Data fusion with a faulty sensor

Author

Akkihal, Shivanand I. ; Wise, Gary L.

Author_Institution

Dept. of Electr. & Comput. Eng., Texas Univ., Austin, TX, USA

Volume

1

fYear

1995

fDate

21-23 Jun 1995

Firstpage

638

Abstract

A set theoretic approach to data fusion when one of the sensors is faulty is considered in this paper. If the fusion center implements a likelihood ratio test and if it does not have any knowledge of this fault, the performance of the system is different than its nominal performance. The changes in the error probabilities depend on the type of the fault and on the threshold value of the fusion center test. Expressions for the changes in the values of the error probabilities are derived for the cases of dissimilar and similar sensors. It is shown that the system false alarm probability increases significantly for some type of faults, whereas the system detection probability decreases significantly for some other faults. A numerical example is given to illustrate the results

Keywords

decision theory; probability; random processes; sensor fusion; set theory; data fusion; dissimilar sensors; error probabilities; false alarm probability; faulty sensor; fusion center test; set theoretic approach; similar sensors; threshold value; Fault detection; Sensor fusion; Sensor systems; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

American Control Conference, Proceedings of the 1995

Conference_Location

Seattle, WA

Print_ISBN

0-7803-2445-5

Type

conf

DOI

10.1109/ACC.1995.529328

Filename

529328