Scaling Deep Learning-Based Decoding of Polar Codes via Partitioning

TLDR

Deep learning-based channel decoders have training complexity that grows exponentially with codebook size and number of information bits, limiting neural network decoding to very short block lengths. This study aims to enhance the conventional iterative decoding algorithm for polar codes by replacing sub‑blocks of the decoder with neural‑network components. The authors partition the polar‑code encoding graph into smaller sub‑blocks, train each block individually to approximate maximum‑a‑posteriori performance, and then connect the blocks through remaining conventional belief‑propagation stages. The resulting non‑iterative decoder achieves competitive bit‑error‑rate performance while enabling high parallelization, and its performance is comparable to state‑of‑the‑art polar decoders such as successive‑cancellation list and belief‑propagation decoding.

Abstract

The training complexity of deep learning-based channel decoders scales exponentially with the codebook size and therefore with the number of information bits. Thus, neural network decoding (NND) is currently only feasible for very short block lengths. In this work, we show that the conventional iterative decoding algorithm for polar codes can be enhanced when sub-blocks of the decoder are replaced by neural network (NN) based components. Thus, we partition the encoding graph into smaller sub-blocks and train them individually, closely approaching maximum a posteriori (MAP) performance per sub-block. These blocks are then connected via the remaining conventional belief propagation decoding stage(s). The resulting decoding algorithm is non-iterative and inherently enables a highlevel of parallelization, while showing a competitive bit error rate (BER) performance. We examine the degradation through partitioning and compare the resulting decoder to state-of-the art polar decoders such as successive cancellation list and belief propagation decoding.