Deep Learning of Geometric Constellation Shaping including Fiber Nonlinearities

Rasmus Thomas Jones, Tobias A. Eriksson, Metodi Plamenov Yankov, Darko Zibar

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

481 Downloads (Pure)

Abstract

A new geometric shaping method is proposed, leveraging unsupervised machinelearning to optimize the constellation design. The learned constellationmitigates nonlinear effects with gains up to 0.13 bit/4D when trained with asimplified fiber channel model.
Original languageEnglish
Title of host publicationProceedings of the 44rd European Conference and Exhibition on Optical Communications (ECOC 2018)
Number of pages3
PublisherIEEE
Publication date2018
ISBN (Print)9781538648629
DOIs
Publication statusPublished - 2018
Event44th European Conference on Optical Communication - Fiera Roma, Rome, Italy
Duration: 23 Sep 201827 Sep 2018

Conference

Conference44th European Conference on Optical Communication
LocationFiera Roma
CountryItaly
CityRome
Period23/09/201827/09/2018

Cite this

Jones, R. T., Eriksson, T. A., Yankov, M. P., & Zibar, D. (2018). Deep Learning of Geometric Constellation Shaping including Fiber Nonlinearities. In Proceedings of the 44rd European Conference and Exhibition on Optical Communications (ECOC 2018) IEEE. https://doi.org/10.1109/ECOC.2018.8535453