Change Detection and Non Stationary Signals Tracking by Adaptive Filtering

In this paper we consider the problem of change detection and non stationary signals tracking. Using parametric estimation of signals based on least square lattice adaptive filters we consider for change detection statistical parametric methods using likelihood ratio and hypothesis tests. In order to track signals dynamics, we introduce a compensation procedure in the adaptive estimation. This will improve the adaptive estimation performances and fasten it-s convergence after changes detection.

Authors:

• Mounira RouaÐùnia
• Noureddine Doghmane

Keywords:

• Change detection
• Hypothesis test
• likelihood ratioleast square lattice adaptive filters.

References:

[1] P. Diniz, Adaptive filtering: Algorithms and practical
implementation. Kluwer academic pub, 2002.
[2] M.Basseville, I.Nikiforov. Detection of abrupt changes: Theory
and application. Prentice Hall, 1997.
[3] M. Basseville. Detecting changes in signals and systems - A
survey-. Automatica, V-24, 1988. p 309-325.
[4] I. Nikiforov, Two strategies in the problem of change detection
and isolation. IEEE Trans on information theory. V43(2) 1997 p
770-776.
[5] M.Seck, R.Blouet, F.Bimbot. Comparaison de critères de
segmentation par détection de rupture sur un signal sonore.
17ème colloque GRETSI sur le traitement du signal et des
imagesV4(1)1999 p 989-992.
[6] G. V. Moustakides. Optimal stopping times for detecting
changes in distribution. The annals of statistics. V 14.1986 p
1479-1487.
[7] Ramdani. Détection de rupture: outil de diagnostic
d-adaptativité dans le cas non stationnaire. ICSS Alger, 1994. p
III45-III49.
[8] D. Siegmund, Sequential analysis tests and confidence intervals.
Series in statistics, Springer, 1985.