Abstract
Recurrent neural network based equalizer with parallel outputs is proposed and experimentally demonstrated. Result shows that it outperforms fully-connected neural network for 100Gbps/λ PON using 25G-class optics. Hardware implementation issues are also discussed.
© 2021 The Author(s)
PDF Article | Presentation VideoMore Like This
Stephen Murphy, Paul D. Townsend, and Cleitus Antony
SW4E.1 CLEO: Science and Innovations (CLEO:S&I) 2022
Zhaopeng Xu, Shuangyu Dong, Chenxin Jiang, Jonathan H. Manton, and William Shieh
M5H.4 Asia Communications and Photonics Conference (ACP) 2021
Xiaoan Huang, Dongxu Zhang, Xiaofeng Hu, and Kaibin Zhang
Tu5.46 European Conference and Exhibition on Optical Communication (ECOC) 2022