Skip to Main content Skip to Navigation
Conference papers

Backward hidden Markov chain for outlier-robust filtering and fixed-interval smoothing

B. Ait-El-Fquih 1 Cedric Gouy-Pailler 1
1 LADIS - Laboratoire d'analyse des données et d'intelligence des systèmes
DM2I - Département Métrologie Instrumentation & Information : DRT/LIST/DM2I
Abstract : This paper addresses the problem of recursive estimation of a process in presence of outliers among the observations. It focuses on deriving robust approximate Kalman-like backward filtering and backward-forward fixed-interval smoothing algorithms in the context of linear hidden Markov chain with heavy-tailed measurement noise. The proposed algorithms are derived based on the backward Markovianity of the model as well as the variational Bayesian approach. In a simulation design, our algorithms are shown to outperform the classical Kalman filter in the presence of outliers.
Keywords : Variational bayesian approaches Algorithms Bayesian networks Markov processes Signal processing Statistics Variational bayes Robust smoothing Robust filtering Recursive estimation Markovian model Hidden Markov chains Fixed-interval smoothing
Document type :
Conference papers
Complete list of metadata
https://hal-cea.archives-ouvertes.fr/cea-01830771
Contributor : Léna Le Roy Connect in order to contact the contributor
Submitted on : Thursday, July 5, 2018 - 1:12:51 PM
Last modification on : Friday, June 25, 2021 - 10:00:02 AM

Identifiers

Collections

CEA | DRT | LIST | DM2I | GS-COMPUTER-SCIENCE

`

Citation

B. Ait-El-Fquih, Cedric Gouy-Pailler. Backward hidden Markov chain for outlier-robust filtering and fixed-interval smoothing. 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, May 2013, Vancouver, BC, Canada. pp.5504-5508, ⟨10.1109/ICASSP.2013.6638716⟩. ⟨cea-01830771⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

130