Improved Low-Density Parity-Check Codes Using Irregular Graphs and Belief Propagation

Authors: Michael G. Luby, Michael Mitzenmacher, M. Amin Shokrollahi, Daniel A. Spielman.

Bibliographic Information: Presented at the 1998 International Symposium on Information Theory, page 117. Incorporated into the journal article Improved Low-Density Parity-Check Codes Using Irregular Graphs.

Abstract

We construct new families of low-density parity-check codes, which we call {\em irregular codes}. When decoded using belief propagation, our codes can correct more errors than previously known low-density codes. Our improved performance comes from using codes based on irregular random bipartite graphs, based on the work of \cite{LMSSS}. Previously studied low-density codes have been derived from regular bipartite graphs. Initial experimental results for our irregular codes suggest that, with improvements, irregular codes may be able to match turbo code performance.
You can download this paper as Postscript, compressed Postscript, PDF, or compressed PDF.