Optimization Algorithms of Neural Networks for Traditional Time-Domain Equalizer in Optical Communications
Source abstract
Abstract
Neural networks (NNs) have been successfully applied to channel equalization for optical communications. In optical fiber communications, the linear equalizer and the nonlinear equalizer with traditional structures might be more appropriate than NNs for performing real-time digital signal processing, owing to its much lower computational complexity. However, the optimization algorithms of NNs are useful in many optimization problems. In this paper, we propose and evaluate the tap estimation schemes for the equalizer with traditional structures in optical fiber communications using the optimization algorithms commonly used in the NNs. The experimental results show that adaptive moment estimation algorithm and batch gradient descent method perform well in the tap estimation of equalizer. In conclusion, the optimization algorithms of NNs are useful in the tap estimation of equalizer with traditional structures in optical communications.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Optical Network Technologies · Blind Source Separation Techniques · Neural Networks and Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Electronic Engineering, College of Information Science and Technology, Jinan University, Guangzhou 510632, China
Huawei Technologies Duesseldorf GmbH, European Research Center, 80992 Munich, Germany
Department of Electronic and Information Engineering, The Hong Kong Polytechnic University, Hong Kong, China
State Key Laboratory of Optoelectronic Materials and Technologies, School of Electronics and Information Technology, Sun Yat-sen University, Guangzhou 510275, China
Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Zhuhai 519082, China
Funding
The Science and Technology Planning Project of Guangdong Province · Local Innovation and Research Teams Project of Guangdong Pearl River Talents Program · National Science Foundation of China (NSFC) · The Fundamental Research Funds for the Central Universities
References (30)
Show source-provided references
1. Deep learning10.1038/nature14539 ↗
2. Deep learning in neural networks: An overview10.1016/j.neunet.2014.09.003 ↗
3. Artificial neural network nonlinear equalizer for coherent optical OFDM10.1109/LPT.2014.2375960 ↗
4. Radial basis function neural network nonlinear equalizer for 16-QAM coherent optical OFDM10.1109/LPT.2016.2601901 ↗
5. Ye, C., Zhang, D., Huang, X., Feng, H., and Zhang, K. (2017, January 17–21). Demonstration of 50Gbps IM/DD PAM4 PON over 10GHz class optics using neural network based nonlinear equalization. Proceedings of the 2017 European Conference on Optical Communication (ECOC), Gothenburg, Sweden.10.1109/ECOC.2017.8346196 ↗
6. Gaiarin, S., Pang, X., Ozolins, O., Jones, R.T., Da Silva, E.P., Schatz, R., Westergren, U., Popov, S., Jacobsen, G., and Zibar, D. (2016, January 2–5). High speed PAM-8 optical interconnects with digital equalization based on neural network. Proceedings of the 2016 Asia Communications and Photonics Conference (ACP), Wuhan, China.10.1364/ACPC.2016.AS1C.1 ↗
7. A nonlinear ANN equalizer with mini-batch gradient descent in 40Gbaud PAM-8 IM/DD system10.1016/j.yofte.2018.09.015 ↗
8. Effective denoising and adaptive equalization of indoor optical wireless channel with artificial light using the discrete wavelet transform and artificial neural network10.1109/JLT.2009.2024432 ↗
9. Visible light communications: 170 Mb/s using an artificial neural network equalizer in a low bandwidth white light configuration10.1109/JLT.2014.2314635 ↗
10. Modulation format identification in heterogeneous fiber-optic networks using artificial neural networks10.1364/OE.20.012422 ↗
11. Fiber echo state network analogue for high-bandwidth dual-quadrature signal processing10.1364/OE.27.002387 ↗
12. 112-Gb/s SSB 16-QAM signal transmission over 120-km SMF with direct detection using a MIMO-ANN nonlinear equalizer10.1364/OE.27.012794 ↗
13. Applying neural networks in optical communication systems: Possible pitfalls10.1109/LPT.2017.2755663 ↗
14. Bottou, L., and Bousquet, O. (2007, January 3–8). The tradeoffs of large scale learning. Proceedings of the Neural Information Processing Systems Conference, Vancouver, BC, Canada.
15. A few useful things to know about machine learning10.1145/2347736.2347755 ↗
16. Adaptive subgradient methods for online learning and stochastic optimization
17. Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude
18. Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv.
19. First-and second-order methods for learning: between steepest descent and Newton’s method10.1162/neco.1992.4.2.141 ↗
20. Xiong, X., and De la Torre, F. (2013, January 23–28). Supervised descent method and its applications to face alignment. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Portland, OR, USA.10.1109/CVPR.2013.75 ↗
21. Ruder, S. (2016). An overview of gradient descent optimization algorithms. arXiv.
22. OFC 2004 workshop on optical and electronic mitigation of impairments10.1109/JLT.2004.840342 ↗
23. Performance of single-mode fiber links using electronic feed-forward and decision feedback equalizers10.1109/LPT.2005.856326 ↗
24. Haykin, S.S. (2005). Adaptive Filter Theory, Pearson Education India.
25. Wang, H., Zhou, J., Li, F., Liu, L., Yu, C., Yi, X., Huang, X., Liu, W., and Li, Z. (2019, January 5–10). Variable-step DD-FTN algorithm for PAM8-based short-reach optical interconnects. Proceedings of the CLEO: Science and Innovations, San Jose, CA, USA. paper SW4O.10.1364/CLEO_SI.2019.SW4O.3 ↗
26. Joint FDE and MLSD Algorithm for 56-Gbit/s Optical FTN-PAM4 System Using 10G-Class Optics10.1109/JLT.2019.2915386 ↗
27. Approaching Nyquist limit in WDM systems by low-complexity receiver-side duobinary shaping10.1109/JLT.2012.2190972 ↗
28. Digital signal processing for short-reach optical communications: A review of current technologies and future trends10.1109/JLT.2018.2793881 ↗
29. Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep Learning, MIT Press.
30. An algorithm for least-squares estimation of nonlinear parameters10.1137/0111030 ↗