Adaptive Bayesian spectrum estimation

Author

Demoment, G. ; Houacine, A. ; Herment, A. ; Mouttapa, I.

Author_Institution

CNRS, Gif-sur-Yvette, France

fYear

1988

fDate

3-5 Aug 1988

Firstpage

33

Lastpage

38

Abstract

A method is described for the numerical analysis of nonstationary signals whose spectral distributions vary slowly enough with time for an adaptive method to be used. The data is processed by blocks. The theory describes the signal by an autoregressive series of high order with stochastic coefficients having a Gaussian distribution. The coefficients are considered as state variables and can thus be estimated in real time by a special fast Kalman filter. This raises the problem of determining an essential tuning parameter which is a regularization coefficient. This is done according to maximum likelihood. Numerical experiments on synthetic signals have demonstrated the feasibility of the method

Keywords

Bayes methods; Kalman filters; numerical analysis; spectral analysis; stochastic processes; tuning; Gaussian distribution; adaptive Bayesian spectrum estimation; autoregressive series; fast Kalman filter; maximum likelihood; nonstationary signals; numerical analysis; spectral distributions; state variables; stochastic coefficients; synthetic signals; tuning parameter; Bayesian methods; Cardiology; Filters; Gaussian distribution; Maximum likelihood estimation; Numerical analysis; Signal processing; Spectral analysis; State estimation; Stochastic processes;

fLanguage

English

Publisher

ieee

Conference_Titel

Spectrum Estimation and Modeling, 1988., Fourth Annual ASSP Workshop on

Conference_Location

Minneapolis, MN

Type

conf

DOI

10.1109/SPECT.1988.206158

Filename

206158