基于BO-BiGRU的后均衡器在水下高速可见光通信系统中的应用

doi:10.11959/j.issn.1000-0801.2023107

电信科学 ›› 2023, Vol. 39 ›› Issue (5): 11-19.doi: 10.11959/j.issn.1000-0801.2023107

基于BO-BiGRU的后均衡器在水下高速可见光通信系统中的应用

张昊宇, 姚力, 陈超旭, 施剑阳, 沈超, 迟楠

复旦大学电磁波信息科学教育部重点实验室，上海 200438

修回日期:2023-05-08 出版日期:2023-05-20 发布日期:2023-05-01
作者简介:张昊宇（2000- ），男，复旦大学博士生，主要研究方向为人工智能、光通信与光器件
姚力（2000- ），男，复旦大学博士生，主要研究方向为人工智能、光通信与光器件
陈超旭（2000- ），男，复旦大学博士生，主要研究方向为光通信与光器件
施剑阳（1992- ），男，复旦大学在站博士后，主要研究方向为高速光通信系统中的高阶调制解调和均衡技术。入选博士后创新人才支持计划
沈超（1989- ），男，复旦大学青年研究员、博士生导师，IEEE Photonics Journal 副主编， APL Photonics期刊ECEAB委员，主要研究方向为宽带半导体器件设计与工艺、光电子器件与光子集成芯片、半导体激光器、高性能超辐射发光芯片与可见光通信技术
迟楠（1974- ），女，复旦大学信息学院院长、教授、博士生导师，美国光学学会会士（OSA Fellow），主要研究方向为高谱效率多维多阶光调制技术和数字信号处理技术。长期从事高速光通信和高速可见光通信方面的研究，荣获教育部自然科学奖二等奖、中国产学研合作创新成果奖一等奖、中国国际工业博览会创新奖等，发表 SCI 检索论文260余篇、ESI高被引论文4篇，出版专著6部，授权发明专利18项，多项技术入选国家标准和IEEE标准提案
基金资助:
国家重点研发计划项目(2022YFB2802803);国家自然科学基金资助项目(61925104);国家自然科学基金资助项目(62031011);国家自然科学基金资助项目(62201157)

Application of BO-BiGRU based post equalizer in high-speed underwater visible light communication system

Haoyu ZHANG, Li YAO, Chaoxu CHEN, Jianyang SHI, Chao SHEN, Nan CHI

Key Laboratory for Information Science of Electromagnetic Waves, Ministry of Education, Fudan University, Shanghai 200438, China

Revised:2023-05-08 Online:2023-05-20 Published:2023-05-01
Supported by:
The National Key Research and Development Program of China(2022YFB2802803);The National Natural Science Foundation of China(61925104);The National Natural Science Foundation of China(62031011);The National Natural Science Foundation of China(62201157)

摘要/Abstract

摘要：

信号在水下可见光通信（UVLC）信道传输的过程中易受到非线性效应的影响，为了提高系统通信性能，对接收的信号进行均衡是至关重要的。提出了一种基于贝叶斯优化算法的双向门控循环单元（BO-BiGRU）模型作为 UVLC 系统中的后均衡器，其能够自动寻找模型最优超参数，以实现模型的最佳性能。BO-BiGRU模型应用于1.2 m水下实验平台，采用64正交调幅（QAM）-无载波幅相调制（CAPM），在系统的误码率（BER）低于3.8×10^-3的7%前向纠错（FEC）阈值的情况下，实现了3.10 Gbit/s的数据传输速率，比传统的后均衡方法提高了128 Mbit/s。

关键词: 水下可见光通信, 非线性效应, 波形级别后均衡器, 贝叶斯优化, 双向门控循环单元

Abstract:

Signals are susceptible to nonlinear effects in the process of underwater visible light communication (UVLC) channel transmission.In order to improve the communication performance of the system, it is very important to recover the received signals.A bidirectional gated recurrent unit model based on Bayesian optimization algorithm (BO-BiGRU) was proposed as post equalizer in UVLC system, which could automatically find the optimal hyperparameters.BO-BiGRU model was applied to the 1.2 m underwater experimental platform.64 quadrature amplitude modulation (QAM)-carrierless amplitude-and-phase modulation (CAPM) was adopted, and 3.10 Gbit/s rate of data transfer was achieved under the condition that the bit error ratio (BER) of the system was less than 3.8×10^-37% forward error correction (FEC), which is 128 Mbit/s higher than the traditional post equalization method.

Key words: underwater visible light communication, nonlinear effect, waveform level post equalizer, Bayesian optimization, bidirectional gated recurrent unit

中图分类号:

TN911.3，TP18

张昊宇, 姚力, 陈超旭, 施剑阳, 沈超, 迟楠. 基于BO-BiGRU的后均衡器在水下高速可见光通信系统中的应用[J]. 电信科学, 2023, 39(5): 11-19.

Haoyu ZHANG, Li YAO, Chaoxu CHEN, Jianyang SHI, Chao SHEN, Nan CHI. Application of BO-BiGRU based post equalizer in high-speed underwater visible light communication system[J]. Telecommunications Science, 2023, 39(5): 11-19.

图/表 7

图1

图2

图3

图4

表1

表2

图5

参考文献 20

[1]	CHI N , ZHOU Y J , WEI Y R ,et al. Visible light communication in 6G:advances,challenges,and prospects[J]. IEEE Vehicular Technology Magazine, 2020,15(4): 93-102.
[2]	SHI J Y , NIU W Q , LI Z W ,et al. Optimal adaptive waveform design utilizing an end-to-end learning-based pre-equalization neural network in an UVLC system[J]. Journal of Lightwave Technology, 2023,41(6): 1626-1636.
[3]	OUBEI H M , SHEN C , KAMMOUN A ,et al. Light based underwater wireless communications[J]. Japanese Journal of Applied Physics, 2018,57(8S2): 08PA06.
[4]	BURSE K , YADAV R N , SHRIVASTAVA S C . Channel equalization using neural networks:a review[J]. IEEE Transactions on Systems,Man,and Cybernetics,Part C (Applications and Reviews), 2010,40(3): 352-357.
[5]	HAIGH P A , GHASSEMLOOY Z , RAJBHANDARI S ,et al. Visible light communications:170 Mb/s using an artificial neural network equalizer in a low bandwidth white light configuration[J]. Journal of Lightwave Technology, 2014,32(9): 1807-1813.
[6]	KECHRIOTIS G , ZERVAS E , MANOLAKOS E S . Using recurrent neural networks for adaptive communication channel equalization[J]. IEEE Transactions on Neural Networks, 1994,5(2): 267-278.
[7]	CHI N , ZHAO Y H , SHI M ,et al. Gaussian kernel-aided deep neural network equalizer utilized in underwater PAM8 visible light communication system[J]. Optics Express, 2018,26(20): 26700-26712.
[8]	CHI N , HU F C , LI G Q ,et al. AI based on frequency slicing deep neural network for underwater visible light communication[J]. Science China Information Sciences, 2020,63(6): 1-8.
[9]	CHEN H , ZHAO Y H , HU F C ,et al. Nonlinear resilient learning method based on joint time-frequency image analysis in underwater visible light communication[J]. IEEE Photonics Journal, 2020,12(2): 1-10.
[10]	ZHAO Z Y , VURAN M C , GUO F J ,et al. Deep-waveform:a learned OFDM receiver based on deep complex-valued convolutional networks[J]. IEEE Journal on Selected Areas in Communications, 2021,39(8): 2407-2420.
[11]	CHEN H , NIU W Q , ZHAO Y H ,et al. Adaptive deep-learning equalizer based on constellation partitioning scheme with reduced computational complexity in UVLC system[J]. Optics Express, 2021,29(14): 21773-21782.
[12]	LU X Y , LU C , YU W X ,et al. Memory-controlled deep LSTM neural network post-equalizer used in high-speed PAM VLC system[J]. Optics Express, 2019,27(5): 7822-7833.
[13]	SHI J Y , NIU W Q , LI Z W ,et al. Optimal adaptive waveform design utilizing an end-to-end learning-based pre-equalization neural network in an UVLC system[J]. Journal of Lightwave Technology, 2023,41(6): 1626-1636.
[14]	ZHAO Y H , ZOU P , YU W X ,et al. Two tributaries heterogeneous neural network based channel emulator for underwater visible light communication systems[J]. Optics Express, 2019,27(16): 22532-22541.
[15]	GAO Z P , WANG Y H , LIU X D ,et al. FFDNet-based channel estimation for massive MIMO visible light communication systems[J]. IEEE Wireless Communications Letters, 2020,9(3): 340-343.
[16]	YU T , ZHU H . Hyper-parameter optimization:a review of algorithms and applications[J]. arXiv preprint, 2020,arXiv:2003.05689.
[17]	SNOEK J , LAROCHELLE H , ADAMS R P . Practical Bayesian optimization of machine learning algorithms[C]// Proceedings of the 25th International Conference on Neural Information Processing Systems - Volume 2. New York:ACM Press, 2012: 2951-2959.
[18]	CHO H , KIM Y , LEE E ,et al. Basic enhancement strategies when using Bayesian optimization for hyperparameter tuning of deep neural networks[J]. IEEE Access, 2020(8): 52588-52608.
[19]	CHUNG J , GULCEHRE C , CHO K ,et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[J]. arXiv preprint,2014,arXiv:1412. 3555.
[20]	LIU X Y , WANG Y J , WANG X S ,et al. Bi-directional gated recurrent unit neural network based nonlinear equalizer for coherent optical communication system[J]. Optics Express, 2021,29(4): 5923-5933.

参数名称	参数范围	最优参数值
第一隐藏层GRU数	(16,32, 48, 64)	32
第二隐藏层GRU数	(64, 96, 128)	128
优化器	(Adam, SGD)	Adam
学习率	(1.0×10^-3,1.0×10^-4, 1.0×10^-5)	1.0×10^-4

LMS-Volterra参数	参数取值
LMS阶数	35
LMS步长	0.014
Volterra阶数	9
Volterra步长	0.000 2
滑窗长度（输入层节点数）	45
第一隐藏层节点数	128
第二隐藏层节点数	64
输出层节点数	1

基于BO-BiGRU的后均衡器在水下高速可见光通信系统中的应用

Application of BO-BiGRU based post equalizer in high-speed underwater visible light communication system

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 20

相关文章 1

Metrics

推荐阅读 0