Abstract
We propose transfer-learning (TL) aided convolutional neural network (CNN) equalizer for short-reach IM/DD systems. Simulations of PMA4 systems with varying distances and data rates show that TL-aided CNN outperforms TL-aided ANN and CNN without TL.
© 2021 The Author(s)
PDF ArticleMore Like This
Hao Chen, Xing Liu, Zhaoquan Fan, Chengju Hu, and Jian Zhao
STh1F.5 CLEO: Science and Innovations (CLEO:S&I) 2021
Jing Zhang, Lianshan Yan, Lin Jiang, Anlin Yi, Yan Pan, Wei Pan, and Bin Luo
M4A.320 Asia Communications and Photonics Conference (ACP) 2020
Mingyuan Li, Wenjia Zhang, and Zuyuan He
M4E.4 Asia Communications and Photonics Conference (ACP) 2021