An MCMC sampling approach to estimation of nonstationary hidden Markov models

Hidden Markov models (HMMs) represent a very important tool for analysis of signals and systems. In the past two decades, HMMs have attracted the attention of various research communities, including the ones in statistics, engineering, and mathematics. Their extensive use in signal processing and, in particular, speech processing is well documented. A major weakness of conventional HMMs is their inflexibility in modeling state durations. This weakness can be avoided by adopting a more complicated class of HMMs known as nonstationary HMMs. We analyze nonstationary HMMs whose state transition probabilities are functions of time that indirectly model state durations by a given probability mass function and whose observation spaces are discrete. The objective of our work is to estimate all the unknowns of a nonstationary HMM, which include its parameters and the state sequence. To that end, we construct a Markov chain Monte Carlo (MCMC) sampling scheme, where sampling from all the posterior probability distributions is very easy. The proposed MCMC sampling scheme has been tested in extensive computer simulations on finite discrete-valued observed data, and some of the simulation results are presented.

[1]  Peter Green,et al.  Bayesian choice among Markov models of ion channels using Markov chain Monte Carlo , 1999, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[2]  Vikram Krishnamurthy,et al.  Hidden Markov model algorithms for narrowband interference suppression in CDMA spread spectrum systems , 1999, Signal Process..

[3]  C. Robert,et al.  Computational and Inferential Difficulties with Mixture Posterior Distributions , 2000 .

[4]  K Karplus,et al.  Predicting protein structure using only sequence information , 1999, Proteins.

[5]  Jen-Tzung Chien,et al.  Online hierarchical transformation of hidden Markov models for speech recognition , 1999, IEEE Trans. Speech Audio Process..

[6]  Gary A. Churchill,et al.  Bayesian Restoration of a Hidden Markov Chain with Applications to DNA Sequencing , 1999, J. Comput. Biol..

[7]  Jin H. Kim,et al.  Nonstationary hidden Markov model , 1995, Signal Process..

[8]  Gareth O. Roberts,et al.  Convergence assessment techniques for Markov chain Monte Carlo , 1998, Stat. Comput..

[9]  S. Chib.  Calculating posterior distributions and modal estimates in Markov mixture models , 1996 .

[10]  Peter Green,et al.  Markov chain Monte Carlo in Practice , 1996 .

[11]  Lawrence Carin,et al.  Hidden Markov models for multiaspect target classification , 1999, IEEE Trans. Signal Process..

[12]  Torbjørn Svendsen,et al.  Maximum likelihood modelling of pronunciation variation , 1999, Speech Commun..

[13]  Sean R. Eddy,et al.  Biological Sequence Analysis: Probabilistic Models of Proteins and Nucleic Acids , 1998 .

[14]  John M. Olin.  Calculating posterior distributions and modal estimates in Markov mixture models , 1996 .

[15]  David R. Brillinger,et al.  Some wavelet-based analyses of Markov chain data , 2000, Signal Processing.

[16]  M. Stephens.  Dealing with label switching in mixture models , 2000 .

[17]  K OrJ.  Numerical Bayesian methods applied to signal processing , 1996 .

[18]  Subhrakanti Dey,et al.  Blind equalization of IIR channels using hidden Markov models and extended least squares , 1995, IEEE Trans. Signal Process..

[19]  Christophe Andrieu,et al.  An Introduction to Monte Carlo Methods for Bayesian Data Analysis , 2001 .

[20]  L. Carin,et al.  Target identification with wave-based matched pursuits and hidden Markov models , 1999 .

[21]  William D. Penny,et al.  Dynamic Models for Nonstationary Signal Segmentation , 1999, Comput. Biomed. Res..

[22]  Holger Stahl,et al.  Speech understanding and speech translation by maximum a-posteriori semantic decoding , 1999, Artif. Intell. Eng..

[23]  Michael J. Korenberg,et al.  Parallel cascade identification as a means for automatically classifying protein sequences into structure/function groups , 2000, Biological Cybernetics.

[24]  K. Riedel.  Numerical Bayesian Methods Applied to Signal Processing , 1996 .

[25]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[26]  C. Robert,et al.  Bayesian estimation of hidden Markov chains: a stochastic implementation , 1993 .

[27]  S. V. Vaseghi.  State duration modelling in hidden Markov models , 1995, Signal Process..

[28]  Christian P. Robert,et al.  Bayesian Inference in Hidden Markov Models through Jump Markov Chain Monte Carlo , 1999 .

[29]  Rong Chen,et al.  Blind restoration of linearly degraded discrete signals by Gibbs sampling , 1995, IEEE Trans. Signal Process..

[30]  T. Rydén,et al.  Stylized Facts of Daily Return Series and the Hidden Markov Model , 1998 .

[31]  Jianxin Wang,et al.  Spatio-temporal target identification method of high-range resolution radar , 2000, Pattern Recognit..