Bit Error and Block Error Rate Training for ML-Assisted Communication

TL;DR

This paper explores specialized loss functions for training ML-assisted communication systems, focusing on minimizing block error rate and improving performance across varying SNRs, with demonstrated benefits in simulations.

Contribution

It introduces new loss functions tailored for coded communication systems and proposes SNR deweighting to enhance robustness over different SNR levels.

Findings

Loss functions targeting block error rate improve training outcomes.

SNR deweighting enhances performance across a range of SNRs.

Proposed methods outperform traditional BCE loss in simulations.

Abstract

Even though machine learning (ML) techniques are being widely used in communications, the question of how to train communication systems has received surprisingly little attention. In this paper, we show that the commonly used binary cross-entropy (BCE) loss is a sensible choice in uncoded systems, e.g., for training ML-assisted data detectors, but may not be optimal in coded systems. We propose new loss functions targeted at minimizing the block error rate and SNR deweighting, a novel method that trains communication systems for optimal performance over a range of signal-to-noise ratios. The utility of the proposed loss functions as well as of SNR deweighting is shown through simulations in NVIDIA Sionna.