Near Perfect Decoding of LDPC Codes

TL;DR

This paper introduces a cooperative optimization approach for decoding LDPC codes, achieving significantly better error rates, faster decoding, and lower error floors compared to traditional methods.

Contribution

It applies cooperative optimization to LDPC decoding, demonstrating substantial improvements over the sum-product algorithm in accuracy and speed.

Findings

Error rate reduced threefold

Decoding speed increased six times

Error floors lowered dramatically

Abstract

Cooperative optimization is a new way for finding global optima of complicated functions of many variables. It has some important properties not possessed by any conventional optimization methods. It has been successfully applied in solving many large scale optimization problems in image processing, computer vision, and computational chemistry. This paper shows the application of this optimization principle in decoding LDPC codes, which is another hard combinatorial optimization problem. In our experiments, it significantly out-performed the sum-product algorithm, the best known method for decoding LDPC codes. Compared to the sum-product algorithm, our algorithm reduced the error rate further by three fold, improved the speed by six times, and lowered error floors dramatically in the decoding.