Optimal Packet Length for Free-Space Optical Communications with Average SNR Feedback Channel

Alnwaimi, Ghassan; Boujemaa, Hatem; Arshad, Kamran

doi:https://doi.org/10.1155/2019/4703284

Journal of Computer Networks and Communications

On this page

Abstract Introduction Conclusion Appendix Data Availability Conflicts of Interest References Copyright Related Articles

Research Article | Open Access

Volume 2019 | Article ID 4703284 | https://doi.org/10.1155/2019/4703284

Optimal Packet Length for Free-Space Optical Communications with Average SNR Feedback Channel

Ghassan Alnwaimi,¹Hatem Boujemaa,²and Kamran Arshad³

Academic Editor: Djamel F. H. Sadok

Received22 Oct 2018

Revised14 Apr 2019

Accepted05 May 2019

Published16 May 2019

Abstract

In this article, a method to enhance data rates of free-space optical (FSO) systems using packet length optimization is proposed. The average signal-to-noise ratio (ASNR) is measured at the receiver and sent back to the transmitter to optimize packet length. In addition, the length of packet is optimized to enhance the average throughput. We concluded that packet length can be reduced at low ASNR. However, packet length should be increased at higher values of received ASNR. For each ASNR, we also choose the optimal modulation and coding scheme (MCS) and optimal packet length to maximize the throughput. Different MCSs are investigated such as 4-pulse amplitude modulation (PAM) with and without channel coding, 8-PAM, 16-PAM, and 32-PAM. The proposed method gives 0.8–1.9 dB gain with respect to conventional FSO with adaptive modulation and coding (AMC) and fixed packet length. This is the first paper to deal with packet length optimization for FSO systems.

1. Introduction

FSO communications allow high data rates as compared to radio frequency (RF) communications [1–5]. It is easy to install and does not require uncovering of walkways to introduce fiber joins. Moreover, FSO spectrum is license free. Therefore, FSO communications become viable due to its low implementation cost from one side and to overcome RF spectrum scarcity on the other side. However, the performance of FSO systems can be degraded due to rain, mist, tide, warm, or pointing errors [1–10]. To circumvent this drawback, we can use cooperative or spatial diversity combined with error control coding [11–17]. In cooperative FSO, relays amplify or decode the source message before transmitting it to the destination. Cooperative protocols allow us to benefit from cooperative diversity as the same information is transmitted over independent relayed channels [11–17]. Cooperative communications and channel coding techniques allow us to combat the effects of atmospheric turbulence and improve the performance of FSO systems.

Adaptive modulation and coding (AMC) allows us to increase data rates in free-space optical (FSO) communications. In fact, the best modulation and coding scheme (MCS) is selected for each instantaneous or average SNR. At high SNRs, we can use 32-pulse amplitude modulation (32-PAM) to increase data rates since its spectral efficiency is 5 bit/s/Hz. At low SNR, robust on-off keying (OOK) modulation with channel coding can be used. It is well known that channel coding is required to overcome channel impairments and noise. Different MCSs can be deployed for each range of SNR: i.e., is used if . In this paper, we will provide a methodology to derive the values of thresholds for fixed and optimal packet length. The thresholds for optimal packet length are denoted by , and thresholds for fixed packet length are denoted by .

In all previous studies, packet length is fixed and only the MCS is varied where different-PAM can be used with different channel encoders [18–26]. The main contribution of this paper is to optimize packet length in order to maximize the system throughput. For each average signal-to-noise ratio (ASNR), we choose the optimal packet length and MCS to have the largest throughput. The proposed solution allows 0.8–1.9 dB gain with respect to conventional FSO systems with AMC and constant packet length. This is the first paper to deal with packet length optimization for FSO systems.

The system model is presented in Section 2. The packet error probability is analyzed in Section 3 in the absence and presence of channel coding. The optimal packet length is derived in Section 4. Adaptive packet length and MCS using ASNR is described in Section 5. Numerical results are given in Section 6. Section 7 concludes the paper.

2. Signal Model

In FSO communications, the received signal is written as follows:where is the transmitted signal, l is the symbol index, is the atmospheric fading coefficient, is the effect of pointing errors, is an AWGN with normalized variance, and is the average SNR.

The average SNR can be expressed as follows:where is the transmitted energy per bit and is the power spectral density of the noise.

The instantaneous photo SNR is expressed as follows:

We denoteas the atmospheric turbulence.

Using a Gamma-Gamma distribution for atmospheric turbulence and in the presence of pointing errors, the probability density function (PDF) is equal to [27]where G is Meijer’s G function and b is the ratio between the equivalent beam radius at the receiver and pointing error displacement standard deviation. For small pointing errors, i.e., the above distribution converges to the nonpointing error case, where and correspond to strong turbulence, and correspond to moderate turbulence, and and correspond to weak turbulence [28, 29].

The cumulative distribution (CDF) of SNR is given by the following equation [27]:

3. Preliminary Results

The average packet error probability (PEP) can be tightly upper bounded by the following equation [30]:where is the PDF of SNR and is a waterfall threshold.

Equation (7) shows that the PEP for a given instantaneous SNR is approximated by 1. However, the PEP for a given instantaneous SNR is approximated by 0.

The waterfall threshold can be written as follows [30]:where is the PEP for a given instantaneous SNR γ.

3.1. Without Channel Coding

For uncoded M-PAM, we havewhere N is the number of data bits per packet, is the number of parity bits per packet, and is the symbol error probability (SEP) of M-PAM [31].using

Using (10) and (11), we havewhere

3.2. With Channel Coding

For convolutionally coded communications, we havewherein which and are, respectively, the free distance and distance spectrum.where is the rate of convolutional encoding.

Using equation (11), we havewhere

We havewhere for coded communications and for uncoded communications.

3.3. Waterfall Threshold

Appendix shows that waterfall threshold is written aswherewhere is the Euler constant.

4. Optimal Packet Length Using ASNR

The number of trials of HARQ is [32–34] calculated as follows:

The throughput in bit/s/Hz is expressed as follows:where is the used bandwidth and is the symbol period. In equation (23), we used the expression of PEP provided in equation (7).

The optimal packet length maximizing the throughput can be obtained using the gradient algorithm:

Usually, the gradient search is used to minimize a function f, and we write . Here, we aim to maximize the throughput which is equivalent to minimize .

The derivative of throughput with respect to packet length is equal to

This expression is valid for Gamma-Gamma atmospheric turbulence with pointing errors.

The used packet length is the closed solution to that obtained by gradient search such that is a multiple of so that bits can be converted to M-PAM symbols.

5. Optimal Packet Length with AMC

Figure 1 shows the average throughput of FSO communications with respect to ASNR using optimal N obtained with gradient search equation (24). From Figure 1, we can deduce the following strategy:(i): use convolutionally coded 4-PAM (ii): use uncoded 4-PAM(iii): use uncoded 8-PAM(iv): use uncoded 16-PAM(v): use uncoded 32-PAM

Thresholds correspond to intersection of different curves in Figure 1.

As shown in Figure 2, a similar approach is used when packet length is fixed, and with thresholds dB, , and .

Figure 3 shows the proposed adaptive packet length and adaptive MCS using ASNR . The ASNR is measured at the receiver to select the optimal packet length and MCS. The ASNR is compared to thresholds , , , and to determine the optimal MCS. Packet length is computed using the gradient algorithm as explained in equations (24) and (25) .

6. Numerical Results

Numerical results were obtained in the presence of both atmospheric turbulence and pointing errors. Except in Figure 4, the ratio b between the equivalent beam radius at receiver and pointing error displacement standard deviation is 12. Weak, moderate, and strong turbulence were considered. Figures 5 and 6 show, respectively, the average throughput of FSO communications for 4-PAM, ASNR = 15 dB (respectively, ASNR = 20 dB) with respect to packet length N. In all results, . We notice that throughput can be maximized by optimizing packet length. In fact, we notice that we can choose the value of packet length N to maximize the throughput. As shown in Figure 5, increasing packet length does not mean increasing the throughput all the time and there is a certain length that maximizes the throughput. The optimal packet length increases as the average SNR increases, as shown in Figures 5 and 6. The optimal packet length is 60 (respectively, 120) for (respectively, 20 dB). By comparing the results of Figures 5 and 6, we concluded that packet length should be increased as the ASNR increases.

Figure 7 compares the average throughput of 4-PAM when packet length is optimal with respect to the fixed packet length (FPL) N of 200, 400, and 1000. We notice that packet length optimization allows 0.8 dB gains with respect to N = 200 and throughput equal to 0.5 bit/s/Hz. Optimal packet length allows 1.3 dB gains with respect to N = 400 and throughput equal to 0.5 bit/s/Hz. Optimal packet length allows 1.9 dB gains with respect to N = 1000 and throughput equal to 0.5 bit/s/Hz. At high ASNR, N = 1000 allows higher throughput than N = 400 and N = 200. At low ASNR, N = 200 allows higher throughput than N = 400 and N = 1000.

Figure 1 shows the average throughput of FSO communications with optimal packet length for different modulations and coding schemes (MCSs): 4-PAM with convolutional coding; 4-PAM, 8-PAM, 16-PAM, and 32-PAM without channel coding. The following AMC strategy is suggested:(i): use convolutionally coded 4-PAM with (ii): use uncoded 4-PAM(iii): use uncoded 8-PAM(iv): use uncoded 16-PAM(v): use uncoded 32-PAM

The thresholds correspond to the intersection of different curves.

Figure 2 shows the average throughput of FSO communications with for different modulations and coding schemes (MCSs): 4-PAM with convolutional coding; 4-PAM, 8-PAM, 16-PAM, and 32-PAM without channel coding. If 4-PAM is used, the packet of length is converted to 420/2 = 210 symbols. If 8-PAM is used, the packet is converted to 420/3 = 140 symbols. If 16-PAM is used, the packet is converted to 420/4 = 105 symbols. If 32-PAM is used, the packet is converted to 420/5 = 84 symbols. Therefore, packet length should be a multiple of 60.

The following AMC approach is suggested:(i): use 4-PAM with convolutional coding (ii): use uncoded 4-PAM(iii): use uncoded 8-PAM(iv): use uncoded 16-PAM(v): use uncoded 32-PAM. Packet length is equal to N = 410

Thresholds correspond to intersection of different curves.

4-PAM transmits 2 bits per symbol, whereas 5 bits are coded in one 32-PAM symbol. As the number of bits per symbol increases, the spectral efficiency increases, but the PEP also increases since the symbols are closer to each other for a fixed transmit energy per bit.

Figure 8 shows the throughput of FSO systems with adaptive modulation and coding (AMC) for fixed and adaptive packet length. Optimal packet length offers 1 dB gain with respect to . The curve of optimal packet length with AMC is the upper bound of 5 curves of Figure 1. The curve of fixed packet length with AMC is the upper bound of 5 curves of Figure 2.

The effects of pointing errors are investigated in Figure 4 for and . b is the ratio between the equivalent beam radius at the receiver and pointing error displacement standard deviation. We notice that the proposed optimal packet length allows us to increase the throughput even in the presence of pointing errors. When b is small, the pointing errors increase and the throughput decreases.

We compare in Figure 9 the throughput derived in our paper for N = 400 to that obtained from [27]. We used the PDF of SNR [27] to compute the PEP and throughput of 4-PAM as an integral as follows:where is the PEP for SNR equal to x, as given in equation (12).

Another contribution of the paper is to show that PEP can be deduced from CDF of SNR and yields close results to the PEP computed using the above integral. Figure 9 shows that our results are very close to those of [27]. We have provided in equation (10) a tight upper bound of PEP. Therefore, it is a tight lower bound on throughput.

7. Conclusion

In this paper, we have optimized the throughput for FSO communications. For each average SNR (ASNR), we choose the optimal packet length and MCS to have the largest throughput. Optimal packet length was obtained using the gradient search algorithm and yields 0.8–1.9 dB gain with respect to conventional FSO. We have shown that the optimal packet length should be increased with average SNR to obtain higher throughput. We have also identified threshold to select the appropriate MCS for each average SNR and ensure higher data rates. The average SNR is compared to thresholds: when average SNR (ASNR) is less than , 4-PAM with half-rate channel coding is used. When 14.5 dB < ASNR ≤ 22.2 dB, uncoded 4-PAM is the best choice. When 22.2 dB < ASNR ≤ 29.2 dB, uncoded 8-PAM should be used. When 29.2 dB < ASNR ≤ 35.9 dB, uncoded 16-PAM is the best choice. When 35.9 dB < ASNR, uncoded 32-PAM is the best choice. In all ranges of SNR, an optimal packet length has been used with each optimal MCS.

Appendix

We have

Let , and we deduce

We deduce

For , we can write

Combining equations (A.5)–(A.7), we obtain (20).

Data Availability

No data were used to support this study.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

References

I. B. Djordjevic, “Adaptive modualtion and coding for free space optical channel,” Journal of Optical Communications and Networking, vol. 2, no. 5, p. 221, 2010.
View at: Publisher Site | Google Scholar
E. Zedini, A. Chelly, and M. S. Alouini, “On the performance analysis of hybrid ARQ with incremental redundancy and with code combining over free space optical channels with pointing errors”,” IEEE Photonics Journal, vol. 6, no. 4, pp. 1–18, 2014.
View at: Publisher Site | Google Scholar
E. Zedini, A. Chelly, and M. S. Alouini, “Unified performance analysis of hybrid arq with incremental redundancy over free space optical channels,” in Proceedings of the 2014 IEEE 25th Annual International Symposium on Personal, Indoor, and Mobile Radio Communication (PIMRC2014), Bologna, Italy, September 2014.
View at: Publisher Site | Google Scholar
S. Aghajanzadeh and M. Uysal, “Outage analysis of hybrid arq protocols in coherent free space optical communications,” in Proceedings of the 2011 IEEE 22nd International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC), Toronto, Canada, 2011.
View at: Google Scholar
S. Aghajanzadeh and M. Uysal, “Information theoretic analysis of hybrid ARQ protocols in coherent free space optical communications,” IEEE Transactions on Communications, vol. 60, no. 5, pp. 1432–1442, 2012.
View at: Publisher Site | Google Scholar
M. R. Bhatnagar and M. K. Arti, “Performance analysis of hybrid satellite-terrestrial FSO cooperative system,” IEEE Photonics Technology Letters, vol. 25, no. 22, pp. 2197–2200, 2013.
View at: Publisher Site | Google Scholar
N. A. Mohd Nor, Z. Ghassemlooy, J. Bohata et al., “Experimental investigation of all-optical relay-assisted 10 Gb/s FSO link over the atmospheric turbulence channel,” Journal of Lightwave Technology, vol. 35, no. 1, pp. 45–53, 2017.
View at: Publisher Site | Google Scholar
A. Jaiswal, M. R. Bhatnagar, and V. K. Jain, “Performance evaluation of space shift keying in free-space optical communication,” Journal of Optical Communications and Networking, vol. 9, no. 2, pp. 149–160, 2017.
View at: Publisher Site | Google Scholar
A. A. Farid and S. Hranilovic, “Outage capacity optimization for free-space optical links with pointing errors,” Journal of Lightwave Technology, vol. 25, no. 7, pp. 1702–1710, 2007.
View at: Publisher Site | Google Scholar
M. R. Bhatnagar and Z. Ghassemlooy, “Performance analysis of gamma-gamma fading FSO MIMO links with pointing errors,” Journal of Lightwave Technology, vol. 34, no. 9, pp. 2158–2169, 2016.
View at: Publisher Site | Google Scholar
E. Zedini and M.-S. Alouini, “Multihop communications over CSI-assisted relay IM/DD FSO systems with pointing errors,” in Proceedings of the IEEE International Conference on Communication Workshop (ICCW), pp. 937–942, 2015.
View at: Publisher Site | Google Scholar
L. Yang, X. Gao, and M.-S. Alouini, “Performance analysis of relay-assisted all-optical FSO networks over strong atmospheric turbulence channels with pointing errors”,” Journal of Lightwave Technology, vol. 32, no. 23, pp. 4613–4620, 2014.
View at: Publisher Site | Google Scholar
M. R. Bhatnagar, “Average BER analysis of relay selection based decode-and-forward cooperative communication over gamma-gamma fading FSO links,” in Proceedings of the IEEE International Conference on Communications (ICC), pp. 3142–3147, Budapest, Hungary, 2013.
View at: Publisher Site | Google Scholar
S. Anees and M. R. Bhatnagar, “Performance of an amplify-and-forward dual-hop asymmetric RF-FSO communication system,” Journal of Optical Communications and Networking, vol. 7, no. 2, pp. 124–135, 2015.
View at: Publisher Site | Google Scholar
I. S. Ansari, F. Yilmaz, and M.-S. Alouini, “Impact of pointing errors on the performance of mixed RF/FSO dual-hop transmission systems,” IEEE Wireless Communications Letters Year, vol. 2, no. 3, pp. 351–354, 2013.
View at: Publisher Site | Google Scholar
E. Lee, J. Park, D. Han, and G. Yoon, “Performance analysis of the asymmetric dual-hop relay transmission with mixed RF/FSO links,” IEEE Photonics Technology Letters, vol. 23, no. 21, pp. 1642–1644, 2011.
View at: Publisher Site | Google Scholar
B. He and R. Schober, “Bit-interleaved coded modulation for hybrid RF/FSO systems,” IEEE Transactions on Communications, vol. 57, no. 12, pp. 3753–3763, 2009.
View at: Publisher Site | Google Scholar
V. V. Mai, T. C. Thang, and A. T. Pham, “Performance analysis of TCP over free space optical links with ARQ-SR,” in Proceedings of the 18-th European Conference on Network and Optical Communications, Graz, Austria, July 2013.
View at: Publisher Site | Google Scholar
G. Prakash and M. Kulkarmi, “On the performance of luby transform codes over selective repeat ARQ in turbulent free space optical links,” in Proceedings of the IEEE International Conference on Computational Science and Engineering, Sydney, Australia, 2013.
View at: Google Scholar
K. Fatima, S. Sheikh Muhammad, and E. Leitgeb, “Adaptive coded modulation for FSO links,” in Proceedings of the 8th International Symposium on Communication Systems, Networks Digital Signal Processing (CSNDSP), pp. 1–4, Poznan, Poland, July 2012.
View at: Publisher Site | Google Scholar
A. Jaiswal, V. K. Jain, and S. Kar, “Adaptive coding and modulation (ACM) technique for performance enhancement of FSO Link,” in Proceedings of the IEEE First International Conference on Control, Measurement and Instrumentation (CMI), pp. 53–57, Kolkata, India, 2016.
View at: Publisher Site | Google Scholar
G.-H. Gho and J. M. Kahn, “Rate-adaptive modulation and low-density parity-check coding for optical fiber transmission systems”,” IEEE/OSA Journal of Optical Communications and Networking, vol. 4, no. 10, pp. 760–768, 2012.
View at: Publisher Site | Google Scholar
G.-H. Gho and J. M. Kahn, “Rate-adaptive modulation and coding for optical fiber transmission systems,” Journal of Lightwave Technology, vol. 30, no. 12, pp. 1818–1828, 2012.
View at: Publisher Site | Google Scholar
J. L. Gao, “A study of adaptive coding and modulation over free space optical link using OCTL data,” in Proceedings of the 2016 IEEE Aerospace Conference, pp. 1–7, Big Sky, MT, USA, 2016.
View at: Publisher Site | Google Scholar
Y. Zeng, R. J. Green, and M. S. Leeson, “Adaptive pulse amplitude and position modulation for optical wireless channels,” in Proceedings of the 2nd IET International Conference on Access Technologies, pp. 13–16, Cambridge, UK, June 2006.
View at: Publisher Site | Google Scholar
S. Hasan Hariq, N. Odabasioglu, and M. Uysal, “An adaptive modulation scheme for coded free-space optical systems,” in Proceedings of the 19th European Conference on Networks and Optical Communications—(NOC), pp. 132–135, Milan, Italy, June 2014.
View at: Publisher Site | Google Scholar
I. S. Ansari, F. Yilmaz, and M.-S. Alouini, “On the performance of mixed RF/FSO variable gain dual-hop transmission systems with pointing errors,” in Proceedings of the IEEE 78th Vehicular Technology Conference (VTC Fall), Las Vegas, NV, USA, 2013.
View at: Google Scholar
H. G. Sandalidis, T. A. Tsiftsis, G. K. Karagiannidis, and M. Uysal, “BER performance of FSO links over strong atmospheric turbulence channels with pointing errors,” IEEE Communications Letters, vol. 12, no. 1, pp. 44–46, 2008.
View at: Publisher Site | Google Scholar
J. Park, E. Lee, and G. Yoon, “Average bit-error rate of the Alamouti scheme in gamma-gamma fading channels,” IEEE Photonics Technology Letters, vol. 23, no. 4, pp. 269–271, 2011.
View at: Publisher Site | Google Scholar
Y. Xi, A. Burr, J. Wei, and D. Grace, “A general upper bound to evaluate packet error rate over quasi-static fading channels,” IEEE Transactions on Wireless Communications, vol. 10, no. 5, pp. 1373–1377, 2011.
View at: Publisher Site | Google Scholar
J. G. Proakis, Digital Communications, McGraw-Hill, New York, NY, USA, 3rd edition, 1995.
H. Boujemaa, “Delay analysis of cooperative truncated HARQ with opportunistic relaying,” IEEE Transactions on Vehicular Technology, vol. 58, no. 9, pp. 4795–4804, 2009.
View at: Publisher Site | Google Scholar
J. A. Anguita, I. B. Djordjevic, M. A. Neifeld, and B. V. Vasic, “Shannon capacities and error-correction codes for optical atmospheric turbulent channels,” Journal of Optical Networking, vol. 4, no. 9, pp. 586–601, 2005.
View at: Publisher Site | Google Scholar
S. Al-Ahmadi and H. Yanikomeroglu, “On the approximation of the generalized-Κ distribution by a gamma distribution for modeling composite fading channels,” IEEE Transactions on Wireless Communications, vol. 9, no. 2, pp. 706–713, 2010.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2019 Ghassan Alnwaimi et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

1368

Downloads

1023

Citations