Concatenated Classic and Neural (CCN) Codes: ConcatenatedAE

TL;DR

This paper introduces a novel concatenated coding scheme combining neural networks and classic codes, significantly improving error correction performance and robustness over traditional neural-only codes in noisy channels.

Contribution

It proposes a concatenated neural and classic coding architecture that enhances error correction and robustness, extending neural code dimensions using repeated NNs with shared parameters.

Findings

Significant reduction in block error probabilities for Gaussian noise channels

Enhanced robustness to channel model variations

Effective extension of neural code dimensions using concatenation

Abstract

Small neural networks (NNs) used for error correction were shown to improve on classic channel codes and to address channel model changes. We extend the code dimension of any such structure by using the same NN under one-hot encoding multiple times, then serially-concatenated with an outer classic code. We design NNs with the same network parameters, where each Reed-Solomon codeword symbol is an input to a different NN. Significant improvements in block error probabilities for an additive Gaussian noise channel as compared to the small neural code are illustrated, as well as robustness to channel model changes.