Using Deep Neural Networks to Predict and Improve the Performance of Polar Codes

TL;DR

This paper presents a novel approach using deep neural networks to predict and optimize the performance of polar codes by designing better frozen bit sequences, improving their efficiency in communication systems.

Contribution

The paper introduces a methodology that trains neural networks to predict polar code performance and uses gradient-based algorithms to generate more efficient frozen bit sequences.

Findings

Neural networks can accurately predict polar code frame error rates.

Gradient-based algorithms can generate improved frozen bit sequences.

Proposed method outperforms existing codes in efficiency.

Abstract

Polar codes can theoretically achieve very competitive Frame Error Rates. In practice, their performance may depend on the chosen decoding procedure, as well as other parameters of the communication system they are deployed upon. As a consequence, designing efficient polar codes for a specific context can quickly become challenging. In this paper, we introduce a methodology that consists in training deep neural networks to predict the frame error rate of polar codes based on their frozen bit construction sequence. We introduce an algorithm based on Projected Gradient Descent that leverages the gradient of the neural network function to generate promising frozen bit sequences. We showcase on generated datasets the ability of the proposed methodology to produce codes more efficient than those used to train the neural networks, even when the latter are selected among the most efficient ones.