Advanced methods to mitigate fiber nonlinearies using neural networks and probabilistic shaping

O. S. Sidelnikov*, A. S. Skidin, S. Sygletos, M. P. Fedoruk

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose a combined approach to mitigate nonlinear fiber effects based on both the probabilistic shaping and static neural networks. We show that such combination can expand the system reach by 25-35%.

Original languageEnglish
Title of host publicationBragg Gratings, Photosensitivity and Poling in Glass Waveguides and Materials, BGPPM 2018
PublisherOptical Society of America
VolumePart F98-BGPPM 2018
ISBN (Electronic)9781557528209
DOIs
Publication statusPublished - 5 Jul 2018
EventBragg Gratings, Photosensitivity and Poling in Glass Waveguides and Materials, BGPPM 2018 - Zurich, Switzerland
Duration: 2 Jul 20185 Jul 2018

Conference

ConferenceBragg Gratings, Photosensitivity and Poling in Glass Waveguides and Materials, BGPPM 2018
CountrySwitzerland
CityZurich
Period2/07/185/07/18

    Fingerprint

Cite this

Sidelnikov, O. S., Skidin, A. S., Sygletos, S., & Fedoruk, M. P. (2018). Advanced methods to mitigate fiber nonlinearies using neural networks and probabilistic shaping. In Bragg Gratings, Photosensitivity and Poling in Glass Waveguides and Materials, BGPPM 2018 (Vol. Part F98-BGPPM 2018). Optical Society of America. https://doi.org/10.1364/BGPPM.2018.JTu5A.48