Channel Modeling and Performance Analysis of Modul

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/314288405
Channel Modeling and Performance Analysis of Modulating Retroreﬂector

FSO Systems Under Weak Turbulence Conditions
Article in IEEE Photonics Journal · March 2017

DOI: 10.1109/JPHOT.2017.2677501
CITATIONS READS
27 234
7 authors, including:
Guowei Yang Li Zhipeng

Hangzhou Dianzi University Zhejiang University
63 PUBLICATIONS 488 CITATIONS 3 PUBLICATIONS 30 CITATIONS
SEE PROFILE SEE PROFILE
Meihua Bi Xuefang Zhou

Shanghai Jiao Tong University Hangzhou Dianzi University
191 PUBLICATIONS 993 CITATIONS 56 PUBLICATIONS 268 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Multi-wavelength Brillouin erbium-doped fiber laser sensor with high tunable temperature sensing coefficient View project
Fronthaul View project
All content following this page was uploaded by Guowei Yang on 21 March 2017.
The user has requested enhancement of the downloaded file.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JPHOT.2017.2677501, IEEE
Photonics Journal
IEEE Photonics Journal Volume XXX
Channel Modeling and Performance Analysis of

Modulating Retroreflector FSO Systems under
Weak Turbulence Conditions
Guowei Yang1 *, Zhipeng Li1 , Meihua Bi1 , Xuefang Zhou1 , Ran Zeng1 , Tianshu Wang2 ,
Jing Li3
1
College of Communication Engineering, Hangzhou Dianzi University, Hangzhou 310018, China
2
National and Local Joint Engineering Research Center of Space Optoelectronics Technology,
Changchun University of Science and Technology, Changchun 130022, China
3
CETC Key Laboratory of Aerospace Information Applications, Shijiazhuang 050081, China
*Corresponding author:guowei.yang@hdu.edu.cn
Abstract: This paper presents the fading channel of double-pass modulating retroreflector (MRR)
free-space optical (FSO) systems under weak turbulence conditions that can be modeled by the
distribution of the weighted product of two correlated Lognormal random variables (RVs). We firstly
testify the proposed channel model through using wave-optics simulation. Then, the probability density
function (PDF) and cumulative distribution function (CDF) of the channel model are derived in closed
form. The system bit error rate (BER) and outage probability (OP) are calculated by averaging the
conditional BER of on-off keying (OOK) modulation over the PDF and substituting the signal-to-noise
ratio (SNR) threshold into the CDF, respectively. The obtained analytical formnulas for the system
performance are validated by Monte-Carlo simulations. Moreover, the effect of the fading correlation
and the aperture averaging on the performance is investigated.
Index Terms: Free-space optics, Modulating retroreflector, Double-pass channel, Fading correlation,
Performance analysis.
1. Introduction
Free-space optical (FSO) communication systems have been extensively studied to provide high
speed mobile connectivity that is applicable to such as wireless access network, unmaned aerial
vehicle communications, ground-to-satellite and satellite-to-satellite links. To keep the system
compactness, the single-input single-output (SISO) FSO system is preferred but its terminals
need be equipped with high precise pointing, acquiste and tracking (PAT) systems. The resulting
large power consumption, weight and size are comparable to those of radio frequency (RF) and
millimeter wave (mmWave) systems provided with higher robustness to the weather conditions.
This prevents the SISO FSO systems from being widely used in limited-payload mobile platforms
[1], [2]. One promising solution is to construct an asymmetric FSO system through replacing one
traditional transceiver with a modulating retroreflector (MRR) that is composed of a light modulator
and a retro-reflective device such as a corner cube reflector (CCR) [3]. The small, battery operating
and semi-passive MRR releases the sophisticated PAT in the MRR terminal, and also could be
used to combat the pointing error that is an important concern in conventional SISO FSO systems
[4]–[7].
In the literatures, a great deal of previously-published works focused on MRR fabrication [3],
[8] and MRR FSO systems establishment [2], [9]. For the system design and optimization, the
performance evaluation based on an appropriate channel model is of crucial importance. The
Vol. xx, No. xx, January 2017 Page 1
1943-0655 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Photonics Journal
Fig. 1: Schematic of a typical MRR FSO system.
link budget analysis of MRR FSO systems in vacuum has been introduced in [10]. In turbulence
conditions, a group of papers computed the statistical moments of the optical field in the double-
pass propagation [11]–[13]. However, the channel fading model can not be uniquely determined
by the statistical moments [11]. The experiments of the double-pass propagation conducted in
[11], [14] and [15] implies the channel fading for the double-pass link (including the forward and
backward passes) and only the backwark link could be individually described by the Lognormal
distribution. On the other hand, [16] indicates that the optical beam in the double-pass propagation
is jointly affected by the turbulence effect in the forward pass from the transceiver to the MRR,
the reflection effect of the MRR, and the turbulence effect in the backward pass from the MRR to
the transceiver. The turbulence effect in a single pass can be described very well by Lognormal
distribution in weak turbulence conditions [17]. Thus, the “three-segment” model proposed in [16]
had been used to calculate the bit error rate (BER) of MRR FSO systems in weak turbulence
conditions was reported in [18]. However, its Lognormal-based model had not been testified
and the assumption of uncorrelated fading between the forward and backward passes is not
appropriate in monostatic configuration and most practical systems employing finite-size receiver
apertures, e.g., [19]–[22].
In this paper, the performance of MRR FSO systems under weak turbulence conditions is
analytically evaluated taking into account the fading correlation between the double passes. To
do this, we firstly study the channel of the double-pass MRR FSO link with the aid of wave-optics
simulation and then the channel is modeled by the product of a constant reflection ratio of the
MRR, the Lognormal-distributed fading in forward and backward passes (in Section 2). The BER
and outage probability (OP) are calculated based on the distribution of the weighted product of
two correlated Lognormal random variables (RVs) (in Section 3). Some analytical results verified
by Monte-Carlo simulations are provided to investigate the fading correlation and the aperture
averaging (in Section 4). Finally, we conclude this paper (in Section 5).
2. System configuration and channel model

Fig.1 shows the schematic of a typical simplex MRR FSO system that consists of the transceiver
and MRR terminals. A continuous wave (CW) laser beam, also called interrogating beam, lauched
from the transceiver propagates towards the MRR. A part of the beam incident on the MRR is mod-
ulated by the optical modulator and reflected by the CCR back to the transceiver. The information-
bearing reflected optical beam through the collecting lens is detected by the photodetector with
a large enough size to capture the total collected intensity. The optical signal is converted into
the electrical signal that will be demodulated to recover the information data sent from the MRR.
The optical carrier is emitted from the transceiver to the MRR, and is then modulated and retro-
reflected from the MRR to the transceiver, which refer to the forward- and backward-pass channels
respectively. The single-pass FSO channel could be described very well by Lognormal distribution
under weak turbulence conditions [17]. Intuitively, the forward- and backward-pass channel fading
should also follow the Lognormal distribution in weak turbulent atmosphere.
Here, we use wave-optics simulation method to testify the applicability of the Lognormal-based
Photonics Journal
model and evaluate the fading correlation between the two opposite-pass channels. The wave-
optics simulation for a single-pass FSO link is elaborated based on the split-step beam propagation
method in [23]. The simulation for the double-pass MRR FSO link can be divided into the wave
propagation from the transciver to the MRR, inside the CCR and from the MRR back to the
transciver [21]. A framework of the double-pass link in wave-optics simulation is shown in Fig.2(a),
where a number of random phase screens are arranged between the transceiver and CCR planes.
The intensity and phase of the optical field on all the planes (the screen also treated as the plane)
are discretized over a grid of N ∆d × N ∆d , where ∆d and N denote the grid spacing and the
number of grid points, respectively. At the transceiver plane, a Gaussian-beam wave Uin (j∆d , l∆d )
is launched and the returned optical wave Uout (j∆d , l∆d ) is received, where j, l = 0, 1, 2, · · · , N .
For the phase screens that record the atmospheric turbulence along the propagation path, we
generate random harmonic amplitudes over the N × N grid in the spectral domain based on the
modified von Kármán power spectrum, and then, take the inverse 2D discrete Fourier transform
to obtain the phase fluctuations [17], [24]. The optical wave denoted by UCCR (j∆d , l∆d ) radiates
on the CCR plane, and then the CCR rotates and retro-reflects the indicent wave wrapped in the
CCR aperture, as shown in the two figures on the right hand side of Fig.2(b) (see more details
in [21]). In general, the reflection effect of the CCR with the aperture diameter DCCR could be
expressed as [25]
Reff (θ) = circ(DCCR )η(θ). (1)
The circular step function circ(D) = 1 and 0 when (j∆d )2 +(l∆d )2 ≤ 2

√ D and (j∆
2 2
d ) +(l∆d ) > D ,
2
respectively. The effectivepreflection ratio η(θ) = 2/π sin µ − 2µ tan θref cos θ, where θref =
−1
sin−1 (sin(θ/n)) and µ = 1 − 2 tan2 θref . Also, θ is the incident angle with respect to the CCR
aperture and n denotes the refractive index of the CCR. Finally the split-step beam propagation
method is used to simulate the wave propagation between two adjacent planes and integrate the
effect of each plane [21], [23]. In the i-th simulation realization, the channel coefficients of the
forward and backward passes are respectively calculated as
ΣN N i
j=1 Σl=1 UCCR Reff (θ)
I1i = (2)
ΣN N i
j=1 Σl=1 Uin circ(DTRX )
and
ΣN N i
j=1 Σl=1 Uout circ(DTRX )
I2i = , (3)
ΣN N i
j=1 Σl=1 UCCR Reff (θ)
where DTRX denotes the aperture diameter of the transceiver as shown in Fig.1. To calculate
the fading correlation coefficient ρI between the two channels I1 and I2 , we first calculate the
mean and the standard deviation of the corresponding channels that we denote by I¯ and σI ,
respectively. Then, ρI is obtained as the sample Pearson correlation coefficient [26]
I1i − I¯1 I2i − I¯2

1
ρI = ΣN s
, (4)
Ns − 1 i=1 σI1 σI2
where Ns denotes the total number of simulation realizations.
2
To consider the weak turbulence conditions (i.e., Rytov variance σR = 1.23Cn2 (2π/λ)7/6 L11/6 is
less than one [17]), we set the link span L = 1km and the structure parameter of the air refractive
index Cn2 = 1 × 10−14 m−2/3 . Note that one Lognormal distribution describes one unidirectional
2
pass (i.e., the forward or backward pass) and hence we calculate the σR using the link span L
2
but not the signal propagation length 2L for the MRR FSO systems. The resultant σR = 0.2 when
using the optical wave with λ = 1550nm. We take N = 512 with ∆d = 1mm and use Ns = 20 phase
screens. Moreover, we assume perfect alignment between the transceiver and the MRR (i.e., the
light beam is normally incident upon the CCR aperture). Note that the value of Reff decreases
with the incident angle θ. We have run 10000 simulatin realizations to guarantee the accuracy of
our results. In Fig.3, we plot the probability density functions (PDFs) of the channel fading data
Photonics Journal
(a) (b)
Fig. 2: The description of wave-optics simulation (a) the framework of double-pass simulation link
(b) an illustration of simulation results.
2.5 3.5
Simulation data Simulation data
Lognormal−fit curve Lognormal−fit curve
3
2
2.5
1.5
2
PDF
PDF 1.5
1
1
0.5
0.5
0 0
0 2 4 6 8 10 0 2 4 6 8 10
normalized I normalized I
(a) (b)
Fig. 3: The PDFs and Lognormal-fit curves of normalized received intensity (a) in the forward-pass
channel, and (b) in the backward-pass channel.
obtained from wave-optics simulations and the corresponding Lognormal-fit curves for the MRR
FSO link with DCCR = 50cm and DTRX = 75cm. We notice that the Lognormal model could fit the
simulation data very well. Moreover, we evaluated the normalized variance of the irradiance σI2
(i.e., it is equivalent to the square of the standard deviation of the channel coefficient I) and the
fading correlation coefficent ρI . As listed in Table I, the relatively large ρI was also found in the
experiments [15]. Also, the decrease of ρI with the increasing aperture DTRX was consistent with
the theoretical calculation [27]. Here, we could draw a conclusion that the fading in the forward-
and backward-pass channels in MRR FSO systems can be individually modeled by Lognormal
distribution, and are practically correlated.
Now, we proceed to deriving the PDF and cumulative distribution function (CDF) of the channel
fading in double-pass MRR FSO systems. Let us denote the normalized intensity and log amplitude
TABLE I: σI2 and ρI for DCCR = 50cm and DTRX = 5, 25, 50, 75cm.
DTRX σI21 † σI22 † ρI
5cm 0.035 0.092 0.79
25cm 0.035 0.062 0.68
50cm 0.035 0.032 0.45
75cm 0.035 0.013 0.41
† σI21 and σI22 correspond to the forward- and backward-pass channels, respectively.
Photonics Journal
of the received optical field in a single-pass channel by I and X, respectively. We have

I = exp(2X), (5)
where X follows a normal distribution with the mean µX and the variance σX . Accordingly, the
PDF of I is given by [17]
" #
1 (ln I − 2µX )2
f (I) = √ exp − 2 , I > 0. (6)
2 2πIσX 8σX
In practice, we could only estimate the mean µI and the variance σI of I. With the aid of [28,
(44.35)], the corresponding µX and σX can be calculated by
2
1 1 σI

 µ = ln µI − ln +1 ,
 X

 2 4 µ2I
s (7)
σI2

 1
 σX = ln +1 .


2 µ2I
Considering the double-pass MRR FSO systems, the channel fading Ic can be regarded as the
weighted product of the forward- and backward-pass fading [16], [18],
Ic = RI1 I2 = R exp [2(X1 + X2 )] = R exp(2Xc ), (8)
where I1 and I2 are the channel coefficients of the forward and backward passes, respectively,
and they are governed by two correlated Lognormal distributions with a correlation coefficient ρI .
We set Xc = X1 + X2 . Moreover, the normalized reflection ratio of the CCR R = Reff (θ)/Reff (0)
could be a constant for an established system in a certain transimission duration (e.g., the maximal
value of R = 1 corresponding to the perfect alignment between the transceiver and the MRR).
From (8), we find that the double-pass channel fading follows the PDF of the weighted product of
two correlated Lognormal RVs, which can be deduced from the PDF of the sum of two correlated
normal RVs. Note that the sum of Lognormal RVs and the product of independent Lognormal
RVs have been studied (e.g., [29], [30]), but the weighted product of correlated Lognormal RVs
have not been addressed.
To obtain the PDF of Xc , we start from the bivariate normal distribution as given by
(X1 − µX1 )2

1 1
f (X1 , X2 ) = p exp − 2 2
2πσX1 σX2 1 − ρ2X 2(1 − ρX ) σX 1
(X2 − µX2 )2

2ρX (X1 − µX1 )(X2 − µX2 )
+ 2 − , (9)
σX 2
σX1 σX2
where ρX represents the correlation coefficient between X1 and X2 . We then substitute X2 =
Xc − X1 into (9) and take the integration with respect to X1 ,
(X1 − µX1 )2

1 1
Z ∞
f (Xc ) = p exp − 2 2
2πσX1 σX2 1 − ρX 02 2(1 − ρX ) σX 1
(Xc − X1 − µX2 )2

2ρX (X1 − µX1 )(Xc − X1 − µX2 )
+ 2 − dX1 (10)
σX 2
σX1 σX2
(Xc − µXc )2

1
= √ exp − 2 .
2πσXc 2σX c
Where the mean and the variance of Xc are given as


 µXc = µX1 + µX2 ,
q
2 + σ 2 + 2ρ σ
(11)
 σXc = σX X2 X X1 σX2 .
1
Photonics Journal
Fig. 4: An illustration of the proposed modulation scheme.
Interchanging the variables in (8) and then, substituting Xc = (ln Ic − ln R)/2 into (10), the PDF
of Ic can be expressed by
(ln Ic − ln R − 2µXc )2

1
f (Ic ) = √ exp − 2 , (12)
2 2πIc σXc 8σX c
and its CDF is given as

1 ln Ic − ln R − 2µXc
F (Ic ) = 1 + erf √ , (13)
2 2 2σXc
where erf(·) represents the error function. Note that in (11) ρX is not directly measured in practice,
but it can be related to the measurable ρI via

1 ρI σI1 σI2
ρX = ln 2 + σ 2 )] + 1 . (14)
4σX1 σX2 exp[2(µX1 + µX2 + σX 1 X2
Now the parameters µXc and σXc of the PDF and CDF can be calculated by (7), (11) and (14).
3. Derivation of BER and outage probability

We consider a double-pass MRR FSO link using uncoded on-off keying (OOK) modulation. As
shown in Fig.4, the unmodulated CW laser beam propagates through the forward pass and arrives
at the MRR. Then, the laser beam is modulated using the OOK by the optical modulator (e.g.,
multiple quantum well (MQW) modulator [2]), and retro-reflected by the CCR through the backward
pass to the transceiver. We perform optimal detection assuming perfect channel state information
(CSI). Also, we use the narrow pass-band optical filter in front of a PIN photodetector to reduce
the background radiations, and consequently we assume that the dominant noise source at the
detector is the thermal noise and model it by an additive white Gaussian noise with zero mean
and variance σn2 [31]. We set the intensity for signal ‘1’ and the quantum efficiency to one. The
received electrical signal-to-noise ratio (SNR) per bit is defined as γ = Ic2 /4σn2 and its average is
γ = E[Ic ]2 /4σn2 = 1/4σn2 . Then, the average BER is obtained by averaging the conditional BER of
OOK modulation over the PDF of Ic [32],
+∞ +∞
1 Ic 1 Ic γ
Z Z
Pe = f (Ic )erfc √ dIc = f (Ic )erfc √ dIc (15)
2 2 2σn 2 2
0 0
where erfc(·) denotes the complementary error function. The integral (15) can be efficiently
computed by Gauss-Hermite quadrature formula [33]. By the variable change y = (ln Ic − ln R −
Photonics Journal
√
2µXc )/(2 2σXc ), we obtain the BER formula
+∞ √ !
1 γ exp 2 2σXc y + ln R + 2µXc
Z
2
Pe = √ e−y erfc √ dy
2 π 2
−∞
√ (16)
k !
1 X γ exp 2 2σXc yi + ln R + 2µXc
≈ √ wi erfc √ .
2 π i=1
2
Where yi and wi are the ith roots of the Hermite polynomial and the corresponding weight factors,
respectively.
Meanwhile, the OP of communication systems is defined as the probability that the received
SNR γ falls below a specified threshold γth [32]. The formula of the OP Pout can be obtained by
  p 
p 1 ln γth /γ − ln R − 2µXc
Pout = F γth /γ = 1 + erf  √  . (17)
2 2 2σXc
4. Numerical results
At first, to verify the accuracy of the proposed formulas, the BER and OP of the double-pass
MRR FSO systems under different turbulence conditions with different fading correlation ρI are
calculated by the analytical formulas and evaluated by Monte-Carlo simulations as well. For
the Monte-Carlo based performance evaluation, two correlated Lognormal RVs can be easily
generated from two correlated Gaussion RVs based on (8) and (14). Two cases of the considered
turbulence conditions are indicated in Fig.5. Note that the channel parameters of the two cases
are specially set, and in particular the case with σI2 = 1.2 is considered in order to check whether
the accuracy of the proposed analytical formulas is guaranteed over the whole range of the weak
turbulence conditions. Without loss of generality, we have set R = 1 for all case studies. The
curves of system BER and OP are shown in Fig.5 (a) and (b), respectively. We notice that all
analytical results are very well coincide with those obtained by Monte-Carlo simulations. The slight
deviations at very low performance values (e.g., Pout < 10−6 ) should be caused by the limited
samples used in simulations. From both figures, the system performance is impaired by the fading
correlation, and deteriorating with increasing ρI . In the Case (2), ρI = 0.3 and 0.9 respectively
lead to the SNR penalties of 5dB and 13dB at the BER of 10−5 compared to ρI = 0.
Next, we directly use the proposed analytical formulas to study the aperture averaging arising
from the transceiver aperture. The system configurations set in wave-optics simulation are used
here. Consider four cases with different transceiver apertures DTRX = 5, 25, 50, and 75cm, the
corresponding variances of channel fading in the two opposite passes and the fading correlation
coefficients between the two opposite channels were listed in Table I, together with the normalized
channel fading set for all cases. The system BER and OP results are given in Fig.6. As expected,
the performance improvement is obtained through increasing the transceiver aperture (see the
four cases with actual correlation coefficients in Fig.6), and this verifies the effect of aperture
averaging. Compared with the performance results for DTRX = 25, 50, and 75cm with ρI = 0.79 as
plotted in Fig.6, we find that the aperture increase leads to the averaged intensity fluctuation and
the reduced fading correlation (see Table I) that simultaneously improve the system performance.
5. Conclusion
Through the wave-optics simulation for the double-pass MRR FSO link, we confirmed that the
channel fading in the forward and backward passes follows the Lognormal distribution and are
practically correlated. The correlated fading in the double-pass MRR FSO sysntems were con-
sidered as two correlated Lognormal RVs. By deriving the exact PDF and CDF of the weighted
Photonics Journal
10 0 10 0
Case (2): µI1 = 1, σI1 = 0.4;
-1
10 10 -1 µI2 = 1, σI2 = 1.2.
Case (2): µI1 = 1, σI1 = 0.4;
10 -2 µI2 = 1, σI2 = 1.2. 10 -2
Anal., ρI = 0
Anal., ρI = 0
Anal., ρI = 0.3
10 -3
10 -3 Anal., ρI = 0.3
Anal., ρI = 0.5
Anal., ρI = 0.5
Anal., ρI = 0.9
Anal., ρI = 0.9
10 -4 Monte-Carlo Simu. 10 -4
Monte-Carlo Simu.
Pout
Pe
-5
10 10 -5
10 -6 10 -6
10 -7 10 -7
-8
Case (1): µI1 = 1, σI1 = 0.2;
10 10 -8
µI2 = 1, σI2 = 0.4.
Case (1): µI1 = 1, σI1 = 0.2;
10 -9 10 -9 µI2 = 1, σI2 = 0.4.
10 -10 10 -10
-10 0 10 20 30 40 50 60 70 80 -60 -50 -40 -30 -20 -10 0 10 20
γ(dB) γth /γ(dB)
(a) (b)
Fig. 5: Performance results of the double-pass MRR FSO systems under two arbitrarily different
turbulence conditions. (a) Average BER versus average SNR, (b) outage probability versus
SNR threshold. Line and circle correspond to the analytical and Monte-Carlo simulation results,
respectively.
0 0
10 10
D =5cm, actual corr.
TRX
-1 D =25cm, actual corr. -1
10 TRX 10
D =25cm, ρ =0.79
TRX I
TRX
-2 -2
10 D =50cm, ρ =0.79 10
TRX I
D =75cm, actual corr. D =5cm, actual corr.
TRX TRX
-3 D =75cm, ρ =0.79 -3
10 TRX I 10 D =25cm, actual corr.
TRX
D =25cm, ρ =0.79
TRX I
-4 -4
10 10
TRX
D =50cm, ρ =0.79
Pout
TRX I
Pe
-5 -5
10 10
TRX
D =75cm, ρ =0.79
-6 -6 TRX I
10 10
-7 -7
10 10
-8 -8
10 10
-9 -9
10 10
-10 -10
10 10
0 5 10 15 20 25 30 35 40 -30 -25 -20 -15 -10 -5 0 5 10
γ(dB) γth /γ(dB)
(a) (b)
Fig. 6: Performance results of double-pass MRR FSO systems with different transceiver apertures.
(a) Average BER versus average SNR, (b) outage probability versus SNR threshold.
product of two correlated Lognormal RVs, we obtained the formulas for the BER and OP of MRR
FSO systems employing OOK modulation, which are validated by Monte-Carlo simulation results.
Furthermore, we indicate that the fading correlation indeed impaire the system performance, and
the aperture averaging effect of the transceiver arises from the averaged intensity fluctuation and
the reduced fading correlation simultaneously.
Photonics Journal
Acknowledgments
This work was supported by the National Natural Science Foundation of China (Grant No. 61405051,
61501157, 61308061, 11574068), the Zhejiang Provincial Natural Science Foundation of China
(Grant No. LY17F050012, LQ16F050004), and the universities cooperation funding of CETC Key
Laboratory of Aerospace Information Applications.
References
[1] M. A. Khalighi and M. Uysal, “Survey on free space optical communication: A communication theory perspective,”
IEEE Commun. Surveys Tuts., vol. 16, no. 4, pp. 2231–2258, June 2014.
[2] P. G. Goetz, W. S. Rabinovich, R. Mahon, J. L. Murphy, M. S. Ferraro, M. R. Suite, W. R. Smith, H. R. Burris, C. I.
Moore, W. W. Schultz, W. T. Freeman, S. J. Frawley, B. M. Mathieu, K. Hacker, and S. Reese, “Modulating retro-
reflector lasercom systems for small unmanned vehicles,” IEEE J. Sel. Areas Commun., vol. 30, no. 50, pp. 986–992,
June 2012.
[3] M. Achour, “Free-space optical communication by retro-modulation: Concept, technologies and challenges,” Proc.
SPIE, vol. 5614, pp. 52–63, Dec. 2004, London, UK.
[4] M. Hulea, Z. Ghassemlooy, S. Rajbhandari, and X. Tang, “Evaluation of the spherical concave mirror and convex
lens in compensating turbulence effect on fso systems,” IEEE, IET Int. Symp. Comm. Syst., Netw. and Digit. Signal
Process.(CSNDSP), pp. 914–918, July 2014, manchester, UK.
[5] ——, “Compensating for optical beam scattering and wandering in fso communications,” J. Lightw. Technol., vol. 32,
no. 7, pp. 1323–1328, Apr. 2014.
[6] E. Zedini and M. S. Alouini, “On the performance of multihop heterodyne fso systems with pointing errors,” IEEE
Photon. J., vol. 7, no. 2, pp. 1–10, Apr. 2015.
[7] P. V. Trinh, T. C. Thang, and A. T. Pham, “Mixed mmwave rf/fso relaying systems over generalized fading channels
with pointing errors,” IEEE Photon. J., vol. 9, no. 1, pp. 1–14, Feb. 2017.
[8] E. Rosenkrantz and S. Arnon, “1550 nm modulating retroreflector based on coated nanoparticles for free-space
optical communication,” Appl. Opt., vol. 54, no. 17, pp. 5309–5313, June 2015.
[9] E. J. Hällstig, J. Öhgren, L. Allard, L. J. Sjöqvist, D. Engstroem, S. Hard, D. Ågren, S. Junique, Q. Wang, and
B. Noharet, “Retrocommunication utilizing electroabsorption modulators and nonmechanical beam steering,” Opt.
Eng., vol. 44, no. 4, pp. 045 001–1–8, Apr. 2005.
[10] E. Rosenkrantz and S. Arnon, “An innovative modulating retro-reflector for free-space optical communication,” Proc.
SPIE, vol. 8874, pp. 88 740D–88 740D–9, Sep. 2013, San Diego, CA.
[11] G. Y. Patrushev, A. I. Petrov, and V. V. Pokasov, “Intensity fluctuations upon specular reflection of optical beams in
a turbulent atmosphere,” Radiophys. Quantum Electron., vol. 26, no. 7, p. 612619, July 1983.
[12] L. C. Andrews, R. L. Phillips, and W. B. Miller, “Mutual coherence function for a double-passage retroreflected optical
wave in atmospheric turbulence,” Appl. Opt., vol. 36, no. 3, pp. 698–708, Jan. 1997.
[13] L. C. Andrews and R. L. Phillips, “Monostatic lidar in weak-to-strong turbulence,” Waves Random Media, vol. 11,
no. 3, pp. 233–245, Mar. 2001.
[14] W. S. Rabinovich, R. Mahon, M. Ferraro, P. G. Goetz, and J. L. Murphy, “Reduction of scintillation in optical modulating
retro-reflector links,” Opt. Express, vol. 22, no. 23, pp. 28 553–28 565, Nov. 2014.
[15] R. Mahon, M. S. Ferraro, P. G. Goetz, C. I. Moore, J. Murphy, and W. S. Rabinovich, “Irradiance correlations in
retro-reflected beams,” Appl. Opt., vol. 54, no. 31, p. F96, Nov. 2015.
[16] J. L. Bufton, R. S. Iyer, and L. S. Taylor, “Scintillation statistics caused by atmospheric turbulence and speckle in
satellite laser ranging,” Appl. Opt., vol. 16, no. 9, pp. 2408–2413, Sep. 1977.
[17] L. C. Andrews and R. L. Phillips, Laser Beam Propagation Through Random Media, 2nd ed. Bellingham, Washington:
SPIE Press, 2005.
[18] D. Geng, P. Du, W. Wang, G. Gao, T. Wang, and M. Gong, “Single laser free-space duplex communication system
with adaptive threshold technique and ber analysis in weak turbulent atmosphere,” Opt. Lett., vol. 39, no. 13, pp.
3950–3953, July 2014.
[19] J. Minet, M. A. Vorontsov, E. Polnau, and D. Dolfi, “Enhanced correlation of received power-signal fluctuations in
bidirectional optical links,” J. Opt., vol. 15, no. 2, pp. 1–8, Feb. 2013.
[20] N. Perlot and D. Giggenbach, “Scintillation correlation between forward and return spherical waves,” Appl. Opt.,
vol. 51, no. 15, pp. 2888–2893, May 2012.
[21] S. You, G. Yang, M. Bi, Y. Wei, Y. Lu, and X. Zhou, “Wave-optics simulation of the channel fading in modulating
retro-reflector free-space optical link,” 2015 Int. Conf. Wireless Comm. & Signal Process. (WCSP), pp. 1–5, Oct.
2015, Nanjing, China.
[22] M. M. Abadi, Z. Ghassemlooy, S. Zvanovec, M. R. Bhatnagar, M.-A. Khalighi, and Y. Wu, “Impact of link parameters
and channel correlation on the performance of fso systems with the differential signaling technique,” IEEE J. Opt.
Commun. Netw., vol. 9, no. 2, pp. 138–148, Feb. 2017.
[23] J. D. Schmidt, Numerical Simulation of Optical Wave Propagation With Examples in MATLAB. SPIE Press, 2010.
[24] G. Yang, M.-A. Khalighi, S. Bourennane, and Z. Ghassemlooy, “Fading correlation and analytical performance
evaluation of space-diversity free-space optical communication systems,” J. Opt., vol. 16, no. 3, pp. 1–10, Feb.
2014.
[25] P. O. Minott, “Design of retrodirector arrays for laser ranging of satellite,” NASA TM-X-723-74-122, pp. 1–21, Mar.
1974, goddard Space Flight Center.
Photonics Journal
[26] G. Yang, M. A. Khalighi, and S. Bourennane, “Performance of receive diversity FSO systems under realistic beam
propagation conditions,” IEEE, IET Int. Symp. Comm. Syst., Netw. and Digit. Signal Process.(CSNDSP), pp. 1–5,
July 2012, Poznan, Poland.
[27] N. Perlot and D. Giggenbach, “Scintillation correlation between forward and return spherical waves,” Appl. Opt.,
vol. 51, no. 15, pp. 2888–2893, May 2012.
[28] S. Kotz, N. Balakrishnan, and N. L. Johnson, Continuous Multivariate Distributions Volume 1: Models and Applications,
2nd ed. New York, US: Wiley-Interscience, Apr. 2000.
[29] N. C. Beaulieu and Q. Xie, “An optimal lognormal approximation to lognormal sum distributions,” IEEE Trans. Veh.
Technol., vol. 53, no. 2, pp. 479–489, Mar. 2004.
[30] M. D. Renzo, F. Graziosi, and F. Santucci, “A comprehensive framework for performance analysis of cooperative
multi-hop wireless systems over log-normal fading channels,” IEEE Trans. Commun., vol. 58, no. 2, pp. 531–544,
Feb. 2010.
[31] G. Yang, M.-A. Khalighi, Z. Ghassemlooy, and S. Bourennane, “Performance evaluation of receive-diversity free-space
optical communications over correlated Gamma-Gamma fading channels,” Appl. Opt., vol. 52, no. 24, pp. 5903–5911,
Aug. 2013.
[32] G. Yang, M. A. Khalighi, Z. Ghassemlooy, and S. Bourennane, “Performance analysis of space-diversity free-space
optical systems over the correlated Gamma-Gamma fading channel using Padé approximation method,” IET Comm.,
vol. 8, no. 13, pp. 2246–2255, Sep. 2014.
[33] M. K. Simon and M.-S. Alouini, Digital Communication over Fading Channels: A Unified Approach to Performance
Analysis, 1st ed. New York: John Wiley & Sons, Inc., Dec. 2000.
View publication stats http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

Channel Modeling and Performance Analysis of Modul

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Channel Modeling and Performance Analysis of Modul

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Channel Modeling and Performance Analysis of Modulating Retroreﬂector

Article in IEEE Photonics Journal · March 2017

Guowei Yang Li Zhipeng

SEE PROFILE SEE PROFILE

Meihua Bi Xuefang Zhou

SEE PROFILE SEE PROFILE

Fronthaul View project

The user has requested enhancement of the downloaded file.

IEEE Photonics Journal Volume XXX

Channel Modeling and Performance Analysis of

Vol. xx, No. xx, January 2017 Page 1

IEEE Photonics Journal Volume XXX

Fig. 1: Schematic of a typical MRR FSO system.

2. System configuration and channel model

Vol. xx, No. xx, January 2017 Page 2

IEEE Photonics Journal Volume XXX

The circular step function circ(D) = 1 and 0 when (j∆d )2 +(l∆d )2 ≤ 2

Vol. xx, No. xx, January 2017 Page 3

IEEE Photonics Journal Volume XXX

Vol. xx, No. xx, January 2017 Page 4

IEEE Photonics Journal Volume XXX

of the received optical field in a single-pass channel by I and X, respectively. We have

Where the mean and the variance of Xc are given as

Vol. xx, No. xx, January 2017 Page 5

IEEE Photonics Journal Volume XXX

Fig. 4: An illustration of the proposed modulation scheme.

and its CDF is given as

3. Derivation of BER and outage probability

Vol. xx, No. xx, January 2017 Page 6

IEEE Photonics Journal Volume XXX

Vol. xx, No. xx, January 2017 Page 7

IEEE Photonics Journal Volume XXX

Vol. xx, No. xx, January 2017 Page 8

IEEE Photonics Journal Volume XXX

Vol. xx, No. xx, January 2017 Page 9

IEEE Photonics Journal Volume XXX

Vol. xx, No. xx, January 2017 Page 10

You might also like