短距光纤通信系统中基于神经网络的非线性均衡器研究和实现

赵晗祺; 李娜; 吴斌; 吴桂龙; 陈一童; 冯晓芳; 何沛礼; 李蔚

doi:10.37188/CO.2024-0114

短距光纤通信系统中基于神经网络的非线性均衡器研究和实现

doi: 10.37188/CO.2024-0114

1.
中国南方电网电力调度控制中心通信处, 广州510663
2.
华中科技大学武汉光电国家研究中心, 武汉 430074

基金项目: 南方电网科技项目面向新型电力系统的骨干通信承载网关键技术研究与示范应用资助（No. CG0000022001594227）

详细信息

作者简介:
赵晗祺（1973—），女，广州人，中国南方电网电力调度控制中心通信处高级主管，从事电力光通信专业管理、工程建设、安全运行、科技研究、标准化、通信运行评价等工作二十余年，曾获南方电网科技进步奖、成果转化奖、专利奖、技术标准创新贡献奖等10余项奖励。主持完成南网电网OTN、ASON等多个网级光通信网建设。E-mail：zhaohq@csg.cn

李蔚（1968—），女，湖北武汉人，教授，博士，博士生导师，长期从事光纤通信技术和光电子技术方面的研究和教学工作。为第一负责人先后多项国家自然科学基金，国家“973”项目子课题，国家“863”项目等多项国家级项目

中图分类号: TP394.1;TH691.9
计量
- 文章访问数: 110
- HTML全文浏览量: 56
- PDF下载量: 9
- 被引次数: 0
出版历程
- 收稿日期: 2024-06-21
- 录用日期: 2024-08-30
- 网络出版日期: 2024-09-25

Nonlinear equalizer based on neural network in high-speed optical fiber communication systems

1.
CSG Power Dispatching and Control Center, China Southern Power Grid Company, Ltd., Guangzhou 510530, P.R.China
2.
Wuhan National Laboratory for Optoelectronics, Huazhong University of Science and Technology, Wuhan 430074, P.R. China

Funds: Supported by Southern Power Grid Technology Project Funding for Key Technology Research and Demonstration Application of Backbone Communication Carrier Network for New Power Systems (No. CG0000022001594227)

More Information

Corresponding author: XXX

摘要

摘要:
为了实现对短距光纤数据通信系统接收端非线性损伤的低复杂度均衡，提出了一种基于全连接神经网络的接收端均衡算法，这是一种引入判决反馈结构的判决反馈神经网络。非线性畸变由线性工作区与实验系统不匹配的光电探测器引入，在此基础上实现了基于C波段直接调制激光器的56 Gbit/s PAM4信号的20 km传输验证实验，并对判决反馈神经网络和其他均衡方案的均衡性能进行了对比实验。实验结果表明，相比全连接神经网络，改进方案实现了20 km传输时2 dB的灵敏度提升。改进方案可以很好地均衡光电器件的非线性，且计算复杂度更低，具有很好的应用意义。
- 短距光通信 /
- 光电器件非线性畸变 /
- 信号均衡 /
- 神经网络 /
- 判决反馈
Abstract:
In order to achieve low complexity balancing of nonlinear damage at the receiver of short-range fiber optic data communication systems, we propose an equalization structure named Decision Feedback Neural Network which introduce the Decision Feedback Structure into the Fully Connected Neural Network. The nonlinear distortion is introduced by using a photodetector with a linear working area that does not match the experimental system. The experimental system is built based on a 56 Gbit/s PAM4 with a C-band direct-modulated laser, and we compare the equalization performance of decision feedback neural network with other equalization schemes. Experimental results show that compared with the fully connected neural network, the improved scheme achieves a sensitivity improvement of 2 dB at 20 km transmission, and the equalization performance is close to the convolutional neural network with lower complexity. This paper has great significance for the rate and capacity upgrade of short-distance optical fiber communication system, and can be used as a reference for further scientific research and industrial application.
- short-reach optical communication /
- nonlinear distortion of optoelectronic devices /
- signal equalization /
- neural network /
- decision feedback

HTML全文

图 1 全连接神经网络均衡器示意图

Figure 1. Diagram of fully connected neural network equalizer

下载: 全尺寸图片幻灯片

图 2 判决反馈神经网络均衡器示意图

Figure 2. Diagram of decision feedback neural network equalizer

下载: 全尺寸图片幻灯片

图 3 56 Gbit/s PAM4 IM/DD系统实验装置

Figure 3. Experimental setup 56 Gbit/s PAM4 IM/DD system

下载: 全尺寸图片幻灯片

图 4 背靠背传输下均衡器性能对比

Figure 4. Comparison of the performance of the equalizers under back-to-back transmission

下载: 全尺寸图片幻灯片

图 5 各均衡器性能对比（a）10 km；（b）20 km；（c）30 km

Figure 5. Comparison of the performance of the equalizers under (a)10 km; (b)20 km; (c)30 km transmission

下载: 全尺寸图片幻灯片

图 6 眼图（a）无均衡；（b）VNLE；（c）CNN；（d）FCNN；（e）DFNN

Figure 6. Eye diagram (a) without equalization; (b) VNLE; (c) CNN; (d) FCNN; (e) DFNN

下载: 全尺寸图片幻灯片

参考文献(22)

[1]	CHAGNON M. Optical communications for short reach[C]. Proceedings of 2018 European Conference on Optical Communication (ECOC), IEEE, 2018: 1-3.
[2]	LI D, DENG L, YE Y, et al. 4×96 Gbit/s PAM8 for short-reach applications employing low-cost DML without pre-equalization[C]. Proceedings of 2019 Optical Fiber Communications Conference and Exhibition (OFC), IEEE, 2019: 1-3.
[3]	ZHANG F, FANG X S, CHEN X Y. Neural network-based fiber nonlinearity mitigation in high-speed coherent optical transmission systems[C]. Proceedings of 2022 Optical Fiber Communications Conference and Exhibition (OFC), IEEE, 2022: 1-3.
[4]	ALMUFTI A M, MOROZOV O G. 1Tbit/s per lambda high order quadrature amplitude modulation (128-256QAM) coherent optical transmission system design to support (5G+)[C]. Proceedings of 2023 Systems of Signal Synchronization, Generating and Processing in Telecommunications (SYNCHROINFO), IEEE, 2023: 1-4.
[5]	HARSTEAD E, VAN VEEN D, HOUTSMA V, et al. Technology roadmap for time-division multiplexed passive optical networks (TDM PONs)[J]. Journal of Lightwave Technology, 2019, 37(2): 657-664.
[6]	王振宇, 付秀华, 林兆文, 等. 星间通信系统高精度分光镜的研制[J]. 中国光学(中英文), 2024, 17(2): 334-341. WANG ZH Y, FU X H, LIN ZH W, et al. Development of high-precision beam splitter for inter-satellite communication system[J]. Chinese Optics, 2024, 17(2): 334-341. (in Chinese).
[7]	杨秀清, 陈海燕. 光通信技术在物联网中的应用[J]. 中国光学(中英文), 2014, 7(6): 889-896. YANG X Q, CHEN H Y. Application of optical communication technique in the Internet of Things[J]. Chinese Optics, 2014, 7(6): 889-896. (in Chinese).
[8]	ZHOU X, URATA R, LIU H. Beyond 1 Tb/s intra-data center interconnect technology: IM-DD OR coherent?[J]. Journal of Lightwave Technology, 2020, 38(2): 475-484. doi: 10.1109/JLT.2019.2956779
[9]	MANILOFF E, GAREAU S, MOYER M. 400G and beyond: coherent evolution to high-capacity inter data center links[C]. Proceedings of 2019 Optical Fiber Communications Conference and Exhibition (OFC), IEEE, 2019: 1-3.
[10]	PANG X D, UDALCOVS A, SCHATZ R, et al. High-speed short reach optical communications: technological options and challenges[C]. Proceedings of 2020 Asia Communications and Photonics Conference (ACP) and International Conference on Information Photonics and Optical Communications (IPOC), IEEE, 2020: 1-3.
[11]	张帆, 朱逸萧. 面向数据中心光互连的高速光传输技术[J]. 中兴通讯技术, 2019, 25(5): 17-24. ZHANG F, ZHU Y X. High-speed transmission technologies for data center optical interconnection[J]. ZTE Technology Journal, 2019, 25(5): 17-24. (in Chinese).
[12]	ZHONG K P, ZHOU X, GAO Y L, et al. 140-Gb/s 20-km transmission of PAM-4 signal at 1.3 μm for short reach communications[J]. IEEE Photonics Technology Letters, 2015, 27(16): 1757-1760. doi: 10.1109/LPT.2015.2439571
[13]	YANG CH, HU R, LUO M, et al. IM/DD-based 112-Gb/s/lambda PAM-4 transmission using 18-Gbps DML[J]. IEEE Photonics Journal, 2016, 8(3): 7903907.
[14]	MATSUMOTO K, YOSHIDA Y, MARUTA A, et al. On the impact of Tomlinson-Harashima precoding in optical PAM transmissions for intra-DCN communication[C]. Proceedings of 2017 Optical Fiber Communications Conference and Exhibition (OFC), IEEE, 2017: 1-3.
[15]	GAO F, ZHOU SH W, LI X, et al. 2 × 64 Gb/s PAM-4 transmission over 70 km SSMF using O-band 18G-class directly modulated lasers (DMLs)[J]. Optics Express, 2017, 25(7): 7230-7237. doi: 10.1364/OE.25.007230
[16]	YU Y K, BO T W, CHE Y, et al. Low-complexity nonlinear equalizer based on absolute operation for C-band PAM signal generated by using directly modulated laser[C]. Proceedings of 2020 Opto-Electronics and Communications Conference (OECC), IEEE, 2020: 1-3.
[17]	ZHANG J W, LIN ZH R, WU X, et al. Low-complexity sparse absolute-term based nonlinear equalizer for C-band IM/DD systems[J]. Optics Express, 2021, 29(14): 21891-21901. doi: 10.1364/OE.425896
[18]	CHEN Y B, WANG Y, LI W, et al. 106 Gbit/s PAM4 transmission employing a 15 GHz directly modulated laser[C]. Proceedings of 2021 Asia Communications and Photonics Conference (ACP), IEEE, 2021: 1-3.
[19]	ESTARAN J, RIOS-MUELLER R, MESTRE M A, et al. Artificial neural networks for linear and non-linear impairment mitigation in high-baudrate IM/DD systems[C]. Proceedings of the ECOC 2016; 42nd European Conference on Optical Communication, IEEE, 2016: 1-3.
[20]	LI P X, YI L L, XUE L, et al. 56 Gbps IM/DD PON based on 10G-class optical devices with 29 dB loss budget enabled by machine learning[C]. Proceedings of 2018 Optical Fiber Communications Conference and Exposition (OFC), IEEE, 2018: 1-3.
[21]	YE CH J H, ZHANG D X, HU X F, et al. Recurrent Neural Network (RNN) based end-to-end nonlinear management for symmetrical 50Gbps NRZ PON with 29dB+ loss budget[C]. Proceedings of 2018 European Conference on Optical Communication (ECOC), IEEE, 2018: 1-3.
[22]	XU ZH P, SUN CH B W, JI T H, et al. Cascade recurrent neural network enabled 100-Gb/s PAM4 short-reach optical link based on DML[C]. Proceedings of 2020 Optical Fiber Communications Conference and Exhibition (OFC), IEEE, 2020: 1-3.