Abstract
We experimentally investigated four SVM multi-classification methods for machine learning nonlinearity mitigation. The comparison results indicate very close BER performance with significant improvement. Meanwhile, the SVM multi-classifier based on the in-phase and quadrature component has the lowest complexity.
© 2018 The Author(s)
PDF ArticleMore Like This
Chang Wang, Guoyao Chen, Haoyang Wang, Lin Sun, Jiangbing Du, Zuyuan He, Ke Xu, and Bo Liu
Su1D.4 Asia Communications and Photonics Conference (ACP) 2018
Min He, Jie Gao, Dongxu Lu, Yuqiang Yang, Jiahao Huo, and Xian Zhou
M4A.114 Asia Communications and Photonics Conference (ACP) 2019
Danshi Wang, Min Zhang, Ze Li, Yue Cui, Shanguo Huang, and Hongxiang Wang
ASu3F.4 Asia Communications and Photonics Conference (ACP) 2015