TY - JOUR
T1 - Merging Belief Propagation and the Mean Field Approximation: A Free Energy Approach
AU - Riegler, Erwin
AU - Kirkelund, Gunvor Elisabeth
AU - Manchón, Carles Navarro
AU - Badiu, Mihai-Alin
AU - Fleury, Bernard Henri
PY - 2013/1
Y1 - 2013/1
N2 - We present a joint message passing approach that combines belief propagation and the mean field approximation. Our analysis is based on the region-based free energy approximation method proposed by Yedidia et al. We show that the message passing fixed-point equations obtained with this combination correspond to stationary points of a constrained region-based free energy approximation. Moreover, we present a convergent implementation of these message passing fixed-point equations provided that the underlying factor graph fulfills certain technical conditions. In addition, we show how to include hard constraints in the part of the factor graph corresponding to belief propagation. Finally, we demonstrate an application of our method to iterative channel estimation and decoding in an OFDM system.
AB - We present a joint message passing approach that combines belief propagation and the mean field approximation. Our analysis is based on the region-based free energy approximation method proposed by Yedidia et al. We show that the message passing fixed-point equations obtained with this combination correspond to stationary points of a constrained region-based free energy approximation. Moreover, we present a convergent implementation of these message passing fixed-point equations provided that the underlying factor graph fulfills certain technical conditions. In addition, we show how to include hard constraints in the part of the factor graph corresponding to belief propagation. Finally, we demonstrate an application of our method to iterative channel estimation and decoding in an OFDM system.
U2 - 10.1109/TIT.2012.2218573
DO - 10.1109/TIT.2012.2218573
M3 - Journal article
SN - 0018-9448
VL - 59
SP - 588
EP - 602
JO - I E E E Transactions on Information Theory
JF - I E E E Transactions on Information Theory
IS - 1
ER -