Minimum variance signal estimation with adaptive order statistic filters

Author

Clarkson, Peter M. ; Williamson, Geoffrey A.

Author_Institution

Dept. of Electr. & Comput. Eng., Illinois Inst. of Technol., Chicago, IL, USA

Volume

4

fYear

1992

fDate

23-26 Mar 1992

Firstpage

253

Abstract

The authors consider the estimation of a locally constant signal embedded in stationary noise with unknown statistics. They develop iterative algorithms, dubbed adaptive order statistic filters, designed to approximate the minimum variance unbiased order statistic estimator for the signal. The authors give conditions for convergence in the mean to the optimal estimator, discuss convergence rates, and present supporting simulations

Keywords

adaptive filters; convergence of numerical methods; filtering and prediction theory; iterative methods; noise; statistical analysis; adaptive order statistic filters; convergence rates; iterative algorithms; locally constant signal; minimum variance unbiased order statistic estimator; signal estimation; simulations; stationary noise; Adaptive filters; Convergence; Estimation; Iterative algorithms; Lagrangian functions; Noise measurement; Nonlinear filters; Q measurement; Statistics; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on

Conference_Location

San Francisco, CA

ISSN

1520-6149

Print_ISBN

0-7803-0532-9

Type

conf

DOI

10.1109/ICASSP.1992.226438

Filename

226438