Weighted least squares/MFT algorithms for linear differential system identification

Author

Pearson, A.E. ; Shen, Yury

Author_Institution

Div. of Eng., Brown Univ., Providence, RI, USA

fYear

1993

fDate

15-17 Dec 1993

Firstpage

2032

Abstract

Based on two different stochastic signal models, weighted least squares and adaptive weighted least-squares algorithms are developed for linear differential system parameter identification, both utilizing a modulating function technique (MFT). Comparison is made with the well known prediction error method showing that the MFT algorithms give smaller biases and standard deviations for the estimated parameters over a broad range of noise levels

Keywords

functional analysis; identification; least squares approximations; linear differential equations; maximum likelihood estimation; series (mathematics); white noise; Fourier series; SISO differential equation system; adaptive weighted least-squares; biases; linear differential system identification; maximum likelihood estimator; modulated white Gaussian noise; modulating function set; noise levels; parameter estimation; stochastic signal models; Differential equations; Frequency; Least squares methods; Parameter estimation; Reduced order systems; Signal processing; Signal processing algorithms; Stochastic resonance; Stochastic systems; System identification;

fLanguage

English

Publisher

ieee

Conference_Titel

Decision and Control, 1993., Proceedings of the 32nd IEEE Conference on

Conference_Location

San Antonio, TX

Print_ISBN

0-7803-1298-8

Type

conf

DOI

10.1109/CDC.1993.325555

Filename

325555