Neural Decoding with Optimization of Node Activations

TL;DR

This paper introduces a neural decoding method for error-correcting codes that enhances performance by optimizing node activations with novel loss terms, achieving up to 1.1dB improvement without increasing complexity.

Contribution

It proposes two new loss terms to improve neural decoders, one for sparsity and one for mimicking a superior teacher decoder, enhancing decoding accuracy.

Findings

Achieves up to 1.1dB performance gain on BCH codes.

Maintains same runtime and model size as neural Belief Propagation.

Demonstrates effectiveness of activation optimization in neural decoding.

Abstract

The problem of maximum likelihood decoding with a neural decoder for error-correcting code is considered. It is shown that the neural decoder can be improved with two novel loss terms on the node's activations. The first loss term imposes a sparse constraint on the node's activations. Whereas, the second loss term tried to mimic the node's activations from a teacher decoder which has better performance. The proposed method has the same run time complexity and model size as the neural Belief Propagation decoder, while improving the decoding performance by up to $1.1dB$ on BCH codes.