Decoding Short LDPC Codes via BP-RNN Diversity and Reliability-Based Post-Processing

• Published in: IEEE transactions on communications Vol. 70; no. 12; p. 1
• Main Authors: Rosseel, Joachim, Mannoni, Valerian, Fijalkow, Inbar, Savin, Valentin
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.12.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE); Institute of Electrical and Electronics Engineers
• Subjects: Algorithms; Artificial neural networks; belief propagation; Bit error rate; Codes; Computer Science; Decoders; Diversity reception; Entropy (Information theory); Error correcting codes; Iterative decoding; LDPC; Maximum likelihood decoding; Networking and Internet Architecture; Neural and Evolutionary Computing; neural network aided decoding; Neurons; Optimization; ordered statistics decoding post-processing; Recurrent neural networks; short LDPC; Training; Waterfalls; LDPC; ordered statistics decoding postprocessing; belief propagation; short LDPC; neural network aided decoding
• ISSN: 0090-6778; 1558-0857; 1558-0857
• DOI: 10.1109/TCOMM.2022.3218821

This paper investigates decoder diversity architectures for short low-density parity-check (LDPC) codes, based on recurrent neural network (RNN) models of the belief-propagation (BP) algorithm. We propose a new approach to achieve decoder diversity in the waterfall region, by specializing BP-RNN decoders to specific classes of errors, with absorbing set support. We further combine our approach with an ordered statistics decoding (OSD) post-processing step, which effectively leverages the bit-error rate optimization deriving from the use of the binary cross-entropy loss function. We show that a single specialized BP-RNN decoder combines better than BP with the OSD post-processing step. Moreover, combining OSD post-processing with the diversity brought by the use of multiple BP-RNN decoders, provides an efficient way to bridge the gap to maximum likelihood decoding.