paper

Deep Learning of Geometric Constellation Shaping including Fiber Nonlinearities

arXiv:1805.03785

Abstract

A new geometric shaping method is proposed, leveraging unsupervised machine learning to optimize the constellation design. The learned constellation mitigates nonlinear effects with gains up to 0.13 bit/4D when trained with a simplified fiber channel model.

3 pages, 6 figures, submitted to ECOC 2018

Deep Learning of Geometric Constellation Shaping including Fiber Nonlinearities · wovepaper