A Low-Complexity Recurrent Neural Network Based Joint Equalization and Decoding Method for Trellis Coded Modulation Link in Data Center

2020 Opto-Electronics and Communications Conference (OECC) Pub Date : 2020-10-04 DOI:10.1109/OECC48412.2020.9273586

Xin Qin, Pengfei Wang, Chuanchuan Yang, Fan Zhang, Xinyue Wang, Ziyu Wang

引用次数: 1

Abstract

RNN-based joint equalization and decoding methods are proposed for data center with TCM signals. Numerical results show compared with traditional DSP algorithms, our method improves power sensitivity by 2.2dB@BER=3.8×10-3 with 96% reduced complexity.

查看原文本刊更多论文

基于低复杂度递归神经网络的数据中心栅格编码调制链路联合均衡与解码方法

针对具有TCM信号的数据中心，提出了基于rnn的联合均衡与解码方法。数值结果表明，与传统的DSP算法相比，该方法的功率灵敏度提高2.2dB@BER=3.8×10-3，复杂度降低96%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2020 Opto-Electronics and Communications Conference (OECC)

自引率

0.00%

发文量