Electrical and Computer Engineering
2332 Main Mall
Vancouver,
BC
Canada
V6T 1Z4
Tel +1.604.822.2872
Fax +1.604.822.5949
Email:
info@ece.ubc.ca
| Title | On-line identification of hidden Markov models via recursive prediction error techniques |
| Publication Type | Journal Article |
| Year of Publication | 1994 |
| Authors | Collings, I. B., V. Krishnamurthy, and J. B. Moore |
| Journal | Signal Processing, IEEE Transactions on |
| Volume | 42 |
| Pagination | 3535 -3539 |
| Date Published | dec. |
| ISSN | 1053-587X |
| Keywords | discrete state values, error analysis, finite-discrete set, hidden Markov models, HMM, initializations, Markov chain, noise, noise density, observations, on-line identification, parameter estimation, parameter identification, prediction theory, probability, recursive estimation, recursive Kullback-Leibler algorithm, recursive prediction error techniques, signal model, signal processing, simulation studies, transition probabilities |
| Abstract | An on-line state and parameter identification scheme for hidden Markov models (HMMs) with states in a finite-discrete set is developed using recursive prediction error (RPE) techniques. The parameters of interest are the transition probabilities and discrete state values of a Markov chain. The noise density associated with the observations can also be estimated. Implementation aspects of the proposed algorithms are discussed, and simulation studies are presented to show that the algorithms converge for a wide variety of initializations. In addition, an improved version of an earlier proposed scheme (the Recursive Kullback-Leibler (RKL) algorithm) is presented with a parameterization that ensures positivity of transition probability estimates |
| URL | http://dx.doi.org/10.1109/78.340791 |
| DOI | 10.1109/78.340791 |