Title :
Auxiliary signal design for detecting changes towards unknown hypotheses
Author :
F. Kerestecioglu;I. Cetin
Author_Institution :
Dept. of Electr. & Electron. Eng., Bogazici Univ., Istanbul, Turkey
Abstract :
The effects of auxiliary input signals on detecting changes in CARMA models via statistical tests are discussed. It is assumed that parameters of the system before the change are known whereas those describing the after-change situation are unknown. The performance criteria for the input design are based on the local and asymptotic properties of the test. It is concluded that the optimal input spectrum consists of a finite number of frequencies and that a suitable choice for the input can drastically improve the detection performance.
Keywords :
"Signal design","Change detection algorithms","Signal detection","Event detection","Algorithm design and analysis","Sequential analysis","Electronic equipment testing","Frequency","System testing","Electrical equipment industry"
Conference_Titel :
Intelligent Control, 1997. Proceedings of the 1997 IEEE International Symposium on
Print_ISBN :
0-7803-4116-3
DOI :
10.1109/ISIC.1997.626475
