Title :
Detection of changes in the spectrum of a multidimensional process
Author_Institution :
Univ. Paris-Sud, Orsay, France
fDate :
2/1/1993 12:00:00 AM
Abstract :
An algorithm is presented for the sequential detection of changes in the spectrum of a multidimensional process. The asymptotic properties of the statistic used are investigated in the case of a real Gaussian process. The algorithm of detection is based on a sequential likelihood-ratio test. Simulations show very good behavior of the algorithm in the case of Gaussian and non-Gaussian processes. In both cases, changes are detected with good accuracy, while the number of false alarms is small
Keywords :
Monte Carlo methods; random processes; spectral analysis; Monte Carlo simulation; Non Gaussian process; asymptotic properties; false alarms; multidimensional process; real Gaussian process; sequential detection; sequential likelihood-ratio test; spectrum change detection; Acoustic signal detection; Acoustic waves; Change detection algorithms; Distribution functions; Gaussian processes; Multidimensional systems; Sequential analysis; Statistical distributions; Statistics; Stochastic processes;
Journal_Title :
Signal Processing, IEEE Transactions on
