Asymptotic stability of SMM and RGM for insufficient order IIR adaptive filters

Author

Nayeri, Majid ; Fan, Hong

Author_Institution

Dept. of Electr. Eng., Michigan State Univ., East Lansing, MI, USA

fYear

1990

fDate

3-6 Apr 1990

Firstpage

1265

Abstract

The asymptotic behaviors of RGM (recursive gradient method) and SMM (Steiglitz-McBride method) algorithms are studied for the insufficient order case. In particular, it is shown that under a set of sufficient conditions the estimates obtained by both methods are stable. The first-order IIR (infinite impulse response) adaptive filters are shown to have stable behavior for both methods irrespective of the unknown system parameters. An example in a class of adaptive filters is also presented to show instability for the SMM, whereas for the same class the RGM obtains stable solutions

Keywords

adaptive filters; digital filters; stability; IIR adaptive filters; RGM; SMM algorithms; Steiglitz-McBride method; asymptotic stability; first-order filter; insufficient order filter; recursive gradient method; sufficient conditions; system parameters; Adaptive filters; Adaptive signal processing; Additive noise; Asymptotic stability; Filtering algorithms; Gradient methods; IIR filters; Signal processing algorithms; Sufficient conditions; System identification;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1990. ICASSP-90., 1990 International Conference on

Conference_Location

Albuquerque, NM

ISSN

1520-6149

Type

conf

DOI

10.1109/ICASSP.1990.115602

Filename

115602