Particle Based Smoothed Marginal MAP Estimation For General State Space Models
2012 (English)In: IEEE Transactions on Signal Processing, ISSN 1053-587X, E-ISSN 1941-0476, Vol. 61, no 2, 264-273 p.Article in journal (Refereed) Published
We consider the smoothing problem for a general state space system using sequential Monte Carlo(SMC) methods. The marginal smoother is assumed to be available in the form of weighted randomparticles from the SMC output. New algorithms are developed to extract the smoothed marginal maximuma posteriori (MAP) estimate of the state from the existing marginal particle smoother. Our method doesnot need any kernel ﬁtting to obtain the posterior density from the particle smoother. The proposedestimator is then successfully applied to ﬁnd the unknown initial state of a dynamical system and toaddress the issue of parameter estimation problem in state space models
Place, publisher, year, edition, pages
IEEE Signal Processing Society, 2012. Vol. 61, no 2, 264-273 p.
Sequential Monte Carlo, Particle smoother, Maximum a posteriori, Unknown initial conditions
Signal Processing Control Engineering
IdentifiersURN: urn:nbn:se:liu:diva-84714DOI: 10.1109/TSP.2012.2223691ISI: 000314678900004OAI: oai:DiVA.org:liu-84714DiVA: diva2:561304