|n |2 , is characterized by a non-central chi-square PDF

  
An x
1
x + A2n
f| n |2 (x|bn ) =
exp
I0
(21)
2bn
2bn
bn
where An = |A2,n exp(j2,n ) A1,n exp(j1,n )|. Assuming
that bn follows the Gamma with parameters n and b0,n , the
PDF of |n |2 is obtained by averaging (21) with respect to bn ,
i.e.,
f| n |2 (x) =

For M = 2, the conditional bit error probability (CBEP) of

OSM systems when no turbulence induced fading is considered
can be obtained in closed form as [27]



2
 h1 h2 F .
(17)
PE (h1 , h2 ) = Q
4

(n /b0,n ) n
2(n )



n bn
x + A2n
n 2
bn
exp

b0,n
2bn
0
 
An x
I0
(22)
dbn .
bn

The squared Frobenius norm in (17) can be expressed as

|h1,n h2,n |2

(18)

n =0

where hi,n is the nth element of hi , i {1, 2}. When M > 2

transmitters are considered, a tight upper bound for the CBEP
of the above system can be obtained as [18, Eq. (7)]
M 1
PE (H)
log2 (M )

M


M


As was pointed out in [10], f| n |2 (x) cant be expressed in

closed form. Nevertheless, for the special case of n = 1, i.e.,
when one scatterer per branch is considered, and by employing
[10, Eq. (10)], this integral can be evaluated in closed form as

IE
E
Pr E
oo
f

 h1 h2 2F =

N


f| n |2 (x)

 

,
K
2
A
/b
2x/b
I
0
n
0,n
0
0,n
b
0,n


=



I0
2x/b0,n ,
2An /b0,n K0
b0,n

Nb (m1 , m2 )PEP(m1 m2 ) (19)

m 1 =1 m 2 = m 1 =1

where PEP(m1 m2 ) denotes the pairwise error probability

(PEP) related to the pair of transmitters m1 and m2 , where m1
and m2 1, 2, . . . , M , and Nb (m1 , m2 ) is the number of bits in
error which have occurred when the receiver decides incorrectly
that m2 instead of m1 has been active. The PEP(m1 m2 )
can be evaluated as [18, Eq. (8)]



2
 hm 1 hm 2 F .
PEP(m1 m2 ) = Q
(20)
4
B. MGF-Based Approach

When atmospheric turbulence is included in the analysis the

conditional error probabilities in (17) and (19) need to be averaged over the elements of the channel matrix H in order to
evaluate the ABEP. Without loss of generality, let us consider the
case of a 2 N MIMO system. Since hi,n are complex Gaussian RVs, the difference n = h1,n h2,n is also a complex
Gaussian RV having mean equal to the difference of the means
of hi,n and variance equal to the sum of variances of hi,n . In order to deduce a closed form expression for the ABEP, it is further
assumed that hi,n have uncorrelated real and imaginary components with the same variance n2 = bn /2. It is noted that this
assumption has been experimentally verified in the past for link
distances of the order of kilometer and for aperture separation
distances of the order of centimeter [31], [32]. Consequently,

x < A2n

x > A2n .

(23)

Moreover, for the special case where h1,n and h2,n have identical mean value, i.e., when An = 0, (22) yields the well-known
K-distribution with PDF given by

( n 1)/2
n x
(1 n )/2
f| n |2 (x) = 2
(n )
b0,n


2n x
K n 1
.
(24)
b0,n
By employing the MGF-based approach for the performance
analysis of digital communications over fading channels, the
average PEP (APEP) can be obtained as

N
 
1 /2  
M| n |2
d.
(25)
APEP =
0
8 sin2
n =1

Moreover, using the tight approximation for the Gaussian Qfunction presented in [33, Eq. (14)] (i.e., Q(x) 1/12 exp
(x2 ) + 1/4 exp(2x2 /3)), an expression accurately approximating APEP can be deduced as
N
N


1 
1 
M| n |2
M| n |2
+
.
12 n =1
8
4 n =1
6
(26)
In the following, analytical expressions for the MGF of |n |2
will be obtained.

APEP

JOURNAL OF LIGHTWAVE TECHNOLOGY

Proposition 1: An integral representation for the MGF of

|n |2 can be deduced as
M| n |2 (s) =

tj and wj are given as


$
2j 1

+
2J
4


2 sin 2j2J1
%
&.


wj =
4J cos2 4 cos 2j2J1 + 4
#

tj = tan

(n /b0,n )
(n )


 n 1
An s
n b
b
exp

db.
2bs + 1
2bs + 1 b0,n
0
(27)

Proof: By employing the definition of the MGF, M| n |2 (s)

can be obtained as

exp(sx)f| n |2 (x)dx
M| n |2 (s) =

By changing the order of integration, the above equation can be

expressed as
n

M| n |2 (s) =

(n /b0 , n )
n b
b n 2 exp
2(n )
b0 , n
0

!
   "

An x
x + A2n
exp sx
I0
dx db.
2b
b
0

(29)

The inner integral, i.e., with respect to x can be evaluated by

employing [34, Eq. (3.15.2.2)] as
  


An x
x + A2n
I0
dx
exp sx
2b
b
0
#
$
1
2b
=
exp
.
(30)
2sb + 1
2An b(2sb + 1)

Substituting (30) into (29) and after some straightforward mathematical manipulations, (27) is readily deduced thus completing
the proof.

The integral in (27) can be accurately approximated by employing a GaussChebyshev quadrature (GCQ) technique as
[35]
M| n |2 (s)

(n /b0,n )
(n )

(32a)
(32b)


M| n |2 (s) =

n
2sb0,n

 n /2
exp

W n /2,( n 1)/2

n
4sb0,n
n
2sb0,n



.

(33)

This result can be readily deduced by employing the integral

representation of the Whittaker W -function given in [26, Eq.
(9.222)]. It is noted that (33) is in agreement with a previously
known result, namely the analytical expression for the MGF of
the K-distribution [37, Eq. (4)].

IE
E
Pr E
oo
f

(n /b0,n ) n
2(n )


 
n b x + A2n
exp sx

b0,n
2b
0
0
 
An x
I0
b n 2 dbdx.
(28)
b

For the special case of An = 0, (27) can be evaluated in closed

form, as it will be shown next.
Corrolary 1: For the special case of A = 0 the MGF of
|n |2 can be deduced in closed form as

cos
4



An s
n tj
tj n 1
exp

wj

2tj s + 1
2tj s + 1
b0,n
j =0
J


(31)
where J is the number of integration points, tj are the abscissas
and wj the corresponding weights. In [36, eqs. (22) and (23)],

C. Diversity Gain Analysis

The diversity gain of the considered OSM MIMO system can

be obtained by using the approach presented in [38]. In particular, a generic analytical expression, which becomes asymptotically tight at high SNR values, will be derived for the APEP
appearing in (25), as follows:
Proposition 2: For high SNR values, (25) can be approximated by
"
! N

 N
N 1

N + 12
1 2
c
(34)
APEP
4
(N + 1) n =1

where

cn =

( n 1)/2

(n /b0,n )( n +1)/2
(n )


2An n
.
K n 1
b0,n
An
2

(35)

Proof: According to [38, Proposition 3], the asymptotic error

performance of the OSM system depends on the behavior of
M| n |2 (s), as s . To determine an analytical asymptotic
expression for APEP a Taylor series expansion is employed to
approximate M| n |2 (s) as
|M| n |2 (s)| = cn |s|d n + o(|s|d n ),

(36)

where cn and dn are parameters that determine the diversity and

coding gains of the nth diversity branch, respectively. Observe
s
s

that since As/(2sb

+ 1) A/(2b)
and 1/(2sb + 1)

PEPPAS AND MATHIOPOULOS: FREE-SPACE OPTICAL COMMUNICATION WITH SPATIAL MODULATION AND COHERENT DETECTION

1/(2bs), (27) yields

M| n |2 (s)

(n /b0,n ) n
2s(n )



An
n b
n 2

b
exp

db. (37)
2b
b0,n
0

By employing [34, Eq. (2.2.2.1)], (37) can be solved in closed

form yielding

( n 1)/2
An
(n /b0,n )( n +1)/2
M| n |2 (s)
2
s(n )


2An n
.
(38)
K n 1
b0,n

A union bound on the ABEP of a coded communication

system can be obtained as [30]

1
P (X)
q(X, X )PEP(X, X )
(41)
Pub
n

X

X= X

where P (X) is the probability that the coded sequence X is

transmitted, q(X, X ) is the number of information bit errors
in choosing another coded sequence X instead of X n is the
number of information bits per transmission and PEP(X, X )
is the PEP, i.e., the probability of selecting X when X was
actually transmitted.
By employing [30, p. 510], (41) can be efficiently evaluated
as
,
$
 /2 #
,

1

T [D(), N ],,
P (X)
Pub
(42)
n
N
0

N =1

where T [D(), N ] is the transfer function of the employed convolutional code, N is an indicator variable taking into account
the number of the erroneous bits and D() depends on the underlying PEP expression. Furthermore, assuming that uniform
error probability (UEP) codes are considered and taking into
account the symmetry property this code family exhibits, thus
making the distance structure of a UEP code independent of the
transmitted sequence, (42) can be further simplified as [30]
,
#
$

,
1 /2 1
T [D(), N ],,
.
(43)
Pub
0
n N
N =1

IE
E
Pr E
oo
f

By comparing (38) and (36) it is readily deduced that dn = 1

and cn is given by (35). Thus, by substituting (36) into (25),
the asymptotic PEP expression can be obtained as in (34) which
concludes the proof.

From (34) it is clear that the diversity gain achieved by the
considered system is equal to N . It is also evident that the diversity gain depends only on the number of the receive apertures
and is independent of the fading severity. This finding is in agreement with other equivalent observations reported in [27], [39]
and[40], which have been obtained for the case of RF MIMO
wireless systems.
It is noted that for the special case An = 0, i.e., when |n |2
follows the K-distribution, by employing the asymptotic result

x0

Kt (x) ((t)/2) (2/x)t [35], cn can be further simplified as

n
.
(39)
cn =
2b0,n (n 1)

For M = 2, using (17), (18) and Craigs formula for the Q /2

function, i.e., Q(x) = 1/ 0 exp(x2 /2 sin2 )d, D()
can be expressed as
D() =

N


n =1

IV. PERFORMANCE ANALYSIS OF CODED OSM OVER

TURBULENCE CHANNELS

When coded OSM is employed, the input signal s(t) is first

encoded by a convolutional encoder. The encoded data are interleaved by a random block interleaver and transmitted through the
optical wireless channels using spatial modulation. It is assumed
that perfect interleaving at the transmitter and de-interleaving
at the receiver is used. Considering maximum likelihood soft
decision decoding, the log likelihood ratios (LLRs) for the ith
constellation bit when the th transmitting antenna is active are
computed as [18, Eq. (6)]
Pr{i = 1|y}
LLR = log
Pr{i = 0|y}


+
2
i exp  y hs  /N0
L
1


= log +
2
i exp  y hs  /N0
L

M| n |2


8 sin2

(44)

where M| n |2 can be obtained from (27). When M > 2, by

employing [18, Eq. (13)], and using a similar line of arguments
as for the case of M = 2, D() can be written as
M


M


m 1 =1 m 2 = m 1 =1

M| m 1 , m 2 |2


8 sin2

(45)

where |m 1 ,m 2 |2 = hm 1 hm 2 2 . The last MGF can be analytically computed with the help of (27) in a straight-forward
way.
V. DIVERSITY TECHNIQUES FOR COHERENT FSO SYSTEMS

(40)

where L {1 : M } is the set of spatial constellation points, Li1

and Li0 are subsets from L containing the transmitter indices
having 1 and 0 at the ith bit, respectively. The resulting data
are finally decoded by a Viterbi decoder.

Diversity at the transmitter and/or the receiver is a commonly

used technique to mitigate the deleterious impact of atmospheric
turbulence on coherent FSO system performance. In this section, firstly the most popular diversity techniques for coherent
FSO systems will be presented [41]. Such diversity schemes
will be also considered for the FSO communication systems investigated in this paper as follows: Receive diversity by employing MRC or SC and transmit diversity based on the Alamouti
scheme.

JOURNAL OF LIGHTWAVE TECHNOLOGY

The instantaneous SNR at the output of the coherent MRC

receiver assuming equal average SNR per receiving aperture,
can be expressed as [41, Eq. (11)]
M RC =

N


In

(46)

n =1

where In denotes the optical signal irradiance at the nth branch.

The ABEP of coherent MRC receivers with differential phase
shift keying (DPSK) can be deduced as [13, Eq. (14)]
N
1 
MI n ()
PE =
2 n =1

(47)

whereas for binary phase shift keying (BPSK) as [41, Eq. (18)]
1
PE =


0

N
/2 
n =1

MI n

 
d.
sin2

(48)

The instantaneous SNR at the output of the coherent SC

receiver is given by [41, Eq. (15)]
(49)

IE
E
Pr E
oo
f

SC = max{I1 , I2 }.

Fig. 1. ABEP of uncoded OSM for 2 N MIMO H-K turbulent channels as

a function of the average SNR, , for various number of receiving apertures,
N . Simulation Parameters: A 1 , n = 2, A 2 , n = 1, 1 , n = /3, 2 , n = /4,
n = 2, b0 , n = 2.

For the SC case, an analytical expression for the ABEP is much

more difficult to be deduced and, thus for this case the ABEP will
be evaluated exclusively by means of Monte Carlo simulations.
In [16], transmit diversity techniques, based on the Alamouti scheme, were explored for coherent OWC applications by
employing STBC. The instantaneous SNR at the input of the
demodulator of the optical receiver has a similar form as (46)
[16], and therefore the ABEP of BPSK can be readily evaluated
by employing (48).
VI. PERFORMANCE EVALUATION RESULTS AND DISCUSSION

In this section the various performance evaluation results

which have been obtained by numerically evaluating the mathematical expressions presented in Sections III and IV for uncoded and coded OSM systems operating over H-K turbulent
channels will be presented. In particular, for uncoded OSM systems, ABEP vs. SNR for 2 N OSM systems (obtained using
(26) with (27), and (34)) the results are presented in Figs. 14.
For the uncoded schemes, in order to validate the accuracy of
the previously mentioned expressions, comparisons with complementary Monte Carlo simulated performance results are also
included in these figures. As far as the performance of coded
OSM systems is concerned, ABEP upper bounds vs. SNR have
been obtained using (43) with (27) and the results are illustrated
in Fig. 5. Table VI summarizes the different values of the system
parameters used to obtain the simulated performance evaluation
results. Note that in obtaining the various performance results
a system with M = 2 transmitting apertures was considered.
The motivation behind this choice is the inherent design complexities and cost of adopting multiple transmitters in current
state-of-the-art coherent optical wireless links [16], as well as
the fact that the proposed framework yields exact results for
M = 2 transmitters.

Fig. 2. ABEP of uncoded OSM for 2 2 and 2 4 MIMO H-K turbulent

channels as a function of the average SNR, , for various values of link distances,
L. Simulation Parameters: = 1550 nm, C n2 = 1.7 10 1 4 m 2 / 3 , 1 , n =
/3, 2 , n = /4, b0 , n = 1.

In order to demonstrate the impact of the number of the receiving apertures on the system performance, Fig. 1, presents
the ABEP performance as a function of the average SNR, ,
of 2 N MIMO OSM systems with N {1, 2, 3, 4}. Independent and identically distributed branches are considered
with A1,n = 2, A2,n = 1, 1,n = /3, 2,n = /4, n = 2,
b0,n = 2. The obtained results clearly indicate that the ABEP

PEPPAS AND MATHIOPOULOS: FREE-SPACE OPTICAL COMMUNICATION WITH SPATIAL MODULATION AND COHERENT DETECTION

Fig. 5. ABEP upper bounds of convolutional coded OSM for 2 2 and 2

1 H-K turbulent channels as a function of the average SNR, , for various
values of link distances, L. Simulation Parameters: = 1550 nm, C n2 = 1.7
10 1 4 m 2 / 3 , 1 , n = /3, 2 , n = /4, b0 , n = 1.

IE
E
Pr E
oo
f

Fig. 3. ABEP of uncoded OSM for 2 3 MIMO H-K turbulent channels as

a function of the average SNR, , for weak (C n2 = 10 1 5 m 2 / 3 ) and strong
C n2 = 3 10 1 4 m 2 / 3 turbulence and for various values of b0 , n . Simulation
Parameters: = 1550 nm, 1 , n = /3, 2 , n = /4.

TABLE II
SYSTEM PARAMETERS FOR SIMULATION

Fig. 4. ABEP Comparison of 2 2 OSM with 1 2 coherent MRC systems

employing DPSK, as a function of the average SNR, , for various values
of A 1 , n . Simulation Parameters: A 2 , n = 0, 1 , n = 0, 2 , n = 0, n = 1.5,
b0 , n = 1.5.

curves, obtained using (26), are in close agreement with the

performance obtained via simulations, thus verifying the correctness of the proposed analysis. Moreover, it is evident that
the asymptotic ABEP curves accurately predict the diversity
gain achieved by the considered system for all test cases.
In Fig. 2, the dependence on the link distance, L, of the ABEP
of a 2 N MIMO OSM system is illustrated. The considered

Parameter

Value

Operating wavelength, (nm)

Refractive index parameter, C n2 (m 2 / 3 )
Link distance, L (m)
1 , n
2 , n
A1,n
A2,n
b0 , n
Number of Transmitting apertures, M
Number of Receiving apertures, N

1500
{3, 1.7, 0.1} 10 1 4 ,
{500, 1000, 1500, 2000}
/3
/4
{0, 1, 2, 3}
{0, 1}
{0.25, 0.5, 1, 1.5, 2}
2
{1, 2, 3, 4}

system is again equipped with either N = 2 or N = 4 receiving

apertures and identically distributed branches are assumed. The
parameters of the H-K distribution are calculated from (15)
and (16) assuming plane wave propagation. Following [42], it
is further assumed that the operating wavelength is = 1550
nm and Cn2 = 1.7 1014 m2/3 , which, as it was pointed out
in Section II, is a typical value of refractive index for FSO
links operating near the ground plane during daytime [29]. As
expected, the error performance deteriorates as L increases from
L = 500 m to L = 1500 m. Moreover, it is evident that an
increase in L from 500 to 1000 m leading to a more severe
performance deterioration than in the case where L increases
from 1000 to 1500 m. For all cases considered, the analytical
results obtained using (26) are compared with the equivalent
results obtained by means of Monte-Carlo computer simulations
and again they match very well.
In Fig. 3 the impact of the turbulence strength as well as
of the parameter b0 on system performance is illustrated. In

10

JOURNAL OF LIGHTWAVE TECHNOLOGY

in [18], OSM offers increased spectral efficiency by a factor

log2 (M ). Moreover, as only one transmitting aperture is activated at any bit duration, OSM has a lower decoding complexity
as compared to conventional MRC and Alamouti schemes.
In Fig. 5, upper bounds on the ABEP of convolutional coded
2 1 and 2 1 OSM systems are depicted, assuming similar
propagation conditions to those considered in Fig. 2. Considering a convolutional code with rate 1/3 and constraint length of
3, its transfer function is given as [44, Eq. (8.2.6)]
T [D(), N ] =

D()6 N
.
1 2N D()2

(50)

Substituting (50) into (43), a union bound on the ABEP can be

obtained as
 /2
D()6
1

Pub
d.
(51)
log2 (M ) 0
(1 2D()2 )2
The performance results of Fig. 5 clearly show that, as expected,
the incorporation of convolutional coding significantly enhances
the performance of OSM systems, even when a small number
of receive apertures is employed.

IE
E
Pr E
oo
f

particular, Fig. 3 presents the ABEP performance as a function

of the average SNR, , of 2 N MIMO OSM systems with
N = 3 receiving apertures, link distance L = 2000 m, and various values of b0 . Two different values of the refractive index
parameter Cn2 are considered, i.e., Cn2 = 3 1014 m2/3 and
Cn2 = 1015 m2/3 to account for strong and weak turbulence,
respectively [29], [43]. The obtained results clearly show that
the error performance improves as b0 increases and/or Cn2 decreases. Moreover, the analytical results obtained using (26)
agree well with the equivalent results obtained by means of
MonteCarlo computer simulations.
Next we compare the proposed OSM system with two alternative coherent FSO systems that can provide performance
enhancements by means of transmit, i.e., multiple-input singleoutput (MISO) or receive diversity i.e. single-input multipleoutput (SIMO). It is noted that for similar aperture configurations, a fair comparison between coherent and IM/DD systems
seems difficult as the same received laser power leads to different
SNRs for each of these schemes [17]. On the other hand, in order
to perform a fair comparison between OSM and the alternative
MISO or SIMO systems under the same propagation channel
conditions, the aperture configuration of the FSO systems under
comparison should be carefully selected. Specifically, because
of the fact that the diversity gain of OSM equals to only the
number of the receive apertures, i.e., no transmit diversity gain
can be achieved, the number of transmit or receive apertures of
the alternative systems must be hence selected to be equal to the
number of receive apertures of the OSM system. To this end and
for a fair comparison in our paper a 2 2 OSM system is compared with the following two alternative FSO communication
systems which also employ coherent detection:
1) A 1 2 heterodyne FSO communication system which
employs DPSK [13] and MRC or SC;
2) A 2 1 coherent FSO system employing the Alamouti
scheme [16] and BPSK.
In order to simplify the underlying mathematical analysis,
it is assumed that the PDF of In is given by (9) with the parameters An being all zero, i.e., the PDF considered here is the
K-distribution. Thus, MI n () can be readily obtained in closed
form from (33) by replacing b0,n with b0,n /2. In Fig. 4, the
ABEP of 2 2 MIMO OSM links is compared with the ABEP
of 1 2 coherent FSO systems with DPSK considering identically distributed links. In order to compare these systems under
the same propagation conditions, it is assumed that n = 1, 5,
b0,n = 1.5, A2,n = 0 and A1,n = {0, 1, 2, 3}. As it can be observed, when either MRC or SC are employed, although coherent DPSK performs worse than the OSM for values of A1,n up
to approximately 1, it outperforms OSM at lower values of A1,n .
Moreover, although the OSM outperforms the Alamouti scheme
for A1,n = 2 and 3, it performs similarly for high SNR values
when A1,n = 1. It is noted that for A1,n = 1 and lower values
of A1,n the Alamouti scheme yields the best performance of the
considered schemes. However, when more transmit apertures
are employed, this advantage is compensated by the superior
spectral efficiency of OSM and its lower hardware complexity as compared to coherent MRC. Specifically, as pointed out

VII. CONCLUSION

In this paper, the use of spatial modulation technique for

coherent FSO communication systems has been proposed. We
have provided a comprehensive analytical framework for error
performance analysis which is valid for a great deal of atmospheric turbulence scattering channel models but focusing on
the H-K distribution. The proposed framework has revealed important information about the performance of OSM over such
turbulent channels, including the effect of fading severity and
the achievable diversity gain. In particular, it has been shown
that the diversity gain depends on the number of receive apertures only while channel parameters affect the coding gain only.
Upper bounds for the ABEP performance of coded OSM systems have also been derived, demonstrating that coding techniques can greatly enhance the performance of OSM. Extensive
computer simulation performance evaluation results have been
also obtained which have verified the accuracy of the analytical approach. It was shown that OSM can provide significant
performance enhancements in the presence of H-K atmospheric
turbulence. The improvements are comparable to the ones offered by conventional coherent systems with spatial diversity,
while outperforming the latter in terms of spectral efficiency
and hardware complexity. Specifically, OSM offers increased
spectral efficiency by a factor log2 (M ) when multiple transmitting apertures are employed while it completely avoids multiple
communication chains in a similar fashion as in RF systems.
Besides, under specific propagation conditions, OSM can yield
better performance than conventional SIMO systems employing
MRC or SC. The proposed framework, apart from enabling the
computation of accurate performance evaluation results in an
efficient manner, it also provides a useful tool for understanding the performance trend, important properties and tradeoffs of

PEPPAS AND MATHIOPOULOS: FREE-SPACE OPTICAL COMMUNICATION WITH SPATIAL MODULATION AND COHERENT DETECTION

outdoor OSM operating in the presence of atmospheric turbulence.

REFERENCES

[22] R. Y. Mesleh, H. Haas, S. Sinanovic, C. Ahn, and S. Yun, Spatial modulation, IEEE Trans. Veh. Technol., vol. 57, no. 4, pp. 22282241, Jul.
2008.
[23] R. Mesleh, M. D. Renzo, H. Haas, and P. M. Grant, Trellis coded spatial
modulation, IEEE Trans. Wireless Commun., vol. 9, no. 7, pp. 23492361,
Jul. 2010.
[24] S.-H. Hwang and Y. Cheng, SIM/SM-aided free-space optical communication with receiver diversity, J. Lightw. Technol., vol. 32, no. 14,
pp. 24432450, Jul. 2014.
[25] T. Ozbilgin and M. Koca, Optical spatial modulation over atmospheric
turbulence channels, J. Lightw. Technol., vol. 33, no. 11, pp. 23132323,
Jun. 2015.
[26] I. Gradshteyn and I. M. Ryzhik, Tables of Integrals, Series, and Products,
6th ed. New York, NY, USA: Academic, 2000.
[27] M. D. Renzo and H. Haas, Space shift keying (SSK) MIMO over correlated Rician fading channels: Performance analysis and a new method for
transmit-diversity, IEEE Trans. Commun., vol. 59, no. 1, pp. 116129,
Jan. 2011.
[28] J. Jeganathan, A. Ghrayeb, and L. Szczecinski, Spatial modulation: Optimal detection and performance analysis, IEEE Commun. Lett., vol. 12,
no. 8, pp. 545547, Aug. 2008.
[29] J. W. Goodman, Statistical Optics. New York, NY, USA: Wiley, 1985.
[30] M. K. Simon and M. S. Alouini, Digital Communication over Fading
Channels, 2nd ed. New York, NY, USA: Wiley, 2005.
[31] E. Lee and V. Chan, Part 1: Optical communication over the clear turbulent atmospheric channel using diversity, IEEE J. Sel. Areas Commun.,
vol. 22, no. 9, pp. 18961906, Nov. 2004.
[32] N. Letzepis, I. Holland, and W. Cowley, The Gaussian free space optical MIMO channel with Qary pulse position modulation, IEEE Trans.
Wireless Commun., vol. 7, no. 5, pp. 17441753, May 2008.
[33] M. Chiani, D. Dardari, and M. K. Simon, New exponential bounds and
approximations for the computation of error probability in fading channels, IEEE Trans. Wireless Commun., vol. 2, no. 4, pp. 840845, Jul.
2003.
[34] A. P. Prudnikov, Y. A. Brychkov, and O. I. Marichev, Integrals and Series
Volume 4: Direct Laplace Transforms, 1st ed. Boca Raton, FL, USA: CRC,
1992.
[35] M. Abramovitz and I. Stegun, Handbook of Mathematical Functions with
Formulas, Graphs, and Mathematical Tables. New York, NY, USA: Dover,
1964.
[36] F. Yilmaz and M.-S. Alouini, An MGF-based capacity analysis of equal
gain combining over fading channels, in Proc. IEEE 21st Int. Symp. Pers.
Indoor Mobile Radio Commun., Sep. 2010, pp. 945950.
[37] P. Theofilakos, A. G. Kanatas, and G. P. Efthymoglou, Performance of
generalized selection combining receivers in K fading channels, IEEE
Commun. Lett., vol. 12, no. 11, pp. 816818, Nov. 2008.
[38] Z. Wang and G. Giannakis, A simple and general parametrization quantifying performance in fading channels, IEEE Trans. Commun., vol. 51,
no. 8, pp. 13891398, Aug. 2003.
[39] M. D. Renzo and H. Haas, Bit error probability of space modulation over
Nakagami-m fading: Asymptotic analysis, IEEE Commun.Lett., vol. 15,
no. 10, pp. 10261028, Oct. 2011.
[40] K. P. Peppas, M. Zamkotsian, F. Lazarakis, and P. G. Cottis, Asymptotic error performance analysis of spatial modulation under generalized
fading, IEEE Wireless Commun. Lett., vol. 3, no. 4, pp. 421424, Aug.
2014.
[41] M. Niu, J. Cheng, and J. F. Holzman, Error rate performance comparison of coherent and subcarrier intensity modulated optical wireless
communications, J. Opt. Commun. Netw., vol. 5, no. 6, pp. 554564, Jun.
2013.
[42] M. Uysal, J. T. Li, and M. Yu, Error rate performance analysis of coded
free-space optical links over gamma-gamma atmospheric turbulence channels, IEEE Trans. Wireless Commun., vol. 5, no. 6, pp. 12291233, Jun.
2006.
[43] H. E. Nistazakis, E. A. Karagianni, A. D. Tsigopoulos, M. E. Fafalios,
and G. S. Tombras, Average capacity of optical wireless communication systems over atmospheric turbulence channels, J. Lightw. Technol.,
vol. 27, no. 8, pp. 974979, Apr. 2009.
[44] J. G. Proakis, Digital Communications, 3rd ed. New York, NY, USA: Mc
Graw Hill, 1995.

IE
E
Pr E
oo
f

[1] X. Zhu and J. M. Kahn, Free-space optical communications through

atmospheric turbulence channels, IEEE Trans. Commun., vol. 50, no. 8,
pp. 12931300, Aug. 2002.
[2] A. Garcia-Zambrana, Error rate performance for STBC in free-space
optical communications through strong atmospheric turbulence, IEEE
Commun. Lett., vol. 11, no. 5, pp. 390392, May 2007.
[3] S. M. Navidpour, M. Uysal, and M. Kavehrad, Performance of freespace optical transmission with spatial diversity, IEEE Trans. Wireless
Commun., vol. 6, no. 8, pp. 28132819, Aug. 2007.
[4] K. Peppas, A simple, accurate approximation to the sum of gammagamma variates and applications in MIMO free-space optical systems,
IEEE Photon. Technol. Lett., vol. 23, no. 13, pp. 839841, Jul. 2011.
[5] K. Peppas, F. Lazarakis, A. Alexandridis, and K. Dangakis, Simple,
accurate formula for the average bit error probability of multiple-input
multiple-output free-space optical links over negative exponential turbulence channels, Opt. Lett., vol. 37, pp. 32433245, Aug. 2012.
[6] L. C. Andrews, M. A. Al-Habash, C. Y. Hopen, and R. L. Phillips, Theory of optical scintillation: Gaussian beam wave model, Waves Random
Media, vol. 11, pp. 271291, 2001.
[7] M. A. Al-Habash, L. C. Andrews, and R. L. Phillips, Mathematical model
for the irradiance PDF of a laser beam propagating through turbulent
media, Opt. Eng, vol. 40, no. 8, pp. 15541562, 2001.
[8] J. H. Churnside and S. F. Clifford, Log-normal Rician probability density
function of optical scintillations in the turbulent atmosphere, J. Opt. Soc.
Amer., vol. 4, pp. 19231930, 1987.
[9] E. Jakeman, On the statistics of k-distributed noise, J. Phys. A, vol. 13,
pp. 3148, 1980.
[10] L. C. Andrews and R. L. Philips, Mathematical genesis of the I-K distribution for random optical fields, J. Opt. Soc. Amer. A, vol. 3, no. 11, pp.
19121919, 1986.
[11] L. C. Andrews, R. L. Phillips, and K. Shivamoggi, Relations of the
parameters of the I-K distribution for irradiance fluctuations to physical
parameters of the turbulence, Appl. Opt., vol. 27, no. 11, pp. 21502155,
1988.
[12] K. P. Peppas, A. N. Stassinakis, G. K. Topalis, H. E. Nistazakis, and
G. S. Tombras, Average capacity of optical wireless communication
systems over I-K atmospheric turbulence channels, J. Opt. Commun.
Netw., vol. 4, pp. 10261032, 2012.
[13] K. Kiasaleh, Performance of coherent DPSK free-space optical communication systems in k-distributed turbulence, IEEE Trans. Commun.,
vol. 54, no. 4, pp. 604607, Apr. 2006.
[14] M. Niu, J. Cheng, and J. F. Holzman, Error rate analysis of m-ary coherent
free-space optical communication systems with k-distributed turbulence,
IEEE Trans. Commun., vol. 59, no. 3, pp. 664668, Mar. 2011.
[15] S. M. Aghajanzadeh and M. Uysal, Diversity-multiplexing trade-off in
coherent free-space optical systems with multiple receivers, IEEE/OSA
J. Opt. Commun. Netw., vol. 2, no. 12, pp. 10871094, Dec. 2010.
[16] M. Niu, J. Cheng, and J. F. Holzman. (2014). Alamouti-type STBC for
atmospheric optical communication using coherent detection. IEEE Photon. J. [Online]. 6(1). Available: http://ieeexplore.ieee.org/xpls/icp.jsp?
arnumber=6725602
[17] E. Bayaki and R. Schober, Performance and design of coherent and
differential space-time coded FSO systems, J. Lightw. Technol., vol. 30,
no. 11, pp. 15691577, Jun. 2012.
[18] R. Mesleh, H. Elgala, and H. Haas, Optical spatial modulation, J. Opt.
Commun. Netw., vol. 3, no. 3, pp. 234244, Mar. 2011.
[19] R. Mesleh, H. Elgala, R. Mehmood, and H. Haas, Performance of optical
spatial modulation with transmitters-receivers alignment, IEEE Commun.
Lett., vol. 15, no. 1, pp. 7981, Jan. 2011.
[20] T. Fath and H. Haas, Performance comparison of MIMO techniques for
optical wireless communications in indoor environments, IEEE Trans.
Commun.,vol. 61, no. 2, pp. 733742, Feb. 2013.
[21] Y. A. Chau and S.-H. Yu, Space modulation on wireless fading channels,
in Proc. IEEE Veh. Technol. Conf. Fall, 2001, vol. 3, pp. 16681671.

11

12

JOURNAL OF LIGHTWAVE TECHNOLOGY

P. Takis Mathiopoulos (SM 94) received the Ph.D. degree in digital communications from the University of Ottawa, Ottawa, ON, Canada, in 1989.
From 1982 to 1986, he was with Raytheon Canada, Ltd., working in the areas
of air navigational and satellite communications. In 1988, he joined the Department of Electrical and Computer Engineering (ECE), University of British
Columbia (UBC), Vancouver, BC, Canada, where he was a Faculty Member as
a Professor from 2000 to 2003. From 2000 to 2014, he was with the Institute
for Space Applications and Remote Sensing (ISARS), National Observatory of
Athens (NOA), Athens, Greece, first as the Director and then as the Director of
research and established the Wireless Communications Research Group. As ISARS Director during 20002004, he led the Institute to a significant expansion
R&D growth, and international scientific recognition. For these achievements,
ISARS has been selected as a National Center of Excellence for the years 2005
2008. Since 2014, he has been an Adjunct Researcher at the Institute of Astronomy, Astrophysics, Space Applications and Remote Sensing, NOA. Since 2003,
he worked part time at the Department of Informatics and Telecommunications,
University of Athens, Athens, where since 2014, he has been a Professor of
digital communications. From 2008 to 2013, he was a Guest Professor with the
Southwest Jiaotong University, China. He is also appointed as a Guest Professor at the School of Information Engineering, Yangzhou University, Yangzhou,
China, for 20142015 under the Senior Foreign Expert Program of the Government of PR of China. Furthermore he is appointed by Keio University, Tokyo,
Japan, as a Visiting Professor at the Department of Information and Computer
Science for 20152016 under the Top Global University Project of the Ministry
of Education, Culture, Sports, Science and Technology Government of Japan.
For the last 25 years, he has been conducting research mainly on the physical
layer of digital communication systems for terrestrial and satellite applications,
including digital communications over fading and interference environments.
He coauthored a paper in GLOBECOM89 establishing for the first time in the
open technical literature the link between MLSE and multiple (or multisymbol)
differential detection for the AWGN and fading channels. He is also interested in
channel characterization and measurements, modulation and coding techniques,
synchronization, SIMO/MIMO, UWB, OFDM, software/cognitive radios, and
green communications. In addition, since 2010, he has been actively involved
with research activities in the fields of remote sensing, LiDAR systems, and photogrammetry. In these areas, he has coauthored more than 100 journal papers,
mainly published in various IEEE and IET journals, four book chapters, and
more than 120 conference papers. He has been the PI for more than 40 research
grants and has supervised the thesis of 11 Ph.D. and 23 Master students.
Dr. Mathiopoulos has been or currently serves on the editorial board of
several archival journals, including the IET Communications, and the IEEE
TRANSACTIONS ON COMMUNICATIONS from 1993 to 2005. He has regularly
acted as a consultant for various governmental and private organizations. Since
1993, he has served on a regular basis as a Scientific Advisor and a Technical
Expert for the European Commission (EC). In addition, since 2001, he has been
the Greek Representative to high-level committees in the EC and the European
Space Agency. He has been a Member of the TPC of more than 70 international
IEEE conferences, as well as the TPC Vice-Chair for the 2006-S IEEE VTC and
the 2008-F IEEE VTC, as well as the Cochair of the FITCE2011. He has delivered numerous invited presentations, including plenary and keynote lectures,
and has taught many short courses all over the world. As a Faculty Member
at the ECE, UBC, he was elected as an ASI Fellow and a Killam Research
Fellow. He received the two best paper awards for papers published in the 2nd
International Symposium on Communication, Control, and Signal Processing
in 2008), and the 3rd International Conference on Advances in Satellite and
Space Communications in 2011.

IE
E
Pr E
oo
f

Kostas P. Peppas was born in Athens, Greece, in 1975. He received the Diploma
degree in electrical and computer engineering and the Ph.D. degree in wireless
communications from the National Technical University of Athens, Athens, in
1997 and 2004, respectively. From 2004 to 2007, he was with the Department of
Computer Science, University of Peloponnese, Tripolis, Greece, and from 2008
to 2014 with the National Center for Scientific ResearchDemokritos, Institute
of Informatics and Telecommunications as a Researcher. In 2014, he joined
the Department of Telecommunication Science and Technology, University of
Peloponnese, where he is currently a Lecturer. His current research interests
include digital communications over fading channels, MIMO systems, wireless
and personal communication networks, and system level analysis and design.
He has authored more than 70 journal and conference papers.

JOURNAL OF LIGHTWAVE TECHNOLOGY

Free-Space Optical Communication With Spatial

Modulation and Coherent Detection Over H-K
Atmospheric Turbulence Channels
Kostas P. Peppas, Senior Member, IEEE, and P. Takis Mathiopoulos, Senior Member, IEEE

Index TermsAtmospheric turbulence, average bit error probability, coherent detection, free space optical communication systems, H-K distribution, multiple-input multiple-output (MIMO)
systems, optical spatial modulation.

I. INTRODUCTION

REE-SPACE optical (FSO) communication systems have

recently attracted great attention within the research community as well as for commercial use. FSO systems can provide ultra-high data rates (at the order of multiple gigabit per
second), immunity to electromagnetic interference, excellent
security and large unlicensed bandwidth, i.e., hundred or even
thousand times higher than radio-frequency (RF) systems, along
with low installation and operational cost [1].
The challenge in employing such systems is that FSO links are
highly vulnerable due to the detrimental effects of attenuation
under adverse weather conditions (e.g., fog), pointing errors
and atmospheric turbulence [1]. One method to improve the
reliability of the FSO link is to employ spatial diversity, i.e.
multiple-lasers and multiple-apertures to create a multiple-input
multiple-output (MIMO) optical channel. Because of its low
complexity, spatial diversity is a particularly attractive fading

mitigation technique and performance enhancements have been

extensively studied in many past research works in the field of
FSO communications (e.g., see [2][5]).
In order to evaluate the impact of atmospheric turbulence on
the performance of OSM, accurate models for the fading distribution are necessary. For example, the lognormal distribution
is often used to model weak turbulence conditions whereas the
negative exponential and the K-distribution are used to model
strong turbulence conditions [6]. Other more general statistical models have also been proposed to model scintillation over
all turbulence conditions, such as, for example, the Gamma
Gamma [7], the lognormal-Rice (or Beckmann) [8] the homodyned K distribution (H-K) [9] and the I-K [10][12] distributions. The validity of these models is based on the argument
that scintillation is a doubly stochastic random process modelling both small and large scale turbulence effects. Besides,
they agree well with measurement data and simulations for a
wide range of turbulence conditions.
In this paper, the H-K distribution is adopted to model
turbulence-induced fading [9]. The main reason for this choice
is the fact that this distribution is based on a very general scattering model which is valid over a wide range of atmospheric
conditions. It is also noted that the H-K distribution generalizes
existing models such as the K-distribution. The H-K distribution
models the field of the optical wave as the sum of a deterministic component and a random component, the intensity of which
follows the Rice (Nakagami-n) distribution. The average intensity of the random portion of the field is treated as a fluctuating
quantity [10]. It is important to underline that, to the best of our
knowledge, in the open technical literature there have been no
papers published analyzing and evaluating the performance of
FSO systems over such channels, because of the complicated
mathematical form of the required for the analysis probability
density functions (PDF).
Depending on their detection, FSO systems can be classified
into two main categories, namely non-coherent (direct detection) and coherent (heterodyne detection) systems [13][17].
Non-coherent FSO systems employing intensity modulation
and direct detection (IM/DD) have the optical power output
of a source varied in accordance with some characteristic of
the modulating signal. A direct detection receiver only responds
to the intensity of the received field and is oblivious to the
received signal phase. Coherent FSO systems encode the information bits directly onto the electric field of the optical beam.
At the receiver, a local oscillator (LO) is employed to extract
the information encoded on the optical carrier electric field. On

IE
E
Pr E
oo
f

AbstractThe use of optical spatial modulation (OSM) is proposed as a simple low-complexity means of achieving spatial diversity in coherent free-space optical (FSO) communication systems.
In doing so, this paper presents a generic analytical framework for
obtaining the average bit error probability (ABEP) of uncoded and
coded OSM with coherent detection in the presence of turbulenceinduced fading. Although the framework is general enough to accommodate any type of channel models based on turbulence scattering, the focus of the analysis is on the H-K distribution as such
model is very general and valid over a wide range of atmospheric
propagation conditions. Using this framework, it is shown that
OSM can offer comparable performance with conventional coherent FSO schemes employing spatial diversity at the transmitter
or the receiver only, while outperforming the latter in terms of
spectral efficiency and hardware complexity. Furthermore, various numerical performance evaluation results are also presented
and compared with equivalent results obtained by Monte Carlo
simulations which verify the accuracy of the derived analytical
expressions.

Manuscript received March 23, 2015; revised June 21, 2015; accepted July
14, 2015.
K. P. Peppas is with the Department of Informatics and Telecommunications,
University of Peloponnese, Tripoli 22100, Greece (e-mail: peppas@uop.gr).
P. T. Mathiopoulos is with the Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Athens 15784,
Greece (e-mail: mathio@di.uoa.gr).
Digital Object Identifier 10.1109/JLT.2015.2465385

0733-8724 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications standards/publications/rights/index.html for more information.

JOURNAL OF LIGHTWAVE TECHNOLOGY

TABLE I
LIST OF MATHEMATICAL NOTATIONS
j 2 = 1 denotes the imaginary unit
|z | denotes the magnitude of the complex number z
{z } denotes the real part of the complex number z
{z } denotes the real part of the complex number z
f (x )
f (x) = o[g (x)] as x x 0 if lim x x 0 g ( x ) = 0
 2F denotes the square Frobenius norm
() T denotes the matrix transpose
denotes convolution
E denotes expectation
f X () denotes the PDF of the RV X
F X () denotes the cumulative distribution function of the RV X
MX () denotes the MGF of the RV X
I a () is the modified Bessel function of the first kind and order a [26, eq. (8.431)]
K a () is the modified Bessel function of the second kind and order a [26, eq. (8.432)]

x 1
dt is the Gamma function [26, eq. (8.310/1)]
(x) = 0 exp(t)t


Q (x) = 21 x exp(t 2 /2)dt is the Gauss Q-function

W p , q () is the Whittaker function [26, eq. (9.220)]
Pr{} denotes the probability operator
denotes estimated value at the receiver side

specifically and within this new analytical framework, the main

novel research contributions of the paper are as follows:
1) New analytical expressions for the ABEP of coherent OSM
under turbulence conditions modeled by the H-K distribution are derived. When the transmitter is equipped with
two apertures the resulting analytical expressions are exact, whereas for an arbitrary number of transmit apertures
tight upperbounds can be obtained.
2) ABEP bounds for coded OSM systems are derived and
the performance enhancements when channel coding is
employed are presented and analyzed.
The ABEP of OSM is also compared to that of conventional
FSO schemes with transmit or receive diversity only, i.e., when
maximal ratio combining (MRC), selection combining (SC) or
Alamouti-type space-time block codes (STBC) are employed.
It is finally noted that the theoretical analysis is substantiated by
comparing the theoretical and equivalent simulated performance
evaluation results obtained by means of Monte Carlo techniques.
The paper is organized as follows. After this introduction,
Section II outlines the system and channel models. In Section III
analytical expressions for the ABEP of uncoded OSM systems
are presented. Asymptotic ABEP expressions are also derived,
wherefrom the diversity gain of coherent OSM can be readily
deduced. The performance of coded OSM systems is discussed
in Section IV. In Section V an overview of the conventional diversity techniques for coherent OSM is presented. In Section VI
the various performance evaluation results and their interpretations as well as comparisons are presented. Finally, concluding
remarks can be found in Section VII. Notations: A comprehensive list of all mathematical notations used in this paper can be
found in Table I.

IE
E
Pr E
oo
f

the one hand, coherent FSO systems can provide significant

performance enhancements due to spatial temporal selectivity
and heterodyne gain in comparison to direct detection systems.
Moreover, they are more versatile as any kind of amplitude,
frequency, or phase modulation can be employed. On the other
hand, the implementation of coherent receivers is more complex
as the LO field should be spatially and temporally coherent with
the received field. Throughout this work, coherent FSO systems
are only considered.
Recently, the so-called optical spatial modulation (OSM)
has emerged as a power- and bandwidth-efficient single-carrier
transmission technique for optical wireless communication systems [18][20]. This spatial diversity scheme, initially proposed
in [21] and further investigated in [22] and[23], employs a simple modulation mechanism that foresees to activate just one out
of several MIMO transmitters at any time instant and to use the
index of the activated transmitter as an additional dimension for
conveying implicit information. It has been shown that OSM
can increase the data rate by base two logarithm of the number
of transmit units [18]. Also, OSM can increase the data rate by
a factor of 2 and 4, as compared to onoff keying and pulse position modulation, respectively [18], [19]. It is underlined that
such performance gains are obtained with a significant reduction
in receiver complexity.
Because of the above mentioned advantages of OSM over
other more conventional transmission schemes and given the
wide applicability of FSO, it is important to investigate the
potential performance enhancements obtained by incorporating
OSM in FSO systems. However, in general, this research topic
has not been thoroughly investigated. In fact, only recently,
there have been papers published in the open technical literature dealing with performance analysis studies of FSO systems
employing spatial modulation and operating in the presence of
atmospheric turbulence, e.g., see [24] and [25]. Specifically, in
[24], the combination of subcarrier IM and spatial modulation
with receiver diversity was proposed to enhance the performance of IM/DD FSO systems. In [25], another IM/DD based
system FSO system which combines antenna shift keying with
joint pulse position and amplitude modulations was considered.
For this system, which was denoted as spatial pulse position
and amplitude modulation (SPPAM), the atmospheric turbulence channel was modeled as log-normal or GammaGamma
distributions and was evaluated, in terms of bounds, for uncoded and coded signals. Average bit error probability (ABEP)
performance evaluation results have shown that SPPAM offers a
compromise between spectral and power efficiencies as well as
a certain degree of robustness against atmospheric turbulence.
Despite these two papers which deal with non-coherent detection schemes, the potential enhancements of OSM on improving the performance of FSO systems with coherent detection
still remains an open research topic which, to the best of our
knowledge, has not been addressed so far in the open technical
literature.
Motivated by the above, in this paper we present for the first
time a generic analytical framework which can be used to accurately obtain the performance of outdoor OSM with coherent
detection in the presence of turbulence-induced fading. More

II. SYSTEM AND CHANNEL MODEL

In this section, a detailed description of the OSM FSO system model, i.e., transmitter, channel and receiver is provided.
Moreover, the H-K distribution is introduced and analytical

PEPPAS AND MATHIOPOULOS: FREE-SPACE OPTICAL COMMUNICATION WITH SPATIAL MODULATION AND COHERENT DETECTION

expressions for its parameters in terms of equivalent physical parameters of the turbulence phenomenon, such as the refractiveindex structure parameter, optical wave number, and propagation path length, are derived.
A. Preliminaries

B. Receiver Structure

Expanding (2) and ignoring the double-frequency terms that are

filtered out by the bandpass filter, the resulting photocurrent can
be expressed as
2 2
hm ,n + RPLO
in (t) = RPt Em

+2R Pt PLO Em hm ,n cos(IF t m ,n b m )

= iDC (t) + iAC (t).

(3)

2 2
In (3), iDC (t) = RPt Em
hm ,n + RPLO is the dc component
generated
by
the
signal
and
LO fields, respectively, iAC (t) =

2R Pt PLO cos(IF t m ,n b m ) is the ac component in

the received photocurrent which, unlike for direct detection,
contains information about the frequency and phase of the received signal. It is assumed that for coherent detection the IF
IF is nonzero, so that the signal power can be expressed as
2 2
hm ,n .
Ps = 2R2 Pt PLO Em
As in [13][17], we also consider that PLO  Ps and thus, the
dc photocurrent can be approximated as iDC (t) RPLO . The
photodetection process is impaired by shot noise with variance
2
= 2qe RPLO Be where Be is the electrical bandwidth of
shot,L
the photodetector. It is also noted that because of the large value
of RPLO the photocurrent due to thermal noise and the dark
current can be ignored [16].
Following [15] and[17], the sufficient statistics at the nth
coherent receiver can be expressed as

IE
E
Pr E
oo
f

Let us consider an M N MIMO FSO system with M transmit units (lasers) and N coherent receivers. It is assumed that
the receiving apertures are separated by more than a coherence
wavelength to ensure the independency of fading channels. The
basic principle of OSM modulation is as follows [18], [20]:
1) The transmitter encodes blocks of log2 (M ) data bits into
the index of a single transmit unit. Such a block of bits
is hereafter referred to as message and is denoted by
bm , m = 1, 2, . . . , M . It is assumed that the M messages are transmitted by the encoder with equal probability and that the related transmitted signal is denoted by
m = Em exp(jb ). During each time slot, only one
E
m
transmitter , where  = 1, 2, . . . , M is active for data
transmission. The information bits are modulated on the
electric field of an optical signal beam through an external
modulator. During each particular time slot, the remaining
transmit lasers are kept silent, i.e., they do not transmit.
2) At the receiver, the incoming optical field is mixed with a
LO field and the combined wave is first converted by the
photodetector to an electrical one. A bandpass filter is then
employed to extract the intermediate frequency (IF) component of the total output current. Finally, a N -hypothesis
detection problem is solved to retrieve the active transmit
unit index, which results in the estimation of the unique
sequence of bits emitted by the transmitter.

The received electric field at the aperture plane of the nth

receiver after mixing with a LO beam, can be expressed as [16],
[17]

2Pt Z0 Em hm ,n cos(0 t + m ,n + b m )
en (t) =

+ 2PLO Z0 cos(LO t).
(1)

In the above equation, Pt is the transmit laser power, Z0 is the

free space impedance, hm ,n and m ,n denote the magnitude and
the phase of the complex channel coefficient between the mth
transmit and the nth receive aperture, respectively. Furthermore,
PLO denotes the power of the LO, LO = 0 + IF where 0
and IF are the carrier and the intermediate radian frequencies,
respectively.
The output current of the nth photodetector can be mathematically expressed as [16], [17]
R
in (t) =
[en (t)]2
Z0

yn =

hm ,n Em exp[j(m ,n + b m )] + zn

where = RPt /(qe Be ) is the average signal-to-noise ratio

(SNR) and zn is the noise at the nth receiver. Assuming that
the LO power is large and the receiver noise is dominated by
the LO related noise terms, the additive white Gaussian noise
model can be employed as an accurate approximation of the
Poisson photon-counting detection model [16], [17]. Thus, zn
can be modeled as a zero-mean unit variance complex Gaussian
random variable (RV) [17].
Similar to [27], it is assumed that the receiver has knowledge
of the actual fading gains and that the fading process remains
constant over the bit interval and changes from bit to bit in an
independent manner. At the receiver, the optimal spatial modulation detector estimates the active transmitter index, , at a
given time slot according to [28]
 = argmaxpy (y|x, H)


= argmin
(2)

where R = qe /(h0 ) is the responsivity of the photodetector

with qe = 1.6 1019 Cb is the charge of an electron, h =
6.6 1034 J s is the Planck constant, is the photodetector
efficiency, and 0 = 0 /(2) is the optical center frequency.

(4)



 h x 2F 2 yT h x

(5)

where
- x is an M -dimensional vector with elements corresponding
to the electrical field Em exp(jb m ) transmitted over the
optical MIMO channel;

JOURNAL OF LIGHTWAVE TECHNOLOGY

- H(t) is an N M optical MIMO channel defined as

H(t) = [h1 , h2 , . . . , hM ]

h1 1 (t) exp(j1 1 )

h (t) exp(j )
21
21


..

hN 1 (t) exp(jN 1 )

...
...

..

...

h1 M (t) exp(j1 M )

h2 M (t) exp(j2 M )

..

hN M (t) exp(jN M )
(6)

- z is the N -dimensional noise vector;

- py (y|x, H) is the PDF of y conditioned on the transmitted
vector x and the channel H.

form. Specifically, for Aij = 0 (9) reduces to the K-distribution

whereas for = 1, (9) reduces to a special case of the I-K
distribution [10, Eq. (10)].
The th normalized moment of Iij is given by [10, Eq. (23)]
as
 

E{Iij }
(ij + k) (ij ij )
!
=
(1 + )
k
E{Iij }
ij
(ij )
!
ij
k =0
(10)
where ij  A2ij /b0 i j is the coherence parameter, defined as
the power ratio of mean intensities of the constant-amplitude
component and random component of the field in (7) [10], [11].
Using (10), the scintillation index can be readily calculated as
I2i j 

C. Channel Model

(11)

Under the assumption of plane wave propagation, I2i j can be

directly related to atmospheric conditions as [11, Eq. (7), Eq.
(9)]
 2
1 (1 + 0.512 ), 1  1
2
I i j
(12)
4/5
1 + 0.86/1 , 1  1

IE
E
Pr E
oo
f

A discrete scattering channel model is considered, where the

radiation field of an optical wave at a particular point is assumed
to be composed of a number of scattered components that have
propagated through different paths. Under the Ricean assump ij (t) between the
tion [10], the complex channel path gains h
ith transmitter and the jth photodetector can be expressed as
ij (t) = hij (t) exp(jt) where is the radian frequency of the
h
optical wave and

E{I2ij }
ij + 2ij ij + 2
1=
.
E{Iij }2
ij (1 + ij )2

hij (t) = {hij (t)} + j{hij (t)}

= Aij exp[jij (t)] + Rij (t) exp[jij (t)]

(7)

where the term Aij exp[jij (t)] is a deterministic component

and Rij (t) exp[jij (t)] is a circular complex Gaussian RV.
Hence, the amplitude Rij is Rayleigh distributed with scale
2
= bij /2 [10, Eq. (13)] and the phase ij is uniparameter ij
formly distributed over [0, 2). Under the assumption of a doubly stochastic scintillation model [10], the effect of random
fluctuations in the turbulence parameters is modeled by allowing random variations in the parameter bij of the Rayleigh component. Following [10], it is further assumed that bij follows a
Gamma distribution with PDF given by


 i j i j 1

ij
b
b
exp
fb i j (b) =
(8)
b0
(ij )
b0 i j
where is the shaping parameter which represents the effective
number of scatters and b0 i j = E{bij }. Then, the PDF of the
irradiance Iij = |hij (t)|2 , fI i j (I), can be expressed as [10, Eq.
(8)]


ij /b0 i j i j
fI i j (I) =
(ij )



ij b I + Aij 2

b i j 2 exp

b0 i j
b
0


2Aij I
db
(9)
I0
b
which is actually the integral representation of the H-K distribution [10]. It is noted that, with the exception of the special
cases Aij = 0 or = 1, fI i j (I) cannot be expressed in closed

11/6

is the Rytov variance, k =

where 12 = 1.23Cn2 i j k 7/6 Lij
2/ is the optical wave number with being the wavelength, Lij is the link distance and Cn i j denotes the index of
refraction structure parameter. For FSO links which operate
near the ground plane, Cn2 i j 1.7 1014 m2/3 and 8.4
1015 m2/3 for the daytime and night time periods, respectively [29]. Moreover, 1  1 and 1  1 correspond to weak
and strong turbulence conditions, respectively.
Using (12), the parameters of the H-K distribution, and ,
can be directly related to physical parameters of the turbulence
by following a similar line of arguments as in [11], where similar
results have been derived for the I-K distribution. In particular,
on the one hand, weak turbulence conditions are characterized
in the H-K distribution by large values of ij . In this case the
scintillation index given by (11) can be approximated as
I2i j

2
,
ij

with ij  1.

(13)

On the other hand assuming strong turbulence conditions, in

which case ij goes towards zero, (11) can be approximated as
I2i j 1 +

2
,
ij

with ij  1.

(14)

By comparing (13) and (14) with the first and second branches
of (12), respectively, ij and ij can be obtained as
4/5

ij = 2.331 i j
ij =

2
.
12i j (1 + 0.512i j )

(15)
(16)

To the best of our knowledge, the relationship of ij and ij with

1 i j given by (15) and (16) has not been previously published.

PEPPAS AND MATHIOPOULOS: FREE-SPACE OPTICAL COMMUNICATION WITH SPATIAL MODULATION AND COHERENT DETECTION

III. PERFORMANCE ANALYSIS OF UNCODED OSM

In this section, by employing the well-known moment generating function (MGF)-based approach for the performance
analysis of digital communications over fading channels [30],
analytical expressions for the ABEP of uncoded OSM systems
will be derived. In addition, expressions for the diversity and
coding gains of OSM systems are also presented, thus providing useful insight as to how these parameters affect the overall
system performance.
A. Preliminaries

n has uncorrelated components too and its squared envelope,

|n |2 , is characterized by a non-central chi-square PDF

  
An x
1
x + A2n
f| n |2 (x|bn ) =
exp
I0
(21)
2bn
2bn
bn
where An = |A2,n exp(j2,n ) A1,n exp(j1,n )|. Assuming
that bn follows the Gamma with parameters n and b0,n , the
PDF of |n |2 is obtained by averaging (21) with respect to bn ,
i.e.,
f| n |2 (x) =

For M = 2, the conditional bit error probability (CBEP) of

OSM systems when no turbulence induced fading is considered
can be obtained in closed form as [27]



2
 h1 h2 F .
(17)
PE (h1 , h2 ) = Q
4

(n /b0,n ) n
2(n )



n bn
x + A2n
n 2
bn
exp

b0,n
2bn
0
 
An x
I0
(22)
dbn .
bn

The squared Frobenius norm in (17) can be expressed as

|h1,n h2,n |2

(18)

n =0

where hi,n is the nth element of hi , i {1, 2}. When M > 2

transmitters are considered, a tight upper bound for the CBEP
of the above system can be obtained as [18, Eq. (7)]
M 1
PE (H)
log2 (M )

M


M


As was pointed out in [10], f| n |2 (x) cant be expressed in

closed form. Nevertheless, for the special case of n = 1, i.e.,
when one scatterer per branch is considered, and by employing
[10, Eq. (10)], this integral can be evaluated in closed form as

IE
E
Pr E
oo
f

 h1 h2 2F =

N


f| n |2 (x)

 

,
K
2
A
/b
2x/b
I
0
n
0,n
0
0,n
b
0,n


=



I0
2An /b0,n K0
2x/b0,n ,
b0,n

Nb (m1 , m2 )PEP(m1 m2 ) (19)

m 1 =1 m 2 = m 1 =1

where PEP(m1 m2 ) denotes the pairwise error probability

(PEP) related to the pair of transmitters m1 and m2 , where m1
and m2 1, 2, . . . , M , and Nb (m1 , m2 ) is the number of bits in
error which have occurred when the receiver decides incorrectly
that m2 instead of m1 has been active. The PEP(m1 m2 )
can be evaluated as [18, Eq. (8)]



2
 hm 1 hm 2 F .
PEP(m1 m2 ) = Q
(20)
4
B. MGF-Based Approach

When atmospheric turbulence is included in the analysis the

conditional error probabilities in (17) and (19) need to be averaged over the elements of the channel matrix H in order to
evaluate the ABEP. Without loss of generality, let us consider the
case of a 2 N MIMO system. Since hi,n are complex Gaussian RVs, the difference n = h1,n h2,n is also a complex
Gaussian RV having mean equal to the difference of the means
of hi,n and variance equal to the sum of variances of hi,n . In order to deduce a closed form expression for the ABEP, it is further
assumed that hi,n have uncorrelated real and imaginary components with the same variance n2 = bn /2. It is noted that this
assumption has been experimentally verified in the past for link
distances of the order of kilometer and for aperture separation
distances of the order of centimeter [31], [32]. Consequently,

x < A2n

x > A2n .

(23)

Moreover, for the special case where h1,n and h2,n have identical mean value, i.e., when An = 0, (22) yields the well-known
K-distribution with PDF given by

( n 1)/2
n x
(1 n )/2
f| n |2 (x) = 2
(n )
b0,n


2n x
K n 1
.
(24)
b0,n
By employing the MGF-based approach for the performance
analysis of digital communications over fading channels, the
average PEP (APEP) can be obtained as

N
 
1 /2  
M| n |2
d.
(25)
APEP =
0
8 sin2
n =1

Moreover, using the tight approximation for the Gaussian Qfunction presented in [33, Eq. (14)] (i.e., Q(x) 1/12 exp
(x2 ) + 1/4 exp(2x2 /3)), an expression accurately approximating APEP can be deduced as
N
N


1 
1 
M| n |2
M| n |2
+
.
12 n =1
8
4 n =1
6
(26)
In the following, analytical expressions for the MGF of |n |2
will be obtained.

APEP

JOURNAL OF LIGHTWAVE TECHNOLOGY

Proposition 1: An integral representation for the MGF of

|n |2 can be deduced as
M| n |2 (s) =

tj and wj are given as


$
2j 1

+
2J
4


2 sin 2j2J1
%
&.


wj =
4J cos2 4 cos 2j2J1 + 4
#

tj = tan

(n /b0,n )
(n )


 n 1
An s
n b
b
exp

db.
2bs + 1
2bs + 1 b0,n
0
(27)

Proof: By employing the definition of the MGF, M| n |2 (s)

can be obtained as

exp(sx)f| n |2 (x)dx
M| n |2 (s) =

By changing the order of integration, the above equation can be

expressed as
n

M| n |2 (s) =

(n /b0 , n )
n b
b n 2 exp
2(n )
b0 , n
0

!
   "

An x
x + A2n
exp sx
I0
dx db.
2b
b
0

(29)

The inner integral, i.e., with respect to x can be evaluated by

employing [34, Eq. (3.15.2.2)] as
  


An x
x + A2n
I0
dx
exp sx
2b
b
0
#
$
1
2b
=
exp
.
(30)
2sb + 1
2An b(2sb + 1)

Substituting (30) into (29) and after some straightforward mathematical manipulations, (27) is readily deduced thus completing
the proof.

The integral in (27) can be accurately approximated by employing a GaussChebyshev quadrature (GCQ) technique as
[35]
M| n |2 (s)

(n /b0,n )
(n )

(32a)
(32b)


M| n |2 (s) =

n
2sb0,n

 n /2
exp

W n /2,( n 1)/2

n
4sb0,n
n
2sb0,n



.

(33)

This result can be readily deduced by employing the integral

representation of the Whittaker W -function given in [26, Eq.
(9.222)]. It is noted that (33) is in agreement with a previously
known result, namely the analytical expression for the MGF of
the K-distribution [37, Eq. (4)].

IE
E
Pr E
oo
f

(n /b0,n ) n
2(n )


 
n b x + A2n
exp sx

b0,n
2b
0
0
 
An x
b n 2 dbdx.
(28)
I0
b

For the special case of An = 0, (27) can be evaluated in closed

form, as it will be shown next.
Corrolary 1: For the special case of A = 0 the MGF of
|n |2 can be deduced in closed form as

cos
4



An s
n tj
tj n 1
exp

wj

2tj s + 1
2tj s + 1
b0,n
j =0
J


(31)
where J is the number of integration points, tj are the abscissas
and wj the corresponding weights. In [36, eqs. (22) and (23)],

C. Diversity Gain Analysis

The diversity gain of the considered OSM MIMO system can

be obtained by using the approach presented in [38]. In particular, a generic analytical expression, which becomes asymptotically tight at high SNR values, will be derived for the APEP
appearing in (25), as follows:
Proposition 2: For high SNR values, (25) can be approximated by
"
! N

 N
N 1

N + 12
1 2
c
(34)
APEP
4
(N + 1) n =1

where

cn =

( n 1)/2

(n /b0,n )( n +1)/2
(n )


2An n
.
K n 1
b0,n
An
2

(35)

Proof: According to [38, Proposition 3], the asymptotic error

performance of the OSM system depends on the behavior of
M| n |2 (s), as s . To determine an analytical asymptotic
expression for APEP a Taylor series expansion is employed to
approximate M| n |2 (s) as
|M| n |2 (s)| = cn |s|d n + o(|s|d n ),

(36)

where cn and dn are parameters that determine the diversity and

coding gains of the nth diversity branch, respectively. Observe
s
s

that since As/(2sb

+ 1) A/(2b)
and 1/(2sb + 1)

PEPPAS AND MATHIOPOULOS: FREE-SPACE OPTICAL COMMUNICATION WITH SPATIAL MODULATION AND COHERENT DETECTION

1/(2bs), (27) yields

M| n |2 (s)

(n /b0,n ) n
2s(n )



An
n b
n 2

b
exp

db. (37)
2b
b0,n
0

By employing [34, Eq. (2.2.2.1)], (37) can be solved in closed

form yielding
( n 1)/2

An
(n /b0,n )( n +1)/2
M| n |2 (s)
2
s(n )


2An n
.
(38)
K n 1
b0,n

A union bound on the ABEP of a coded communication

system can be obtained as [30]

1
P (X)
q(X, X )PEP(X, X )
(41)
Pub
n

X

X= X

where P (X) is the probability that the coded sequence X is

transmitted, q(X, X ) is the number of information bit errors
in choosing another coded sequence X instead of X n is the
number of information bits per transmission and PEP(X, X )
is the PEP, i.e., the probability of selecting X when X was
actually transmitted.
By employing [30, p. 510], (41) can be efficiently evaluated
as
,
$
 /2 #
,

1

T [D(), N ],,
P (X)
Pub
(42)
n
N
0

N =1

where T [D(), N ] is the transfer function of the employed convolutional code, N is an indicator variable taking into account
the number of the erroneous bits and D() depends on the underlying PEP expression. Furthermore, assuming that uniform
error probability (UEP) codes are considered and taking into
account the symmetry property this code family exhibits, thus
making the distance structure of a UEP code independent of the
transmitted sequence, (42) can be further simplified as [30]
,
#
$

,
1 /2 1
T [D(), N ],,
.
(43)
Pub
0
n N
N =1

IE
E
Pr E
oo
f

By comparing (38) and (36) it is readily deduced that dn = 1

and cn is given by (35). Thus, by substituting (36) into (25),
the asymptotic PEP expression can be obtained as in (34) which
concludes the proof.

From (34) it is clear that the diversity gain achieved by the
considered system is equal to N . It is also evident that the diversity gain depends only on the number of the receive apertures
and is independent of the fading severity. This finding is in agreement with other equivalent observations reported in [27], [39]
and[40], which have been obtained for the case of RF MIMO
wireless systems.
It is noted that for the special case An = 0, i.e., when |n |2
follows the K-distribution, by employing the asymptotic result

x0

Kt (x) ((t)/2) (2/x)t [35], cn can be further simplified as

n
.
(39)
cn =
2b0,n (n 1)

For M = 2, using (17), (18) and Craigs formula for the Q /2

function, i.e., Q(x) = 1/ 0 exp(x2 /2 sin2 )d, D()
can be expressed as
D() =

N


n =1

IV. PERFORMANCE ANALYSIS OF CODED OSM OVER

TURBULENCE CHANNELS

When coded OSM is employed, the input signal s(t) is first

encoded by a convolutional encoder. The encoded data are interleaved by a random block interleaver and transmitted through the
optical wireless channels using spatial modulation. It is assumed
that perfect interleaving at the transmitter and de-interleaving
at the receiver is used. Considering maximum likelihood soft
decision decoding, the log likelihood ratios (LLRs) for the ith
constellation bit when the th transmitting antenna is active are
computed as [18, Eq. (6)]
Pr{i = 1|y}
LLR = log
Pr{i = 0|y}


+
2
i exp  y hs  /N0
L
1


= log +
2
i exp  y hs  /N0
L

M| n |2


8 sin2

(44)

where M| n |2 can be obtained from (27). When M > 2, by

employing [18, Eq. (13)], and using a similar line of arguments
as for the case of M = 2, D() can be written as
M


M


m 1 =1 m 2 = m 1 =1

M| m 1 , m 2 |2


8 sin2

(45)

where |m 1 ,m 2 |2 = hm 1 hm 2 2 . The last MGF can be analytically computed with the help of (27) in a straight-forward
way.
V. DIVERSITY TECHNIQUES FOR COHERENT FSO SYSTEMS

(40)

where L {1 : M } is the set of spatial constellation points, Li1

and Li0 are subsets from L containing the transmitter indices
having 1 and 0 at the ith bit, respectively. The resulting data
are finally decoded by a Viterbi decoder.

Diversity at the transmitter and/or the receiver is a commonly

used technique to mitigate the deleterious impact of atmospheric
turbulence on coherent FSO system performance. In this section, firstly the most popular diversity techniques for coherent
FSO systems will be presented [41]. Such diversity schemes
will be also considered for the FSO communication systems investigated in this paper as follows: Receive diversity by employing MRC or SC and transmit diversity based on the Alamouti
scheme.

JOURNAL OF LIGHTWAVE TECHNOLOGY

The instantaneous SNR at the output of the coherent MRC

receiver assuming equal average SNR per receiving aperture,
can be expressed as [41, Eq. (11)]
M RC =

N


In

(46)

n =1

where In denotes the optical signal irradiance at the nth branch.

The ABEP of coherent MRC receivers with differential phase
shift keying (DPSK) can be deduced as [13, Eq. (14)]
PE =

N
1 
MI n ()
2 n =1

(47)

whereas for binary phase shift keying (BPSK) as [41, Eq. (18)]
1
PE =


0

N
/2 
n =1

MI n

 
d.
sin2

(48)

The instantaneous SNR at the output of the coherent SC

receiver is given by [41, Eq. (15)]
(49)

IE
E
Pr E
oo
f

SC = max{I1 , I2 }.

Fig. 1. ABEP of uncoded OSM for 2 N MIMO H-K turbulent channels as

a function of the average SNR, , for various number of receiving apertures,
N . Simulation Parameters: A 1 , n = 2, A 2 , n = 1, 1 , n = /3, 2 , n = /4,
n = 2, b0 , n = 2.

For the SC case, an analytical expression for the ABEP is much

more difficult to be deduced and, thus for this case the ABEP will
be evaluated exclusively by means of Monte Carlo simulations.
In [16], transmit diversity techniques, based on the Alamouti scheme, were explored for coherent OWC applications by
employing STBC. The instantaneous SNR at the input of the
demodulator of the optical receiver has a similar form as (46)
[16], and therefore the ABEP of BPSK can be readily evaluated
by employing (48).
VI. PERFORMANCE EVALUATION RESULTS AND DISCUSSION

In this section the various performance evaluation results

which have been obtained by numerically evaluating the mathematical expressions presented in Sections III and IV for uncoded and coded OSM systems operating over H-K turbulent
channels will be presented. In particular, for uncoded OSM systems, ABEP vs. SNR for 2 N OSM systems (obtained using
(26) with (27), and (34)) the results are presented in Figs. 14.
For the uncoded schemes, in order to validate the accuracy of
the previously mentioned expressions, comparisons with complementary Monte Carlo simulated performance results are also
included in these figures. As far as the performance of coded
OSM systems is concerned, ABEP upper bounds vs. SNR have
been obtained using (43) with (27) and the results are illustrated
in Fig. 5. Table VI summarizes the different values of the system
parameters used to obtain the simulated performance evaluation
results. Note that in obtaining the various performance results
a system with M = 2 transmitting apertures was considered.
The motivation behind this choice is the inherent design complexities and cost of adopting multiple transmitters in current
state-of-the-art coherent optical wireless links [16], as well as
the fact that the proposed framework yields exact results for
M = 2 transmitters.

Fig. 2. ABEP of uncoded OSM for 2 2 and 2 4 MIMO H-K turbulent

channels as a function of the average SNR, , for various values of link distances,
L. Simulation Parameters: = 1550 nm, C n2 = 1.7 10 1 4 m 2 / 3 , 1 , n =
/3, 2 , n = /4, b0 , n = 1.

In order to demonstrate the impact of the number of the receiving apertures on the system performance, Fig. 1, presents
the ABEP performance as a function of the average SNR, ,
of 2 N MIMO OSM systems with N {1, 2, 3, 4}. Independent and identically distributed branches are considered
with A1,n = 2, A2,n = 1, 1,n = /3, 2,n = /4, n = 2,
b0,n = 2. The obtained results clearly indicate that the ABEP

PEPPAS AND MATHIOPOULOS: FREE-SPACE OPTICAL COMMUNICATION WITH SPATIAL MODULATION AND COHERENT DETECTION

Fig. 5. ABEP upper bounds of convolutional coded OSM for 2 2 and 2

1 H-K turbulent channels as a function of the average SNR, , for various
values of link distances, L. Simulation Parameters: = 1550 nm, C n2 = 1.7
10 1 4 m 2 / 3 , 1 , n = /3, 2 , n = /4, b0 , n = 1.

IE
E
Pr E
oo
f

Fig. 3. ABEP of uncoded OSM for 2 3 MIMO H-K turbulent channels as

a function of the average SNR, , for weak (C n2 = 10 1 5 m 2 / 3 ) and strong
C n2 = 3 10 1 4 m 2 / 3 turbulence and for various values of b0 , n . Simulation
Parameters: = 1550 nm, 1 , n = /3, 2 , n = /4.

TABLE II
SYSTEM PARAMETERS FOR SIMULATION

Fig. 4. ABEP Comparison of 2 2 OSM with 1 2 coherent MRC systems

employing DPSK, as a function of the average SNR, , for various values
of A 1 , n . Simulation Parameters: A 2 , n = 0, 1 , n = 0, 2 , n = 0, n = 1.5,
b0 , n = 1.5.

curves, obtained using (26), are in close agreement with the

performance obtained via simulations, thus verifying the correctness of the proposed analysis. Moreover, it is evident that
the asymptotic ABEP curves accurately predict the diversity
gain achieved by the considered system for all test cases.
In Fig. 2, the dependence on the link distance, L, of the ABEP
of a 2 N MIMO OSM system is illustrated. The considered

Parameter

Value

Operating wavelength, (nm)

Refractive index parameter, C n2 (m 2 / 3 )
Link distance, L (m)
1 , n
2 , n
A1,n
A2,n
b0 , n
Number of Transmitting apertures, M
Number of Receiving apertures, N

1500
{3, 1.7, 0.1} 10 1 4 ,
{500, 1000, 1500, 2000}
/3
/4
{0, 1, 2, 3}
{0, 1}
{0.25, 0.5, 1, 1.5, 2}
2
{1, 2, 3, 4}

system is again equipped with either N = 2 or N = 4 receiving

apertures and identically distributed branches are assumed. The
parameters of the H-K distribution are calculated from (15)
and (16) assuming plane wave propagation. Following [42], it
is further assumed that the operating wavelength is = 1550
nm and Cn2 = 1.7 1014 m2/3 , which, as it was pointed out
in Section II, is a typical value of refractive index for FSO
links operating near the ground plane during daytime [29]. As
expected, the error performance deteriorates as L increases from
L = 500 m to L = 1500 m. Moreover, it is evident that an
increase in L from 500 to 1000 m leading to a more severe
performance deterioration than in the case where L increases
from 1000 to 1500 m. For all cases considered, the analytical
results obtained using (26) are compared with the equivalent
results obtained by means of Monte-Carlo computer simulations
and again they match very well.
In Fig. 3 the impact of the turbulence strength as well as
of the parameter b0 on system performance is illustrated. In

10

JOURNAL OF LIGHTWAVE TECHNOLOGY

in [18], OSM offers increased spectral efficiency by a factor

log2 (M ). Moreover, as only one transmitting aperture is activated at any bit duration, OSM has a lower decoding complexity
as compared to conventional MRC and Alamouti schemes.
In Fig. 5, upper bounds on the ABEP of convolutional coded
2 1 and 2 1 OSM systems are depicted, assuming similar
propagation conditions to those considered in Fig. 2. Considering a convolutional code with rate 1/3 and constraint length of
3, its transfer function is given as [44, Eq. (8.2.6)]
T [D(), N ] =

D()6 N
.
1 2N D()2

(50)

Substituting (50) into (43), a union bound on the ABEP can be

obtained as
 /2
D()6
1

Pub
d.
(51)
log2 (M ) 0
(1 2D()2 )2
The performance results of Fig. 5 clearly show that, as expected,
the incorporation of convolutional coding significantly enhances
the performance of OSM systems, even when a small number
of receive apertures is employed.

IE
E
Pr E
oo
f

particular, Fig. 3 presents the ABEP performance as a function

of the average SNR, , of 2 N MIMO OSM systems with
N = 3 receiving apertures, link distance L = 2000 m, and various values of b0 . Two different values of the refractive index
parameter Cn2 are considered, i.e., Cn2 = 3 1014 m2/3 and
Cn2 = 1015 m2/3 to account for strong and weak turbulence,
respectively [29], [43]. The obtained results clearly show that
the error performance improves as b0 increases and/or Cn2 decreases. Moreover, the analytical results obtained using (26)
agree well with the equivalent results obtained by means of
MonteCarlo computer simulations.
Next we compare the proposed OSM system with two alternative coherent FSO systems that can provide performance
enhancements by means of transmit, i.e., multiple-input singleoutput (MISO) or receive diversity i.e. single-input multipleoutput (SIMO). It is noted that for similar aperture configurations, a fair comparison between coherent and IM/DD systems
seems difficult as the same received laser power leads to different
SNRs for each of these schemes [17]. On the other hand, in order
to perform a fair comparison between OSM and the alternative
MISO or SIMO systems under the same propagation channel
conditions, the aperture configuration of the FSO systems under
comparison should be carefully selected. Specifically, because
of the fact that the diversity gain of OSM equals to only the
number of the receive apertures, i.e., no transmit diversity gain
can be achieved, the number of transmit or receive apertures of
the alternative systems must be hence selected to be equal to the
number of receive apertures of the OSM system. To this end and
for a fair comparison in our paper a 2 2 OSM system is compared with the following two alternative FSO communication
systems which also employ coherent detection:
1) A 1 2 heterodyne FSO communication system which
employs DPSK [13] and MRC or SC;
2) A 2 1 coherent FSO system employing the Alamouti
scheme [16] and BPSK.
In order to simplify the underlying mathematical analysis,
it is assumed that the PDF of In is given by (9) with the parameters An being all zero, i.e., the PDF considered here is the
K-distribution. Thus, MI n () can be readily obtained in closed
form from (33) by replacing b0,n with b0,n /2. In Fig. 4, the
ABEP of 2 2 MIMO OSM links is compared with the ABEP
of 1 2 coherent FSO systems with DPSK considering identically distributed links. In order to compare these systems under
the same propagation conditions, it is assumed that n = 1, 5,
b0,n = 1.5, A2,n = 0 and A1,n = {0, 1, 2, 3}. As it can be observed, when either MRC or SC are employed, although coherent DPSK performs worse than the OSM for values of A1,n up
to approximately 1, it outperforms OSM at lower values of A1,n .
Moreover, although the OSM outperforms the Alamouti scheme
for A1,n = 2 and 3, it performs similarly for high SNR values
when A1,n = 1. It is noted that for A1,n = 1 and lower values
of A1,n the Alamouti scheme yields the best performance of the
considered schemes. However, when more transmit apertures
are employed, this advantage is compensated by the superior
spectral efficiency of OSM and its lower hardware complexity as compared to coherent MRC. Specifically, as pointed out

VII. CONCLUSION

In this paper, the use of spatial modulation technique for

coherent FSO communication systems has been proposed. We
have provided a comprehensive analytical framework for error
performance analysis which is valid for a great deal of atmospheric turbulence scattering channel models but focusing on
the H-K distribution. The proposed framework has revealed important information about the performance of OSM over such
turbulent channels, including the effect of fading severity and
the achievable diversity gain. In particular, it has been shown
that the diversity gain depends on the number of receive apertures only while channel parameters affect the coding gain only.
Upper bounds for the ABEP performance of coded OSM systems have also been derived, demonstrating that coding techniques can greatly enhance the performance of OSM. Extensive
computer simulation performance evaluation results have been
also obtained which have verified the accuracy of the analytical approach. It was shown that OSM can provide significant
performance enhancements in the presence of H-K atmospheric
turbulence. The improvements are comparable to the ones offered by conventional coherent systems with spatial diversity,
while outperforming the latter in terms of spectral efficiency
and hardware complexity. Specifically, OSM offers increased
spectral efficiency by a factor log2 (M ) when multiple transmitting apertures are employed while it completely avoids multiple
communication chains in a similar fashion as in RF systems.
Besides, under specific propagation conditions, OSM can yield
better performance than conventional SIMO systems employing
MRC or SC. The proposed framework, apart from enabling the
computation of accurate performance evaluation results in an
efficient manner, it also provides a useful tool for understanding the performance trend, important properties and tradeoffs of

PEPPAS AND MATHIOPOULOS: FREE-SPACE OPTICAL COMMUNICATION WITH SPATIAL MODULATION AND COHERENT DETECTION

outdoor OSM operating in the presence of atmospheric turbulence.

REFERENCES

[22] R. Y. Mesleh, H. Haas, S. Sinanovic, C. Ahn, and S. Yun, Spatial modulation, IEEE Trans. Veh. Technol., vol. 57, no. 4, pp. 22282241, Jul.
2008.
[23] R. Mesleh, M. D. Renzo, H. Haas, and P. M. Grant, Trellis coded spatial
modulation, IEEE Trans. Wireless Commun., vol. 9, no. 7, pp. 23492361,
Jul. 2010.
[24] S.-H. Hwang and Y. Cheng, SIM/SM-aided free-space optical communication with receiver diversity, J. Lightw. Technol., vol. 32, no. 14,
pp. 24432450, Jul. 2014.
[25] T. Ozbilgin and M. Koca, Optical spatial modulation over atmospheric
turbulence channels, J. Lightw. Technol., vol. 33, no. 11, pp. 23132323,
Jun. 2015.
[26] I. Gradshteyn and I. M. Ryzhik, Tables of Integrals, Series, and Products,
6th ed. New York, NY, USA: Academic, 2000.
[27] M. D. Renzo and H. Haas, Space shift keying (SSK) MIMO over correlated Rician fading channels: Performance analysis and a new method for
transmit-diversity, IEEE Trans. Commun., vol. 59, no. 1, pp. 116129,
Jan. 2011.
[28] J. Jeganathan, A. Ghrayeb, and L. Szczecinski, Spatial modulation: Optimal detection and performance analysis, IEEE Commun. Lett., vol. 12,
no. 8, pp. 545547, Aug. 2008.
[29] J. W. Goodman, Statistical Optics. New York, NY, USA: Wiley, 1985.
[30] M. K. Simon and M. S. Alouini, Digital Communication over Fading
Channels, 2nd ed. New York, NY, USA: Wiley, 2005.
[31] E. Lee and V. Chan, Part 1: Optical communication over the clear turbulent atmospheric channel using diversity, IEEE J. Sel. Areas Commun.,
vol. 22, no. 9, pp. 18961906, Nov. 2004.
[32] N. Letzepis, I. Holland, and W. Cowley, The Gaussian free space optical MIMO channel with Qary pulse position modulation, IEEE Trans.
Wireless Commun., vol. 7, no. 5, pp. 17441753, May 2008.
[33] M. Chiani, D. Dardari, and M. K. Simon, New exponential bounds and
approximations for the computation of error probability in fading channels, IEEE Trans. Wireless Commun., vol. 2, no. 4, pp. 840845, Jul.
2003.
[34] A. P. Prudnikov, Y. A. Brychkov, and O. I. Marichev, Integrals and Series
Volume 4: Direct Laplace Transforms, 1st ed. Boca Raton, FL, USA: CRC,
1992.
[35] M. Abramovitz and I. Stegun, Handbook of Mathematical Functions with
Formulas, Graphs, and Mathematical Tables. New York, NY, USA: Dover,
1964.
[36] F. Yilmaz and M.-S. Alouini, An MGF-based capacity analysis of equal
gain combining over fading channels, in Proc. IEEE 21st Int. Symp. Pers.
Indoor Mobile Radio Commun., Sep. 2010, pp. 945950.
[37] P. Theofilakos, A. G. Kanatas, and G. P. Efthymoglou, Performance of
generalized selection combining receivers in K fading channels, IEEE
Commun. Lett., vol. 12, no. 11, pp. 816818, Nov. 2008.
[38] Z. Wang and G. Giannakis, A simple and general parametrization quantifying performance in fading channels, IEEE Trans. Commun., vol. 51,
no. 8, pp. 13891398, Aug. 2003.
[39] M. D. Renzo and H. Haas, Bit error probability of space modulation over
Nakagami-m fading: Asymptotic analysis, IEEE Commun.Lett., vol. 15,
no. 10, pp. 10261028, Oct. 2011.
[40] K. P. Peppas, M. Zamkotsian, F. Lazarakis, and P. G. Cottis, Asymptotic error performance analysis of spatial modulation under generalized
fading, IEEE Wireless Commun. Lett., vol. 3, no. 4, pp. 421424, Aug.
2014.
[41] M. Niu, J. Cheng, and J. F. Holzman, Error rate performance comparison of coherent and subcarrier intensity modulated optical wireless
communications, J. Opt. Commun. Netw., vol. 5, no. 6, pp. 554564, Jun.
2013.
[42] M. Uysal, J. T. Li, and M. Yu, Error rate performance analysis of coded
free-space optical links over gamma-gamma atmospheric turbulence channels, IEEE Trans. Wireless Commun., vol. 5, no. 6, pp. 12291233, Jun.
2006.
[43] H. E. Nistazakis, E. A. Karagianni, A. D. Tsigopoulos, M. E. Fafalios,
and G. S. Tombras, Average capacity of optical wireless communication systems over atmospheric turbulence channels, J. Lightw. Technol.,
vol. 27, no. 8, pp. 974979, Apr. 2009.
[44] J. G. Proakis, Digital Communications, 3rd ed. New York, NY, USA: Mc
Graw Hill, 1995.

IE
E
Pr E
oo
f

[1] X. Zhu and J. M. Kahn, Free-space optical communications through

atmospheric turbulence channels, IEEE Trans. Commun., vol. 50, no. 8,
pp. 12931300, Aug. 2002.
[2] A. Garcia-Zambrana, Error rate performance for STBC in free-space
optical communications through strong atmospheric turbulence, IEEE
Commun. Lett., vol. 11, no. 5, pp. 390392, May 2007.
[3] S. M. Navidpour, M. Uysal, and M. Kavehrad, Performance of freespace optical transmission with spatial diversity, IEEE Trans. Wireless
Commun., vol. 6, no. 8, pp. 28132819, Aug. 2007.
[4] K. Peppas, A simple, accurate approximation to the sum of gammagamma variates and applications in MIMO free-space optical systems,
IEEE Photon. Technol. Lett., vol. 23, no. 13, pp. 839841, Jul. 2011.
[5] K. Peppas, F. Lazarakis, A. Alexandridis, and K. Dangakis, Simple,
accurate formula for the average bit error probability of multiple-input
multiple-output free-space optical links over negative exponential turbulence channels, Opt. Lett., vol. 37, pp. 32433245, Aug. 2012.
[6] L. C. Andrews, M. A. Al-Habash, C. Y. Hopen, and R. L. Phillips, Theory of optical scintillation: Gaussian beam wave model, Waves Random
Media, vol. 11, pp. 271291, 2001.
[7] M. A. Al-Habash, L. C. Andrews, and R. L. Phillips, Mathematical model
for the irradiance PDF of a laser beam propagating through turbulent
media, Opt. Eng, vol. 40, no. 8, pp. 15541562, 2001.
[8] J. H. Churnside and S. F. Clifford, Log-normal Rician probability density
function of optical scintillations in the turbulent atmosphere, J. Opt. Soc.
Amer., vol. 4, pp. 19231930, 1987.
[9] E. Jakeman, On the statistics of k-distributed noise, J. Phys. A, vol. 13,
pp. 3148, 1980.
[10] L. C. Andrews and R. L. Philips, Mathematical genesis of the I-K distribution for random optical fields, J. Opt. Soc. Amer. A, vol. 3, no. 11, pp.
19121919, 1986.
[11] L. C. Andrews, R. L. Phillips, and K. Shivamoggi, Relations of the
parameters of the I-K distribution for irradiance fluctuations to physical
parameters of the turbulence, Appl. Opt., vol. 27, no. 11, pp. 21502155,
1988.
[12] K. P. Peppas, A. N. Stassinakis, G. K. Topalis, H. E. Nistazakis, and
G. S. Tombras, Average capacity of optical wireless communication
systems over I-K atmospheric turbulence channels, J. Opt. Commun.
Netw., vol. 4, pp. 10261032, 2012.
[13] K. Kiasaleh, Performance of coherent DPSK free-space optical communication systems in k-distributed turbulence, IEEE Trans. Commun.,
vol. 54, no. 4, pp. 604607, Apr. 2006.
[14] M. Niu, J. Cheng, and J. F. Holzman, Error rate analysis of m-ary coherent
free-space optical communication systems with k-distributed turbulence,
IEEE Trans. Commun., vol. 59, no. 3, pp. 664668, Mar. 2011.
[15] S. M. Aghajanzadeh and M. Uysal, Diversity-multiplexing trade-off in
coherent free-space optical systems with multiple receivers, IEEE/OSA
J. Opt. Commun. Netw., vol. 2, no. 12, pp. 10871094, Dec. 2010.
[16] M. Niu, J. Cheng, and J. F. Holzman. (2014). Alamouti-type STBC for
atmospheric optical communication using coherent detection. IEEE Photon. J. [Online]. 6(1). Available: http://ieeexplore.ieee.org/xpls/icp.jsp?
arnumber=6725602
[17] E. Bayaki and R. Schober, Performance and design of coherent and
differential space-time coded FSO systems, J. Lightw. Technol., vol. 30,
no. 11, pp. 15691577, Jun. 2012.
[18] R. Mesleh, H. Elgala, and H. Haas, Optical spatial modulation, J. Opt.
Commun. Netw., vol. 3, no. 3, pp. 234244, Mar. 2011.
[19] R. Mesleh, H. Elgala, R. Mehmood, and H. Haas, Performance of optical
spatial modulation with transmitters-receivers alignment, IEEE Commun.
Lett., vol. 15, no. 1, pp. 7981, Jan. 2011.
[20] T. Fath and H. Haas, Performance comparison of MIMO techniques for
optical wireless communications in indoor environments, IEEE Trans.
Commun.,vol. 61, no. 2, pp. 733742, Feb. 2013.
[21] Y. A. Chau and S.-H. Yu, Space modulation on wireless fading channels,
in Proc. IEEE Veh. Technol. Conf. Fall, 2001, vol. 3, pp. 16681671.

11

12

JOURNAL OF LIGHTWAVE TECHNOLOGY

P. Takis Mathiopoulos (SM 94) received the Ph.D. degree in digital communications from the University of Ottawa, Ottawa, ON, Canada, in 1989.
From 1982 to 1986, he was with Raytheon Canada, Ltd., working in the areas
of air navigational and satellite communications. In 1988, he joined the Department of Electrical and Computer Engineering (ECE), University of British
Columbia (UBC), Vancouver, BC, Canada, where he was a Faculty Member as
a Professor from 2000 to 2003. From 2000 to 2014, he was with the Institute
for Space Applications and Remote Sensing (ISARS), National Observatory of
Athens (NOA), Athens, Greece, first as the Director and then as the Director of
research and established the Wireless Communications Research Group. As ISARS Director during 20002004, he led the Institute to a significant expansion
R&D growth, and international scientific recognition. For these achievements,
ISARS has been selected as a National Center of Excellence for the years 2005
2008. Since 2014, he has been an Adjunct Researcher at the Institute of Astronomy, Astrophysics, Space Applications and Remote Sensing, NOA. Since 2003,
he worked part time at the Department of Informatics and Telecommunications,
University of Athens, Athens, where since 2014, he has been a Professor of
digital communications. From 2008 to 2013, he was a Guest Professor with the
Southwest Jiaotong University, China. He is also appointed as a Guest Professor at the School of Information Engineering, Yangzhou University, Yangzhou,
China, for 20142015 under the Senior Foreign Expert Program of the Government of PR of China. Furthermore he is appointed by Keio University, Tokyo,
Japan, as a Visiting Professor at the Department of Information and Computer
Science for 20152016 under the Top Global University Project of the Ministry
of Education, Culture, Sports, Science and Technology Government of Japan.
For the last 25 years, he has been conducting research mainly on the physical
layer of digital communication systems for terrestrial and satellite applications,
including digital communications over fading and interference environments.
He coauthored a paper in GLOBECOM89 establishing for the first time in the
open technical literature the link between MLSE and multiple (or multisymbol)
differential detection for the AWGN and fading channels. He is also interested in
channel characterization and measurements, modulation and coding techniques,
synchronization, SIMO/MIMO, UWB, OFDM, software/cognitive radios, and
green communications. In addition, since 2010, he has been actively involved
with research activities in the fields of remote sensing, LiDAR systems, and photogrammetry. In these areas, he has coauthored more than 100 journal papers,
mainly published in various IEEE and IET journals, four book chapters, and
more than 120 conference papers. He has been the PI for more than 40 research
grants and has supervised the thesis of 11 Ph.D. and 23 Master students.
Dr. Mathiopoulos has been or currently serves on the editorial board of
several archival journals, including the IET Communications, and the IEEE
TRANSACTIONS ON COMMUNICATIONS from 1993 to 2005. He has regularly
acted as a consultant for various governmental and private organizations. Since
1993, he has served on a regular basis as a Scientific Advisor and a Technical
Expert for the European Commission (EC). In addition, since 2001, he has been
the Greek Representative to high-level committees in the EC and the European
Space Agency. He has been a Member of the TPC of more than 70 international
IEEE conferences, as well as the TPC Vice-Chair for the 2006-S IEEE VTC and
the 2008-F IEEE VTC, as well as the Cochair of the FITCE2011. He has delivered numerous invited presentations, including plenary and keynote lectures,
and has taught many short courses all over the world. As a Faculty Member
at the ECE, UBC, he was elected as an ASI Fellow and a Killam Research
Fellow. He received the two best paper awards for papers published in the 2nd
International Symposium on Communication, Control, and Signal Processing
in 2008), and the 3rd International Conference on Advances in Satellite and
Space Communications in 2011.

IE
E
Pr E
oo
f

Kostas P. Peppas was born in Athens, Greece, in 1975. He received the Diploma
degree in electrical and computer engineering and the Ph.D. degree in wireless
communications from the National Technical University of Athens, Athens, in
1997 and 2004, respectively. From 2004 to 2007, he was with the Department of
Computer Science, University of Peloponnese, Tripolis, Greece, and from 2008
to 2014 with the National Center for Scientific ResearchDemokritos, Institute
of Informatics and Telecommunications as a Researcher. In 2014, he joined
the Department of Telecommunication Science and Technology, University of
Peloponnese, where he is currently a Lecturer. His current research interests
include digital communications over fading channels, MIMO systems, wireless
and personal communication networks, and system level analysis and design.
He has authored more than 70 journal and conference papers.