Bayesian Networks and Particle Filters for Joint Adaptive Equalization and Decoding

Abstract : This thesis is about bayesian networks, particle filters and their application to digital communications. First, we give a rigorous and very general definition of bayesian networks and we formulate the belief propagation algorithm in this context. Then, we present a new type of particle filter, called the 'global sampling particle filter' and we show through numerical simulations that this new algorithm compares favorably with existing filters. Next, we use particle filtering to approximate some of the messages of the belief propagation algorithm. We call the resulting algorithm, which combines belief propagation and particle filtering, the 'turbo particle filtering algorithm'. Finally, we apply these techniques to design methodically a digital communications receiver.
Document type :
Theses
Domain :
Liste complète des métadonnées

Cited literature [18 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/pastel-00000732
Contributor : Ecole Télécom Paristech <>
Submitted on : Monday, September 6, 2004 - 8:00:00 AM
Last modification on : Wednesday, February 20, 2019 - 2:39:12 PM
Document(s) archivé(s) le : Thursday, September 30, 2010 - 6:04:37 PM

Identifiers

  • HAL Id : pastel-00000732, version 1

Citation

Pascal Cheung-Mon-Chan. Bayesian Networks and Particle Filters for Joint Adaptive Equalization and Decoding. domain_other. Télécom ParisTech, 2003. English. ⟨pastel-00000732⟩

Share

Metrics

Record views

276

Files downloads

445