Linear MIMO Precoders Design for Finite Alphabet Inputs via Model-Free Training

TL;DR

This paper introduces a model-free autoencoder-based approach for designing linear MIMO precoders with finite alphabet inputs, eliminating the need for explicit channel knowledge while effectively maximizing mutual information.

Contribution

It proposes a novel autoencoder training method that jointly optimizes precoders and receivers without requiring channel models, advancing MIMO precoder design.

Findings

Achieves comparable mutual information maximization to model-based methods.

Reduces bit error rate effectively in simulations.

Operates without explicit channel knowledge.

Abstract

This paper investigates a novel method for designing linear precoders with finite alphabet inputs based on autoencoders (AE) without the knowledge of the channel model. By model-free training of the autoencoder in a multiple-input multiple-output (MIMO) system, the proposed method can effectively solve the optimization problem to design the precoders that maximize the mutual information between the channel inputs and outputs, when only the input-output information of the channel can be observed. Specifically, the proposed method regards the receiver and the precoder as two independent parameterized functions in the AE and alternately trains them using the exact and approximated gradient, respectively. Compared with previous precoders design methods, it alleviates the limitation of requiring the explicit channel model to be known. Simulation results show that the proposed method works as well as those methods under known channel models in terms of maximizing the mutual information and reducing the bit error rate.