End-to-end Learning for GMI Optimized Geometric Constellation Shape

Rasmus Thomas Jones, Metodi Plamenov Yankov, Darko Zibar

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

42 Downloads (Pure)

Abstract

Autoencoder-based geometric shaping is proposed that includes optimizing bit mappings. Up to 0.2 bits/QAM symbol gain in GMI is achieved for a variety of data rates and in the presence of transceiver impairments. The gains can be harvested with standard binary FEC at no cost w.r.t. conventional BICM.
Original languageEnglish
Title of host publicationProceedings of 45th European Conference on Optical Communication
Number of pages4
PublisherInstitution of Engineering and Technology
Publication date2019
ISBN (Print)978-1-83953-185-9
DOIs
Publication statusPublished - 2019
Event45th European Conference on Optical Communication - Royal Dublin Showground, Dublin, Ireland
Duration: 22 Sep 201926 Sep 2019
Conference number: 45
http://www.ecoc2019.org

Conference

Conference45th European Conference on Optical Communication
Number45
LocationRoyal Dublin Showground
CountryIreland
CityDublin
Period22/09/201926/09/2019
Internet address
Series45th European Conference on Optical Communication (ecoc 2019)

Keywords

  • Geometric constellation shaping
  • Generalized mutual information
  • Fiber channel
  • End-to-end learning

Fingerprint Dive into the research topics of 'End-to-end Learning for GMI Optimized Geometric Constellation Shape'. Together they form a unique fingerprint.

Cite this