Advertisement

Description d'un detecteur sequentiel de changements brusques de dynamiques des modeles arma

  • D. Canon
  • C. Dongarli
Session 4 Detection Of Changes In Systems
  • 121 Downloads
Part of the Lecture Notes in Control and Information Sciences book series (LNCIS, volume 62)

Abstract

The sequential detection of abrupt changes in ARMA models is an important question in multi sensor/multi target tracking or similar problems (failure detection, E.E.G. analysis ...)

The authors propose an original sequential algorithm both testing the model and estimating its parameters simultaneously. This method is based upon a constant order hypothesis with abrupt changes within the parameters of an ARMA model. This realistic hypothesis is a consequence of the experimental capability of a low order ARMA model to represent correctly any Markovian process.

The identification of the system is sequentially performed by an Extended Kalman Filter which provides both the parameters of the model and a "pseudo-innovation" sequence.

That independant sequence is multiply by itself with a link of one step to define a decision variable and to detect a change of level.

Keywords

Extend Kalman Filter Nous Avons ARMA Model Realistic Hypothesis Premier Temp 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. BASSEVILLE, M. (1981 a), Edge detection using sequential methods for change in level. Part. II: Sequential detection of change in mean. I.E.E.E. Trans. on A.S.S.P., Vol. ASSP 29, no 1, pp. 32–50Google Scholar
  2. BASSEVILLE, M. (1982 a), Contribution à la détection séquentielle de ruptures de modèles statistiques. Thèse Docteur ès Sciences, RennesGoogle Scholar
  3. BASSEVILLE, M. and A. BENVENISTE (1982 b), Détection séquentielle de changements brusques des caractéristiques spectrales d'un signal numérique. Rapport INRIA, no 129Google Scholar
  4. BASSEVILLE, M., B. ESPIAU and J. GASNIER (1981 b), Edge detection using sequential method for change in level. Part. I: A sequential edge detection algorithm. IEEE Trans. on ASSP, ASSP 29, no 1, pp. 24–31Google Scholar
  5. HINKLEY, D.V. (1971), Inference about the change point from cumulative sum tests. Biometrica, 58, no 3, pp. 509–523Google Scholar
  6. MEHRA, R.K., and J. PEESCHON (1971), An innovation approach to fault detection and diagnosis in dynamic systems. Automatica, 7, p. 637–640Google Scholar
  7. SHIRYAEV, A.N. (1961), The problem of the most rapid detection of a disturbance in a stationnary process. Sov. Math. Dokl, no 2, pp. 795–799Google Scholar
  8. SHIRYAEV, A.N. (1963), On optimum methods in quickest detection problems. Theory Prob. Appl., 8, no 1, pp. 22–46Google Scholar
  9. SHIRYAEV, A.N. (1965), Some exact formulas in a disorder problem. Theory Prob. Appl., 10, no 3, pp. 348–354Google Scholar
  10. WILLSKY, A.S., CHOW, E.Y., GERSCHWIN, S.B., GREENE, C.S., HOUPT, P.K., and KURKJIAN, A.K. (1980), Dynamic model based techniques for the detection of insidients on Freeways. IEEE Trans. A.C., AC 25, no 3, pp. 347–360Google Scholar
  11. WILLSKY, A.S., and JONES, H.L. (1976), A generalized likelihood ratio approach to the detection and estimation of jumps in linear systems. IEEE Trans. A.C., AC 21, no 1, pp. 108–112Google Scholar

Copyright information

© Springer-Verlag 1984

Authors and Affiliations

  • D. Canon
    • 1
  • C. Dongarli
    • 1
  1. 1.Laboratoire D'AutomatiqueEcole Nationale Supérieure de MécaniqueNantes CédexFrance

Personalised recommendations