Sequential simulation-based estimation of jump Markov linear systems

TitleSequential simulation-based estimation of jump Markov linear systems
Publication TypeConference Paper
Year of Publication2000
AuthorsDoucet, A., N. J. Gordon, and V. Krishnamurthy
Conference NameDecision and Control, 2000. Proceedings of the 39th IEEE Conference on
Pagination1166 -1171 vol.2
Keywordsdiscrete time systems, filtering theory, finite state Markov chain, Gaussian noise, importance sampling, jump Markov linear systems, linear systems, Markov chain Monte Carlo methods, Markov processes, optimal conditional mean state estimates, optimal filtering problem, particle filters, selection scheme, sequential importance sampling, sequential simulation-based estimation, state estimation, statistical structure, variance reduction methods

Jump Markov linear systems (JMLS) are linear systems whose parameters evolve with time according to a finite state Markov chain. Our aim is to recursively compute optimal conditional mean state estimates for JMLS. We present efficient simulation-based algorithms called particle filters to solve the optimal filtering problem. Our algorithms combine sequential importance sampling, a selection scheme and Markov chain Monte Carlo methods. They use several variance reduction methods to make the most of the statistical structure of JMLS


a place of mind, The University of British Columbia

Electrical and Computer Engineering
2332 Main Mall
Vancouver, BC Canada V6T 1Z4
Tel +1.604.822.2872
Fax +1.604.822.5949

Emergency Procedures | Accessibility | Contact UBC | © Copyright 2020 The University of British Columbia