End-to-end Learning of a Constellation Shape Robust to Variations in SNR and Laser Linewidth

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

103 Downloads (Pure)

Abstract

We propose an autoencoder-based geometric shaping that learns a constellation robust to SNR and laser linewidth estimation errors. This constellation maintains shaping gain in mutual information (up to 0.3 bits/symbol) with respect to QAM over various SNR and laser linewidth values.
Original languageEnglish
Title of host publicationProceedings of European Conference on Optical Communications 2021
PublisherIEEE
Publication date2021
ISBN (Print)978-1-6654-3868-1
Publication statusPublished - 2021
EventEuropean Conference on Optical Communications 2021 - Parc des expositions, Bordeaux, France
Duration: 13 Sept 202116 Sept 2021
https://www.ecoc2021.org/

Conference

ConferenceEuropean Conference on Optical Communications 2021
LocationParc des expositions
Country/TerritoryFrance
CityBordeaux
Period13/09/202116/09/2021
Internet address

Fingerprint

Dive into the research topics of 'End-to-end Learning of a Constellation Shape Robust to Variations in SNR and Laser Linewidth'. Together they form a unique fingerprint.

Cite this