Learning Speed of four-layer-DNN-based Nonlinear Equalizer for Optical Communication Systems

Learning Speed of four-layer-DNN-based Nonlinear Equalizer for Optical Communication Systems
复制标题

DOI：
10.23919/oecc/psc53152.2022.9849963
复制
发表时间：
2022-07
期刊：
2022 27th OptoElectronics and Communications Conference (OECC) and 2022 International Conference on Photonics in Switching and Computing (PSC)
影响因子：
--
通讯作者：
Jinya Nakamura;K. Ikuta;Moriya Nakamura
Jinya Nakamura;K. Ikuta;Moriya Nakamura
中科院分区：
其他
文献类型：
--
作者：
Jinya Nakamura;K. Ikuta;Moriya Nakamura

文献摘要

相似文献

比较了用于光纤通信系统非线性补偿的三层神经网络和四层离散神经网络非线性均衡器的学习速度。结果表明，基于DNN的均衡器具有更快的学习特性。
We compared learning speed of three-layer-ANN- and four-laver-DNN-based nonlinear equalizers used for nonlinear compensation in optical communication systems. The results showed that the DNN-based equalizer has faster learning characteristics.