Variational Autoencoders for Precoding Matrices with High Spectral Efficiency

TL;DR

This paper introduces a novel application of variational autoencoders (VAE and CVAE) to generate high spectral efficiency precoding matrices in MIMO systems, providing a computationally efficient approach with minimal quality loss.

Contribution

It presents the first use of VAE and CVAE to model the distribution of high SE precoding matrices, enabling efficient sampling and reconstruction.

Findings

VAE and CVAE can accurately reconstruct the distribution of high SE precoding matrices.

The proposed algorithms are computationally efficient compared to optimal precoding methods.

The distribution of precoding matrices for high spectral efficiency is described for the first time.

Abstract

Neural networks are used for channel decoding, channel detection, channel evaluation, and resource management in multi-input and multi-output (MIMO) wireless communication systems. In this paper, we consider the problem of finding precoding matrices with high spectral efficiency (SE) using variational autoencoder (VAE). We propose a computationally efficient algorithm for sampling precoding matrices with minimal loss of quality compared to the optimal precoding. In addition to VAE, we use the conditional variational autoencoder (CVAE) to build a unified generative model. Both of these methods are able to reconstruct the distribution of precoding matrices of high SE by sampling latent variables. This distribution obtained using VAE and CVAE methods is described in the literature for the first time.