Uniformly Reweighted Belief Propagation for Estimation and Detection in Wireless Networks
Journal article, 2012
In this paper, we propose a new inference algorithm, suitable for distributed processing over wireless networks. The algorithm, called uniformly reweighted belief propagation (URW-BP), combines the local nature of belief propagation with the improved performance of tree-reweighted belief propagation (TRW-BP) in graphs with cycles. It reduces the degrees of freedom in the latter algorithm to a single scalar variable, the uniform edge appearance probability rho. We provide a variational interpretation of URW-BP, give insights into good choices of rho, develop an extension to higher-order potentials, and complement our work with numerical performance results on three inference problems in wireless communication systems: spectrum sensing in cognitive radio, cooperative positioning, and decoding of a low-density parity-check (LDPC) code.
factor graph approach
sensor networks
belief propagation
sum-product algorithm
localization
message passing
distributed detection
Distributed inference
approximate inference
factor graphs
models
multiple sensors
Author
Henk Wymeersch
Chalmers, Signals and Systems, Kommunikationssystem, informationsteori och antenner
Federico Penna
Fraunhofer-Institut fur Nachrichtentechnik Heinrich-Hertz-Institut - HHI
Vladimir Savic
Universidad Politecnica de Madrid
IEEE Transactions on Wireless Communications
1536-1276 (ISSN)
Vol. 11 4 1587-1595 6153324Cooperative Situational Awareness for Wireless Networks (COOPNET)
European Commission (FP7) (EC/FP7/258418), 2011-05-01 -- 2016-04-30.
Robust and Fault-Tolerant Cooperative Positioning
Swedish Research Council (VR) (2010-5889), 2011-01-01 -- 2013-12-31.
Subject Categories (SSIF 2011)
Computer and Information Science
DOI
10.1109/twc.2012.021412.111509