DocumentCode
1298442
Title
Parameter Identification of Sinusoids
Author
Hou, Mengshu
Author_Institution
Dept. of Eng., Univ. of Hull, Kingston upon Hull, UK
Volume
57
Issue
2
fYear
2012
Firstpage
467
Lastpage
472
Abstract
For a biased multi-sinusoidal signal an estimator of order 5n+1 is designed to identify all the 3n+1 parameters including the frequency, amplitude and phase for each of n sinusoids, and the overall offset. As an alternative, a fifth-order parameter estimator is also derived for a biased single sinusoid. The derivations of the estimators are straightforward by using some treatments in the existing frequency estimators and the standard gradient algorithm. Global asymptotic convergence of the parameter estimation is verified. The originality of the work is the proposal of a first solution to the problem of identifying full parameters of a biased multi-sinusoidal signal with globally asymptotic convergence of the parameter estimation.
Keywords
frequency estimation; signal processing; biased multisinusoidal signal; fifth-order parameter estimator; frequency estimators; globally asymptotic convergence; sinusoid parameter identification; standard gradient algorithm; Asymptotic stability; Convergence; Frequency estimation; Noise; Observers; Adaptive observer; frequency/amplitude/phase estimation; identification; signal processing; sinusoidal estimation;
fLanguage
English
Journal_Title
Automatic Control, IEEE Transactions on
Publisher
ieee
ISSN
0018-9286
Type
jour
DOI
10.1109/TAC.2011.2164736
Filename
5985482
Link To Document