Title :
On discrete hidden Markov state estimation
Author_Institution :
Signal & Syst. Technol., Seattle, WA, USA
Abstract :
Hidden Markov models can be used to describe the behavior of a class of dynamic systems that are subject to abrupt changes. Since the Markov state is “hidden” and can only be observed through imperfect observations, its estimation is of practical importance for control and prediction. In this paper, a unified framework is established within which a comparative study of various hidden state estimation filters based on discrete-valued observations is presented
Keywords :
filtering theory; hidden Markov models; state estimation; control; discrete hidden Markov state estimation; discrete-valued observations; hidden Markov models; imperfect observations; prediction; state estimation filters; Control systems; Filters; Hidden Markov models; Hydrogen; Manufacturing automation; State estimation; Target recognition;
Conference_Titel :
American Control Conference, Proceedings of the 1995
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-2445-5
DOI :
10.1109/ACC.1995.529197
