Complex Fully Connected Neural Networks for Nonlinearity Compensation in Long-Haul Transmission Systems

Signal nonlinear impairments have been one of the fundamental limiting factors for the further development of optical communication systems operating at broader bandwidth and longer distances. To tackle this problem, a number of techniques have been proposed. One of the promising approaches is the d...

Full description

Saved in:
Bibliographic Details
Published in2021 Conference on Lasers and Electro-Optics Europe & European Quantum Electronics Conference (CLEO/Europe-EQEC) p. 1
Main Authors Bogdanov, Stepan, Sidelnikov, Oleg
Format Conference Proceeding
LanguageEnglish
Published IEEE 21.06.2021
Subjects
Bandwidth
Equalizers
Europe
Limiting
Neural networks
Optical fiber communication
Online AccessGet full text
DOI10.1109/CLEO/Europe-EQEC52157.2021.9592658

Cover

More Information
Summary:Signal nonlinear impairments have been one of the fundamental limiting factors for the further development of optical communication systems operating at broader bandwidth and longer distances. To tackle this problem, a number of techniques have been proposed. One of the promising approaches is the development of nonlinear equalizers (NLE) based on neural networks. Such equalizers provide high accuracy of symbol classification, requiring acceptable computational resources [1] .
DOI:10.1109/CLEO/Europe-EQEC52157.2021.9592658