Learning to Flip Successive Cancellation Decoding of Polar Codes with LSTM Networks

TL;DR

This paper introduces a deep learning-based SC flip decoding method for polar codes using LSTM networks to more accurately identify error bits and improve decoding performance.

Contribution

It proposes a novel LSTM-based approach with a two-stage training method to enhance SC flip decoding of polar codes, surpassing existing algorithms.

Findings

More accurate error bit identification

Improved decoding performance over state-of-the-art

Effective two-stage training of LSTM networks

Abstract

The key to successive cancellation (SC) flip decoding of polar codes is to accurately identify the first error bit. The optimal flipping strategy is considered difficult due to lack of an analytical solution. Alternatively, we propose a deep learning aided SC flip algorithm. Specifically, before each SC decoding attempt, a long short-term memory (LSTM) network is exploited to either (i) locate the first error bit, or (ii) undo a previous `wrong' flip. In each SC attempt, the sequence of log likelihood ratios (LLRs) derived in the previous SC attempt is exploited to decide which action to take. Accordingly, a two-stage training method of the LSTM network is proposed, i.e., learn to locate first error bits in the first stage, and then to undo `wrong' flips in the second stage. Simulation results show that the proposed approach identifies error bits more accurately and achieves better performance than the state-of-the-art SC flip algorithms.