End-to-End Learning of Geometrical Shaping Maximizing Generalized Mutual Information

Kadir Gumus, Alex Alvarado, Bin Chen, Christian Hager, Erik Agrell

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

3 Citations (Scopus)
12 Downloads (Pure)

Abstract

GMI-based end-to-end learning is shown to be highly nonconvex. We apply gradient descent initialized with Gray-labeled APSK constellations directly to the constellation coordinates. State-of-the-art constellations in 2D and 4D are found providing reach increases up to 26% w.r.t. to QAM.

Original languageEnglish
Title of host publication2020 Optical Fiber Communications Conference and Exhibition, OFC 2020 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers
Chapter3
ISBN (Electronic)9781943580712
ISBN (Print)9781943580712
DOIs
Publication statusPublished - 13 May 2020
Event2020 Optical Fiber Communications Conference and Exhibition, OFC 2020 - San Diego, United States
Duration: 8 Mar 202012 Mar 2020

Conference

Conference2020 Optical Fiber Communications Conference and Exhibition, OFC 2020
CountryUnited States
CitySan Diego
Period8/03/2012/03/20

Fingerprint Dive into the research topics of 'End-to-End Learning of Geometrical Shaping Maximizing Generalized Mutual Information'. Together they form a unique fingerprint.

Cite this