基于高斯过程回归的光纤非线性信道补偿方法  

Nonlinear Channel Equalization based on Gaussian Processes for Regression in Fiber Link

作　　者：吴彪[1] 李嘉浩 张召才[3] WU Biao;LI Jia-hao;ZHANG Zhao-cai(Wuhan NARI Limitied Company of State Grid Electric Power Research Institute,Wuhan 430074,China;School of Electrical Engineering and Automation,Wuhan University,Wuhan 430072,China;Beijing Institute of Space Science and Technology Information,Beijing 100094,China)

机构地区：[1]国网电力科学研究院武汉南瑞有限责任公司,武汉430074 [2]武汉大学电气与自动化学院,武汉430072 [3]北京空间科技信息研究所,北京100094

出　　处：《光通信研究》2022年第6期35-38,76,共5页Study on Optical Communications

摘　　要：为了降低光纤传输系统中非线性噪声带来的影响,文章提出了基于高斯过程回归(GPR)的非线性信道均衡器(CE),并在强度调制和直接检测光纤链路中进行了实验验证。在文章所提方案中,GPR模型用于估计预处理后的传输符号或相应的非线性噪声。实验结果表明,基于GPR的非线性CE比传统的线性或非线性CE具有更好的性能。实验结果还表明,非线性信道均衡过程中的GPR模型可以理解为具有无限宽度的单层神经网络模型。In order to mitigate the effect of nonlinear noise nonlinear Channel Equalizer(CE) based on Gaussian Processes for Regression(GPR) is proposed and experimentally demonstrated in an intensity modulation and direct detection fiber link. In this scheme, the GPR model is used to estimate the transmitted symbols or the corresponding nonlinear noise after pre-processing. The experimental results show that the nonlinear CE based on GPR has better performance than conventional linear and nonlinear filter-based CE. In addition, it is shown that the GPR model in the nonlinear channel equalization process can be understood as an optimized single-layer neural network model with infinite width.

关 键 词：高斯过程回归 信道均衡器 单层神经网络 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]