Abstract

A decision technique using mixture-of-Gaussian (MoG) clustering algorithms is proposed in the context of a coherent optical communication system. For an 80-Gb/s single-carrier polarization-division multiplexed 16 quadrature amplitude modulation (QAM) transmission system at 975 km, an improvement in $Q$ factors up to 0.41 dB is observed for the entire range of the considered optical signal-to-noise SNR. Compared with the traditional minimum Euclidean distance-based decision, the MoG clustering-based decision achieves a transmission distance increase of 175 km at a $Q$ factor of 9.96 dB. Experiment results show that the proposed decision technique is insensitive to the system’s nonlinear impairments and can effectively improve nonlinear tolerance of the system. We also propose a majorization method for decision-directed least mean square (LMS) using the MoG clustering-based decision algorithm, called MoG-LMS. The performance improvement of the MoG-LMS algorithm is verified by experiments.

© 2019 Optical Society of America

K-means-clustering-based fiber nonlinearity equalization techniques for 64-QAM coherent optical communication system

Junfeng Zhang, Wei Chen, Mingyi Gao, and Gangxiang Shen
Opt. Express 25(22) 27570-27580 (2017)

Robust weighted K-means clustering algorithm for a probabilistic-shaped 64QAM coherent optical communication system

Xishuo Wang, Qi Zhang, Xiangjun Xin, Ran Gao, Qinghua Tian, Feng Tian, Chuxuan Wang, Xiaolong Pan, Yongjun Wang, and Leijing Yang
Opt. Express 27(26) 37601-37613 (2019)

Bi-directional gated recurrent unit neural network based nonlinear equalizer for coherent optical communication system

Xinyu Liu, Yongjun Wang, Xishuo Wang, Hui Xu, Chao Li, and Xiangjun Xin
Opt. Express 29(4) 5923-5933 (2021)

DACNN-aided nonlinear equalizer for a probabilistic shaping coherent optical communication system

Yuzhe Li, Huan Chang, Qi Zhang, Ran Gao, Feng Tian, Qinghua Tian, Yongjun Wang, Lan Rao, Dong Guo, Fu Wang, Sitong Zhou, and Xiangjun Xin
Appl. Opt. 63(7) 1881-1887 (2024)

Data-defined naïve Bayes (DNB) based decision scheme for the nonlinear mitigation for OAM mode division multiplexed optical fiber communication

Sitong Zhou, Ran Gao, Qi Zhang, Huan Chang, Xiangjun Xin, Yang Zhao, Jie Liu, and Zhenrui Lin
Opt. Express 29(4) 5901-5914 (2021)