Conference paper
End-to-end Learning for GMI Optimized Geometric Constellation Shape
Department of Photonics Engineering, Technical University of Denmark1
Ultra-fast Optical Communication, Department of Photonics Engineering, Technical University of Denmark2
Coding and Visual Communication, Department of Photonics Engineering, Technical University of Denmark3
Centre of Excellence for Silicon Photonics for Optical Communications, Centers, Technical University of Denmark4
Machine Learning in Photonic Systems, Department of Photonics Engineering, Technical University of Denmark5
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.
Language: | English |
---|---|
Publisher: | Institution of Engineering and Technology |
Year: | 2019 |
Pages: | 560-564 |
Proceedings: | 45th European Conference on Optical Communication |
ISBN: | 1839530073 , 1839531851 , 9781839530074 and 9781839531859 |
Types: | Conference paper |
DOI: | 10.1049/cp.2019.0886 |
ORCIDs: | Yankov, Metodi Plamenov and Zibar, Darko |