Title :
Continuously evolving classification of signals corrupted by an abrupt change
Author :
Robert, Thierry ; Tourneret, Jean Yves
Author_Institution :
ENSEEIHT, Toulouse, France
Abstract :
Bayes decision theory is based on the assumption that the decision problem is posed in probabilistic terms, and that all of the relevant probability values are known. The aim of this paper is to show how blind sliding window AR modeling is corrupted by an abrupt model change and to derive a statistical study of these parameters
Keywords :
Bayes methods; autoregressive processes; decision theory; random processes; statistical analysis; Bayes decision theory; abrupt model change; blind sliding window AR modeling; continuously evolving signal classification; corruption; decision problem; probabilistic terms; statistical study; Decision theory; Equations; Pattern recognition; Predictive models; Probability density function; Random processes; Shape; Signal processing; Statistics; Vectors;
Conference_Titel :
Information Theory and Statistics, 1994. Proceedings., 1994 IEEE-IMS Workshop on
Conference_Location :
Alexandria, VA
Print_ISBN :
0-7803-2761-6
DOI :
10.1109/WITS.1994.513924
