End-to-End Learning of Joint Geometric and Probabilistic Constellation Shaping
arXiv:2112.05050v137 citations
Originality Incremental advance
AI Analysis
This work addresses signal processing challenges in communication systems, representing an incremental advancement in constellation shaping techniques.
The paper tackles the problem of optimizing constellation shaping for coded-modulation systems by introducing an autoencoder-based method that maximizes mutual information or generalized mutual information, achieving improved performance metrics.
We present a novel autoencoder-based learning of joint geometric and probabilistic constellation shaping for coded-modulation systems. It can maximize either the mutual information (for symbol-metric decoding) or the generalized mutual information (for bit-metric decoding).