End-to-End Learning of Geometrical Shaping Maximizing Generalized Mutual Information
arXiv:1912.05638v155 citations
Originality Incremental advance
AI Analysis
This work addresses the challenge of nonconvex optimization in end-to-end learning for communication systems, offering incremental improvements in constellation design.
The paper tackled the problem of designing optimal constellations for communication systems by applying gradient descent to constellation coordinates, resulting in state-of-the-art constellations in 2D and 4D that provide reach increases up to 26% compared to QAM.
GMI-based end-to-end learning is shown to be highly nonconvex. We apply gradient descent initialized with Gray-labeled APSK constellations directly to the constellation coordinates. State-of-the-art constellations in 2D and 4D are found providing reach increases up to 26\% w.r.t. to QAM.