Improved methods for exponential parameter estimation in the presence of known poles and noise

Author

Chen, Hua ; Van Huffel, Sabine ; Vandewalle, Joos

Author_Institution

Dept. of Electr. Eng., Eindhoven Univ. of Technol., Netherlands

Volume

45

Issue

5

fYear

1997

fDate

5/1/1997 12:00:00 AM

Firstpage

1390

Lastpage

1393

Abstract

By incorporating prior knowledge of known signal poles corresponding to some of the closely spaced sinusoids, Dowling et al. (see IEEE Trans. Antennas Propagat., vol.42, no.5, p.590-599, 1994) have proposed an LCTLS-FLP technique, which improves the performance of TLS-FLP. We present two better estimators that incorporate the same prior knowledge but clearly outperform LCTLS-FLP in resolution and parameter accuracy

Keywords

least squares approximations; noise; parameter estimation; poles and zeros; prediction theory; signal resolution; spectral analysis; LCTLS-FLP; TLS-FLP; closely spaced sinusoids; exponential parameter estimation; forward linear prediction; linear constraints; noise; parameter accuracy; signal poles; signal resolution; spectral estimation techniques; total least squares; Cepstral analysis; Deconvolution; Entropy; Nuclear magnetic resonance; Parameter estimation; Radar signal processing; Signal processing; Signal processing algorithms; Signal reconstruction; Signal resolution;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/78.575717

Filename

575717