Comments on why generalized BP serves so remarkably in 2-D channels

Ori Shental, Noam Shental, Shlomo Shamai, Ido Kanter, Anthony J. Weiss, Yair Weiss

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Generalized belief propagation (GBP) algorithm has been shown recently to infer the a-posteriori probabilities of finite-state input two-dimensional (2-D) Gaussian channels with memory in a practically accurate manner, thus enabling near-optimal estimation of the transmitted symbols and the Shannon-theoretic information rates. In this note, a rationalization of this excellent performance of GBP is addressed.

Original languageEnglish
Title of host publication2007 Information Theory and Applications Workshop, Conference Proceedings, ITA
Pages369
Number of pages1
DOIs
StatePublished - 2007
Externally publishedYes
Event2007 Information Theory and Applications Workshop, ITA - San Diego, CA, United States
Duration: 29 Jan 20072 Feb 2007

Publication series

Name2007 Information Theory and Applications Workshop, Conference Proceedings, ITA

Conference

Conference2007 Information Theory and Applications Workshop, ITA
Country/TerritoryUnited States
CitySan Diego, CA
Period29/01/072/02/07

Fingerprint

Dive into the research topics of 'Comments on why generalized BP serves so remarkably in 2-D channels'. Together they form a unique fingerprint.

Cite this