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

Darko Zibar; Francesco Da Ros; Metodi P. Yankov; Ognjen Jovanovic

arxiv: 2106.00431 · v1 · pith:DCJ2CK3L · submitted 2021-06-01 · eess.SP

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

Ognjen Jovanovic , Metodi P. Yankov , Francesco Da Ros , Darko Zibar This is my paper

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:DCJ2CK3L record.json open to challenge →

classification eess.SP

keywords constellationlaserlinewidthrobustshapingautoencoder-basedbitsend-to-end

0 comments

read the original 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.

This paper has not been read by Pith yet.

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

discussion (0)