Bootstrap modeling of a class of nonstationary signals

Author

Zoubir, Abdelhak M. ; Iskander, D. Robert

Author_Institution

Sch. of Electr. & Comput. Eng., Curtin Univ. of Technol., Perth, WA, Australia

Volume

48

Issue

2

fYear

2000

fDate

2/1/2000 12:00:00 AM

Firstpage

399

Lastpage

408

Abstract

The problem of modeling polynomial-phase signals is considered. Techniques based on the bootstrap and multiple hypotheses tests for optimal model selection of constant amplitude polynomial-phase signals embedded in stationary noise are developed. Phase parameter estimators based on both the least-squares method and the polynomial phase transform are used. The proposed techniques are compared with existing ones, including Akaike´s (1978, 1974) information criterion and Rissanen´s (1978) minimum description length criterion and are shown to outperform these procedures for the considered small sample sizes

Keywords

least squares approximations; noise; phase estimation; polynomials; signal sampling; transforms; Akaike´s information criterion; Rissanen´s minimum description length criterion; bootstrap modeling; constant amplitude polynomial-phase signals; least-squares method; multiple hypotheses tests; nonstationary signals; optimal model selection; phase parameter estimators; polynomial phase transform; polynomial-phase signals; small sample sizes; stationary noise; Australia; Doppler radar; Finite impulse response filter; Frequency modulation; Marine technology; Parameter estimation; Polynomials; Radar imaging; Sonar; Testing;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/78.823967

Filename

823967