Discriminative Mutual Information Estimation for the Design of Channel Capacity Driven Autoencoders

TL;DR

This paper introduces novel discriminative mutual information estimators to enhance the design of autoencoders for communication systems, aiming to approach channel capacity and improve end-to-end system performance.

Contribution

It proposes new mutual information estimators specifically for training capacity-approaching autoencoders in communication systems.

Findings

New discriminative mutual information estimators developed

Demonstrated potential for designing capacity-approaching codes

Enhanced estimation of channel capacity in deep learning-based communication systems

Abstract

The development of optimal and efficient machine learning-based communication systems is likely to be a key enabler of beyond 5G communication technologies. In this direction, physical layer design has been recently reformulated under a deep learning framework where the autoencoder paradigm foresees the full communication system as an end-to-end coding-decoding problem. Given the loss function, the autoencoder jointly learns the coding and decoding optimal blocks under a certain channel model. Because performance in communications typically refers to achievable rates and channel capacity, the mutual information between channel input and output can be included in the end-to-end training process, thus, its estimation becomes essential. In this paper, we present a set of novel discriminative mutual information estimators and we discuss how to exploit them to design capacity-approaching codes and ultimately estimate the channel capacity.