Multiresolution parametric spectral estimation

Author

Xie, Min ; Beex, A.A.

Author_Institution

Bradley Dept. of Electr. Eng., Virginia Tech, Blacksburg, VA, USA

fYear

1994

fDate

20-22 Mar 1994

Firstpage

432

Lastpage

436

Abstract

Parametric spectral estimation in conjunction with multiple band analysis is proposed for multiresolution spectral estimation. Several autoregressive estimators are compared in this multiresolution context, with the windowed recursive least squares algorithm providing better frequency tracking than the exponential forgetting least squares algorithm, and requiring fewer computations than the Burg algorithm. A multifrequency signal experiment shows that the multiresolution parametric spectral estimation method can provide better time and frequency resolution than the Morlet wavelet transform

Keywords

least squares approximations; parameter estimation; spectral analysis; stochastic processes; time series; autoregressive estimators; frequency tracking; multiple band analysis; multiresolution parametric spectral estimation; parametric spectral estimation; windowed recursive least squares algorithm; Discrete wavelet transforms; Fourier transforms; Frequency estimation; Least squares approximation; Parameter estimation; Signal analysis; Signal processing; Signal processing algorithms; Signal resolution; Wavelet transforms;

fLanguage

English

Publisher

ieee

Conference_Titel

System Theory, 1994., Proceedings of the 26th Southeastern Symposium on

Conference_Location

Athens, OH

ISSN

0094-2898

Print_ISBN

0-8186-5320-5

Type

conf

DOI

10.1109/SSST.1994.287838

Filename

287838