English

Joint Constellation Shaping Using Gradient Descent Approach for MU-MIMO Broadcast Channel

Information Theory 2024-08-22 v2 Machine Learning Networking and Internet Architecture math.IT

Abstract

We introduce a learning-based approach to optimize a joint constellation for a multi-user MIMO broadcast channel (TT Tx antennas, KK users, each with RR Rx antennas), with perfect channel knowledge. The aim of the optimizer (MAX-MIN) is to maximize the minimum mutual information between the transmitter and each receiver, under a sum-power constraint. The proposed optimization method do neither impose the transmitter to use superposition coding (SC) or any other linear precoding, nor to use successive interference cancellation (SIC) at the receiver. Instead, the approach designs a joint constellation, optimized such that its projection into the subspace of each receiver kk, maximizes the minimum mutual information I(Wk;Yk)I(W_k;Y_k) between each transmitted binary input WkW_k and the output signal at the intended receiver YkY_k. The rates obtained by our method are compared to those achieved with linear precoders.

Keywords

Cite

@article{arxiv.2407.07708,
  title  = {Joint Constellation Shaping Using Gradient Descent Approach for MU-MIMO Broadcast Channel},
  author = {Maxime Vaillant and Alix Jeannerot and Jean-Marie Gorce},
  journal= {arXiv preprint arXiv:2407.07708},
  year   = {2024}
}
R2 v1 2026-06-28T17:35:48.790Z