Title :
Algorithms for treatment of locally stationary multichannel signals-proposition of a method for change detection based on histograms
Author :
Olivier, C. ; Courtellemont, P. ; de Brucq, D. ; Colot, O.
Author_Institution :
LACIS, Rouen Univ., Mont-Saint-Aignan, France
Abstract :
To take into account slow changes of characteristics in signal which are only locally stationary, an adaptive process for parameter modeling is carried out. An adaptive parametric modeling technique using windows is described. A general process with two thresholds for change detection is discussed. A more general method based on the histograms of the prediction errors is suggested
Keywords :
adaptive systems; filtering and prediction theory; signal processing; adaptive process; change detection; histograms; locally stationary multichannel signals; parameter modeling; prediction errors; windows; Change detection algorithms; Convergence; Histograms; Maximum likelihood detection; Maximum likelihood estimation; Parametric statistics; Performance evaluation; Rhythm; Signal processing; Testing;
Conference_Titel :
Industrial Electronics Society, 1990. IECON '90., 16th Annual Conference of IEEE
Conference_Location :
Pacific Grove, CA
Print_ISBN :
0-87942-600-4
DOI :
10.1109/IECON.1990.149144
