Detection of change in periodic, nonstationary data

Author

Gruner, C.M. ; Johnson, D.H.

Author_Institution

Dept. of Electr. & Comput. Eng., Rice Univ., Houston, TX, USA

Volume

5

fYear

1996

fDate

7-10 May 1996

Firstpage

2471

Abstract

Traditional change detection strategies are limited in cases where the data are nonstationary and the distributions are unknown. We present an algorithm for change detection problems in which we do not know the form of the distribution. Our algorithm uses distributed detection with a bank of type-based front end detectors that achieve asymptotically optimal type I error performance. Our simulations indicate that this algorithm performs much better than traditional methods

Keywords

error statistics; optimisation; signal detection; statistical analysis; algorithm; asymptotically optimal type I error performance; change detection; distributed detection; distributions; front end detectors; nonstationary data; periodic data; simulations; Change detection algorithms; Data engineering; Detectors; Information technology; Maximum likelihood detection; Maximum likelihood estimation; Probability distribution; Statistical analysis; Statistical distributions; Testing;

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.1996.547964

Filename

547964