Machine learning methods for optical communications

Introduction

Radio over Fiber (RoF) technology has been realized in different forms ranging from analog to more complex forms [1-6]. The enhancement of capacity and wireless coverage has posed significant challenges to the existing optical and wireless access networks. Machine Learning (ML) methods have given a new direction to meet the ever-increasing challenges in fiber-optic communications. Since, ML-based methods are well known to perform exceptionally well in scenarios where it is too difficult to explicitly describe the underlying physics and mathematics of the problem and the numerical procedures available require significant computational resources/time.

Before the introduction of ML in this field, the reduction of nonlinearities was carried out through different methodologies that were proposed during last 10 years via Optical and Electrical domain methods, with latter getting more attention. Digital Predistortion (DPD) has gained immense attention and therefore it is still looked to be an important methodology [1,7-9]. However, these DPD based linearization techniques are not so straight forward and are rather complex, time consuming and adds extra over heads [10]. Similarly, these methodologies require an extra feedback mechanism which is achievable but increases the complexity of the system. Therefore, reducing such nonlinearities in optical communication links effectively is still a bottle neck in efficient deployment of optical communication systems.

Recent advancements in ML technology have further motivated the researchers to explore true potential of this emerging field in optical communication systems. Firstly, we review some significant research ideas pertaining to the use of ML algorithms in fiber-optic communications.

Then, a experimental study is presented for the mitigation of nonlinear impairments by comparing Reinforcement Learning (RL) based machine learning method with Support Vector Machine (SVM) method and conventional methods. In the proposed system, 10-Gb/s with 256 quadrature amplitude modulation (QAM) signal is injected into distributed feedback (DFB) laser for 50 km single mode fiber transmission. It causes signal to suffer from nonlinearities due to opto-electronic devices in the RoF link as the ideal decision boundary is no longer linear.

Machine learning applications in optical communication system

The application of optical communication systems can be divided with three main subcategories:

i. Nonlinearity Mitigation

ii. Performance Monitoring

iii. Smart Decision/ Prevention

The mitigation of nonlinearities in optical communications with ML is the most important application these days. Recently, RL-SARSA based ML method has been evaluated in [11-12] while SVM and K-Nearest Neighbours (KNN) based ML methods have been utilized too in [13-15]. SBP method has also been evaluated for reducing the impairments in optical communication systems [16].

Optical communication researchers are no strangers to regressions and classifications. Over the last decade, coherent detection and digital signal processing (DSP) techniques have been the cornerstone of optical transceivers in fiber-optic communication systems. Advanced modulation formats such as 256 quadrature amplitude modulation (256-QAM) and above together with DSP-based estimation and compensation of various transmission impairments such as laser phase noise have become the key drivers of innovation. In this context, parameter estimation and symbol detection are naturally regression and classification problems. Currently, most of these parameter estimation and decision rules are derived from probability theory and adequate understanding of the problem’s underlying physics. As high-capacity optical transmission links are increasingly being limited by transmission impairments such as fiber nonlinearity, explicit statistical characterizations of inputs/outputs become difficult. The maximum likelihood decision boundaries in this case are curved and virtually impossible to derive analytically. Consequently, there has been an increasing amount of research on the application of ML techniques for fiber nonlinearity compensation (NLC). Another related area where ML flourishes is short reach direct detection systems that are affected by chromatic dispersion (CD), laser chirp, and other transceiver components imperfections, which render the overall communication system hard to analyze.

The other important application of ML is monitoring the performance of optical communication system. Many methodologies have been evaluated among which SVM [17], ANN [18] and Kernel based methods are most significant [19]. Recently, Parzen window method to mitigate the fiber nonlinearity in the context of dispersion managed and dispersion unmanaged systems has been proposed as well [20].

Finally, the other important application of ML method is smart decision in order to prevent any errors by fault detection. Reliable operation of an optical network requires incorporation of an early protection mechanism into the network. Recently, several ML-based techniques have been developed for cognitive fault detection/prevention in fiber-optic networks. A combination of Double Exponential Smoothing (DES) and SVM for network equipment failure prediction has been employed in [20]. In [17,21,22], an ANN is trained to learn historical fault patterns in networks and is subsequently used for detecting significant network faults with much better accuracies and proactive reaction times as compared to traditional threshold-based fault detection methods. The ML methodologies can be implemented in recent real time S-DRoF systems as explained in [23].

ML techniques offer the opportunity to optimize the design of individual physical components as well as complete end-to-end fiber-optic communication systems. Recently we have seen some noticeable research works in this regard with quite encouraging results. In [24], a complete optical communication system including transmitter, receiver and nonlinear channel is modeled as an end-to-end fully-connected DNN. This approach enables the optimization of transceivers in a single end-to-end process where the transmitter learns waveform representations that are robust to channel impairments while the receiver learns to equalize channel distortions. The results for 42 Gb/s intensity modulation/direct detection (IM/DD) systems show that the DL-based optimization outperforms the solutions based on two- and four-level pulse amplitude modulation (PAM2/PAM4) and conventional receiver equalization, for a range of transmission distances. Jones, et al. [25] proposed an ANN-based receiver for Nonlinear Frequency-Division Multiplexing (NFDM) optical communication systems. Unlike standard nonlinear Fourier transform (NFT) based receivers which are vulnerable to losses and noise in NFDM systems, the ANN-based receiver tackles these impairments by learning the distortion characteristics of previously transmitted pulses and applying them for inference for future decisions. The results demonstrate improved BER performance as compared to conventional NFT-based receivers for practical link configurations. In Ref. [26], an ANN is used in receiver DSP for mitigating linear and nonlinear impairments in IM/DD systems. The ANN infers linear and nonlinear channel responses simultaneously, which are then exploited for increasing the demodulation reliability beyond the capability of linear equalization techniques. Using an ANN along with standard feed-forward equalization (FFE), up to 10 times BER improvement over FFE-only configurations is demonstrated for 84 GBd/s PAM4 transmission over 1.5 km SSMF.

Figure 1 presents the summary of the applications of ML methods in optical communication systems.

Figure 1: Key applications of ML methods in optical fiber systems.

Role of Machine Learning in Future

Looking to the future, we can foresee a vital role played by ML-based mechanisms across several diverse functional areas in optical networks, for example, network planning and performance prediction, network maintenance and fault prevention, network resources allocation and management, etc. ML can also aid cross-layer optimization in future optical networks requiring big data analytics, since it can inherently learn and uncover hidden patterns and unknown correlations in big data, which can be extremely beneficial in solving complex network optimization problems. The ultimate objective of ML-driven next-generation optical networks will be to provide infrastructures that can monitor themselves, diagnose and resolve their problems, and provide intelligent and efficient services to the end users [24].

Conclusion

The article provides a recent applications of ML methods in optical communication systems. The survey presents the categories of applications where ML has been successfully being employed. Indeed, the ML methods proposed are quite useful in mitigating the nonlinearities of optical communication systems.

1. Hadi  MU, Nanni J, Polleux JL, Traverso PA, Tartarini G (2019) Direct digital predistortion technique for the compensation of laser chirp and fiber dispersion in long haul radio over fiber links. Opt Quant Electron 51: 205. Link: https://bit.ly/33xdbiN  
2. Hadi MU, Jung H, Ghaffar S, Traverso PA, Tartarini G (2019) Optimized digital radio over fiber system for medium range communication. Optics Communications 443: 177-185. Link: https://bit.ly/35HW7sU
3. Hadi M, Jung H, Traverso P, Tartarini G (2019) Digital Radio Frequency Transport over Optical Fiber for 5G Fronthaul Links. Journal of Optical Communications. Link: https://bit.ly/3krWQmp  
4. Hadi M, Hadi M, Aslam N, Ali R, Khurshid K (2019) Experimental Demonstration of MASH Based Sigma Delta Radio over Fiber System for 5G C-RAN Downlink. Journal of Optical Communications. Link: https://bit.ly/32AQ0VC
5. Hadi MU, Jung H, Traverso PA, Tartarini G (2020) Experimental evaluation of real-time sigma-delta radio over fiber system for fronthaul applications. International Journal of Microwave and Wireless Technologies 1–10. Link: https://bit.ly/3kmMfJh
6. Hadi MU, Traverso PA, Tartarini G, Jung H (2019) Experimental characterization of Sigma Delta Radio over fiber system for 5G C-RAN downlink. ICT Express 6: 23-27. Link: https://bit.ly/2FD7FTs
7. Hadi MU (2015) Extending the benefits of LTE to unlicensed spectrum. 2015 International Conference on Information and Communication Technologies (ICICT), Karachi 1-3. Link: https://bit.ly/32Ba7TG
8. Hadi MU, Nanni J, Traverso PA, Tartarini G, Venrad O, et al. (2018) Experimental evaluation of digital predistortion for VCSEL-SSMF-based Radio-over-Fiber link. 2018 International Topical Meeting on Microwave Photonics (MWP), Toulouse 1-4. Link: https://bit.ly/3mthvby
9. Hadi MU, Traverso PA, Tartarini G, Venard O, Baudoin G, et al. (2018) Digital Predistortion for Linearity Improvement of VCSEL-SSMF-Based Radio-Over-Fiber Links. In IEEE Microwave and Wireless Components Letters. Link: https://bit.ly/3moRbPD
10. Hadi MU, Nanni J, Traverso PA, Tartarini G, Venard O, et al. (2019) Linearity improvement of VCSELs based radio over fiber systems utilizing digital predistortion. Adv Sci Technol Eng Syst J 4: 156-163. Link: https://bit.ly/2RyMBAm
11. Hadi MU (2019) RL-SARSA Machine Learning Based Analog Radio over Fiber System. Preprints. Link: https://bit.ly/3hCNjHl
12. Aslam N, Xia K, Hadi MU (2019) Optimal Wireless Charging Inclusive of Intellectual Routing Based on SARSA Learning in Renewable Wireless Sensor Networks. IEEE Sensors Journal 19: 8340-8351.  Link: https://bit.ly/3iBAlut
13.  Hadi MU (2019) Nonlinearities Diminution in 40 Gb/s 256 QAM Radio over Fiber Link via Machine Learning Method. Preprints. Link: https://bit.ly/2Rwod27
14. Li M, Yu S, Yang J, Chen Z, Han Y, et al. (2013) Nonparameter nonlinear phase noise mitigation by using M-ary support vector machine for coherent optical systems. IEEE Photonics Journal 5.
15.  Wang Z, Zhang M, Wang D, Song C, Liu M, et al. (2017) Failure prediction using machine learning and time series in optical network. Opt Exp 25. Link: https://bit.ly/2ZHhkQ5
16. Jiang N, Gong Y, Karout J, Wymeersch H, Johannisson P, et al. (2011) Stochastic backpropagation for coherent optical communications. Proc ECOC Geneva 81. Link: https://bit.ly/2RwSnSR
17. Skoog RA, Banwell TC, Gannett JW, Habiby SF, Pang M, et al. (2006) Automatic identification of impairments using support vector machine pattern classification on eye diagrams. IEEE Photon Technol Lett 18: 2398-2400. Link: https://bit.ly/32Adni1
18.  Khan FN, Yu Y, Tan MC, Yu C, Lau APT, et al. (2015) Simultaneous OSNR monitoring and modulation format identification using asynchronous single channel sampling. Proc Asia Communications and Photonics Conf Hong Kong. Link: https://bit.ly/33u7H8q
19. Anderson TB, Kowalczyk A, Clarke K, Dods SD, Hewitt D, et al. (2009) Multi impairment monitoring for optical networks. J Lightwave Technol 27: 3729-3736. Link: https://bit.ly/33xFU7c
20. Amari A, Lin X, Dobre OA, Venkatesan R, Alvarado A (2019) A Machine Learning-Based Detection Technique for Optical Fiber Nonlinearity Mitigation. IEEE Photonics Technology Letters 31: 627-630. Link: https://bit.ly/33yBAEv
21. Rafique D, Szyrkowiec T, Grieber H, Autenrieth A, Elbers JP (2018) Cognitive assurance architecture for optical network fault management. J Lightwave Technol 36:  1443–1450. Link: https://bit.ly/2ZEQiZp
22. Hadi M, Jung H, Traverso P, Tartarini G (2020) Experimental evaluation of real-time sigma-delta radio over fiber system for fronthaul applications. International Journal of Microwave and Wireless Technologies 1-10. Link: https://bit.ly/3kA5S0P
23. Hadi M, Aslam N, Jung H (2019) Performance appraisal of sigma delta   modulated radio over fiber system. Journal of Optical Communications. Link: https://bit.ly/35G6jSQ
24. Khan FN, Fan Q, Lu C, Lau APT (2020) Machine learning methods for optical communication systems and networks. Optical Fiber Telecommunications7: 921–978. Link: https://bit.ly/3iELLhd
25. Karanov B, Chagnon M, Thouin F, Eriksson TA, Bulow H, et al. (2018) End-to-end deep learning of optical fiber communications, J. Lightwave Technol 36: 48434855. Link: https://bit.ly/33Cf7qj
26. Jones RT, Gaiarin S, Yankov MP, Zibar D (2018) Time-domain neural network receiver for nonlinear frequency division multiplexed systems. IEEE Photonics Technol Lett 30: 10791082. Link: https://bit.ly/32Apdsz