Asymptotic distribution of recursive subspace estimators

Author

Yang, Ban ; Gersemsky, Frank

Author_Institution

Dept. of Electr. Eng., Ruhr-Univ., Bochum, Germany

Volume

5

fYear

1996

fDate

7-10 May 1996

Firstpage

2856

Abstract

We derive the asymptotic distribution of recursive subspace estimators. In particular, we study the PAST algorithm for tracking the signal subspace and the Oja (1982) rule for updating the eigenvector corresponding to the largest eigenvalue. Both the decreasing gain and the constant gain case are considered. It turns out that their asymptotic distributions differ from that of the batch eigenvalue decomposition. The asymptotic rate of convergence is also addressed

Keywords

convergence of numerical methods; eigenvalues and eigenfunctions; recursive estimation; signal processing; statistical analysis; tracking; Oja rule; PAST algorithm; asymptotic convergence rate; asymptotic distribution; asymptotic statistics; batch eigenvalue decomposition; constant gain; decreasing gain; eigenvector updating; recursive subspace estimators; signal processing; signal subspace tracking; Convergence; Covariance matrix; Eigenvalues and eigenfunctions; Performance analysis; Recursive estimation; Signal processing; Signal processing algorithms; Singular value decomposition; Statistical distributions; Statistics;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1996. ICASSP-96. Conference Proceedings., 1996 IEEE International Conference on

Conference_Location

Atlanta, GA

ISSN

1520-6149

Print_ISBN

0-7803-3192-3

Type

conf

DOI

10.1109/ICASSP.1996.550149

Filename

550149