Learning to decode LDPC codes with finite-alphabet message passing

Bane V Vasic, Xin Xiao, Shu Lin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Scopus citations

Abstract

In this paper, we discuss the perspectives of utilizing deep neural networks (DNN) to decode Low-Density Parity Check (LDPC) codes. The main idea is to build a neural network to learn and optimize a conventional iterative decoder of LDPC codes. A DNN is based on Tanner graph, and the activation functions emulate message update functions in variable and check nodes. We impose a symmetry on weight matrices which makes it possible to train the DNN on a single codeword and noise realizations only. Based on the trained weights and the bias, we further quantize messages in such DNN-based decoder with 3-bit precision while maintaining no loss in error performance compared to the min-sum algorithm. We use examples to present that the DNN framework can be applied to various code lengths. The simulation results show that, the trained weights and bias make the iterative DNN decoder converge faster and thus achieve higher throughput at the cost of trivial additional decoding complexity.

Original languageEnglish (US)
Title of host publication2018 Information Theory and Applications Workshop, ITA 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728101248
DOIs
StatePublished - Oct 23 2018
Event2018 Information Theory and Applications Workshop, ITA 2018 - San Diego, United States
Duration: Feb 11 2018Feb 16 2018

Other

Other2018 Information Theory and Applications Workshop, ITA 2018
CountryUnited States
CitySan Diego
Period2/11/182/16/18

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Software
  • Information Systems and Management
  • Theoretical Computer Science

Fingerprint Dive into the research topics of 'Learning to decode LDPC codes with finite-alphabet message passing'. Together they form a unique fingerprint.

Cite this