Contextual hidden Markov models for wavelet-domain signal processing
- 21 June 2006
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1 (10586393) , 95-100
- https://doi.org/10.1109/acssc.1997.680036
Abstract
Wavelet-domain hidden Markov models (HMMs) provide a powerful new approach for statistical modeling and processing of wavelet coefficients. In addition to characterizing the statistics of individual wavelet coefficients, HMMs capture some of the key interactions between wavelet coefficients. However, as HMMs model an increasing number of wavelet coefficient interactions, HMM-based signal processing becomes increasingly complicated. In this paper, we propose a new approach to HMMs based on the notion of context. By modeling wavelet coefficient inter-dependencies via contexts, we retain the approximation capabilities of HMMs, yet substantially reduce their complexity. To illustrate the power of this approach, we develop new algorithms for signal estimation and for efficient synthesis of nonGaussian, long-range-dependent network traffic.Keywords
This publication has 10 references indexed in Scilit:
- Signal estimation using wavelet-Markov modelsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2006
- Wavelet-based statistical signal processing using hidden Markov modelsIEEE Transactions on Signal Processing, 1998
- Adaptive Bayesian Wavelet ShrinkageJournal of the American Statistical Association, 1997
- Which stochastic models allow Baum-Welch training?IEEE Transactions on Signal Processing, 1996
- Adapting to Unknown Smoothness via Wavelet ShrinkageJournal of the American Statistical Association, 1995
- On the self-similar nature of Ethernet traffic (extended version)IEEE/ACM Transactions on Networking, 1994
- Fractal estimation from noisy data via discrete fractional Gaussian noise (DFGN) and the Haar basisIEEE Transactions on Signal Processing, 1993
- Wavelet analysis and synthesis of fractional Brownian motionIEEE Transactions on Information Theory, 1992
- Ten Lectures on WaveletsPublished by Society for Industrial & Applied Mathematics (SIAM) ,1992
- A Karhunen-Loeve-like expansion for 1/f processes via waveletsIEEE Transactions on Information Theory, 1990