Performance analysis for a changepoint problem

Author

Hibey, Joseph L. ; Charalambous, Charalambos D.

Author_Institution

Dept. of Electr. Eng., Colorado Univ., Denver, CO, USA

Volume

3

fYear

1997

fDate

4-6 Jun 1997

Firstpage

1365

Abstract

Nonlinear stochastic differential equations are used to model a version of the changepoint problem. State estimates of the minimum-mean-square-error type are used in likelihood-ratio tests to detect the time of change. We then evaluate how well the system performs by deriving theoretical expressions for the false alarm probability and a special case of the miss error probability. The approach uses a Fokker-Planck equation. The overall methodology is general and should be of interest in other applications

Keywords

approximation theory; fault diagnosis; filtering theory; nonlinear differential equations; probability; state estimation; Fokker-Planck equation; changepoint problem; false alarm probability; likelihood-ratio tests; minimum-mean-square-error estimates; miss error probability; nonlinear stochastic differential equations; performance analysis; state estimates; Differential equations; Error probability; Filtering; Kalman filters; Mathematical model; Performance analysis; State estimation; Stochastic processes; Testing; Yield estimation;

fLanguage

English

Publisher

ieee

Conference_Titel

American Control Conference, 1997. Proceedings of the 1997

Conference_Location

Albuquerque, NM

ISSN

0743-1619

Print_ISBN

0-7803-3832-4

Type

conf

DOI

10.1109/ACC.1997.610641

Filename

610641