Autoencoder-Based Unequal Error Protection Codes

TL;DR

This paper introduces an autoencoder-based method for designing codes with unequal error protection, offering improved performance and flexibility over traditional coding schemes for different importance classes.

Contribution

The paper proposes a novel autoencoder-based framework for UEP code design that generalizes loss functions and demonstrates superior performance compared to existing methods.

Findings

Autoencoder-based UEP codes outperform traditional schemes.

The method provides flexible trade-offs in error protection.

Superior performance in message-wise and bit-wise UEP scenarios.

Abstract

We present a novel autoencoder-based approach for designing codes that provide unequal error protection (UEP) capabilities. The proposed design is based on a generalization of an autoencoder loss function that accommodates both message-wise and bit-wise UEP scenarios. In both scenarios, the generalized loss function can be adjusted using an associated weight vector to trade off error probabilities corresponding to different importance classes. For message-wise UEP, we compare the proposed autoencoder-based UEP codes with a union of random coset codes. For bit-wise UEP, the proposed codes are compared with UEP rateless spinal codes and the superposition of random Gaussian codes. In all cases, the autoencoder-based codes show superior performance while providing design simplicity and flexibility in trading off error protection among different importance classes.