Title :
New finite-dimensional filters and smoothers for noisily observed Markov chains
Author :
Elliott, Robert J.
Author_Institution :
Dept. of Stat. & Appl. Probability, Alberta Univ., Edmonton, Alta., Canada
fDate :
1/1/1993 12:00:00 AM
Abstract :
New finite-dimensional filters and smoothers that are related to the Wonham filter of a noisily observed Markov chain are obtained. In particular, finite-dimensional, recursive filters and smoothers are given for the number of jumps from state i to state j, for the occupation time of state i, and for a stochastic integral related to the drift in the observations. These filters allow easy application of the EM algorithm for the estimation of the parameters of the Markov chain and observation process
Keywords :
Markov processes; filtering and prediction theory; parameter estimation; EM algorithm; Wonham filter; expectation-maximisation algorithm; finite-dimensional filters; noisily observed Markov chains; observation process; parameter estimation; recursive filters; smoothers; stochastic integral; Equations; Filters; Parameter estimation; Random variables; Signal processing; Signal processing algorithms; State estimation; State-space methods; Statistics; Stochastic processes;
Journal_Title :
Information Theory, IEEE Transactions on
