Time discretization of continuous time filters for HMM parameters estimation - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 1992

Time discretization of continuous time filters for HMM parameters estimation

Abstract

We propose numerical techniques for parameter estimation of fast-sampled homogeneous Markov chains observed in white Gaussian noise. Recently, continuous-time filters that estimate the quantities used in the EM algorithm for MLE parameter estimation were obtained by Elliott. Our numerical work is based on the robust discretization of these filters. The advantage of using filters in the EM algorithm is that they have negligible memory requirements; indeed, independent of the number of observations. In comparison, standard discrete time EM algorithms (Baum-Welch re-estimation equations) are based on smoothers and require the use of the forward-backward algorithm, which is a fixed-interval algorithm and has memory requirements proportional to the number of observations. Although the computational complexity of our filters at each time instant is 0(N^4) (for a N state Markov chain) compared to O(N^2) for the forward-backward scheme, the filters are suitable for parallel implementation. A careful analysis of our techniques might reduce the computational complexity. We present simulations to illustrate the satisfactory performance of our algorithms.
No file

Dates and versions

hal-00912053 , version 1 (20-12-2013)

Identifiers

Cite

Matthew R. James, Vikram Krishnamurthy, François Le Gland. Time discretization of continuous time filters for HMM parameters estimation. Proceedings of the 31st IEEE Conference on Decision and Control, Tucson 1992, IEEE--CSS, Dec 1992, Tucson, United States. pp.3305--3310, ⟨10.1109/CDC.1992.371026⟩. ⟨hal-00912053⟩
180 View
0 Download

Altmetric

Share

Gmail Facebook X LinkedIn More