Using Deep Neural Networks to Predict and Improve the Performance of Polar Codes
This work addresses the practical optimization problem for polar code design in communication systems, representing an incremental improvement through a hybrid machine learning approach.
The paper tackles the challenge of designing efficient polar codes for specific communication contexts by training deep neural networks to predict frame error rates based on frozen bit sequences, and introduces a Projected Gradient Descent algorithm that uses these predictions to generate improved codes, achieving better performance than the training codes in generated datasets.
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.