Model-Based Deep Learning of Joint Probabilistic and Geometric Shaping for Optical Communication
This work addresses the challenge of improving data transmission efficiency in optical communication systems, representing an incremental advancement in constellation shaping techniques.
The paper tackled the problem of optimizing geometric and probabilistic constellation shaping for optical coherent communication using autoencoder-based deep learning, resulting in a performance gain of 0.05 bits/4D-symbol mutual information over a 256 QAM Maxwell-Boltzmann distribution for 64 GBd transmission over 170 km SMF link.
Autoencoder-based deep learning is applied to jointly optimize geometric and probabilistic constellation shaping for optical coherent communication. The optimized constellation shaping outperforms the 256 QAM Maxwell-Boltzmann probabilistic distribution with extra 0.05 bits/4D-symbol mutual information for 64 GBd transmission over 170 km SMF link.