Abstract
France Télécom’s interest for the French telephone network is to secure a constant maximal flow of traffic. This is why a real time network traffic management is desired, which requires the knowledge of the system state at each moment. In order to perform this purpose a diagnostic approach, based upon fuzzy pattern recognition, to the detection and following of the state evolution of a dynamic system is presented. As an example, this method is used to diagnose, in real time, the functional state of an exchange in a model of the French long distance telephone network.
Résumé
Le souci de France Télécom est d’assurer en permanence un écoulement maximal du trafic dans le réseau téléphonique franÇais. C’est pourquoi une gestion en temps réel de ce trafic est recherchée, qui nécessite à tout instant la connaissance de l’état du système. Afin d’atteindre cet objectif, une approche de type diagnostic par reconnaissance des formes floues est présentée pour la détection et le suivi de l’évolution d’un système dynamique. Dans le cadre d’un exemple, cette méthode est appliquée au diagnostic en temps réel de l’état d’un autocommutateur de transit dans un modèle du réseau téléphonique franÇais à longue distance.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Dubuisson (B.), Masson (M.). A statistical decision rule with incomplete knowledge about classes. Pattern Recognition (1993),26, n° 1, pp. 155–165.
Fukunaga (K.). Introduction to statistical pattern recognition. Second edition,Academic Press, San Diego (1990).
Dubuisson (B.). Decision with reject option. In5th European Signal Processing Conference, Barcelone (1990).
Boutleux (E.), Dubuisson (B.). A decision system to detect a state evolution of a complex system. InProc. IEEE Systems, Man and Cybernetics, Vancouver, Canada (Oct. 1995),1, pp. 742–747.
Didelet (E.), Dubuisson (B.). USing neural trees for the diagnosis of the French long distance transit network. In Proc. Artificial Neural Networks in Engineering (ANNIE’91), Saint-Louis, Missouri (1991).
Stern (D.). Methods to detect traffic disturbances for realtime network management. European Cooperation on Network Traffic Management (1993).
Mandal (D. B.), Murthy (C. A.), Pal (S. K.). Formulation of a multivalued recognition system. IEEE Trans. SMC (1992),22, n° 4, pp. 607–620.
Mandal (D. B.), Murthy (C. A.), Pal (S. K.). Theoretical performance of a multivalued recognition system. IEEE Trans. SMC (1994),24, n° 7, pp. 1001–1021.
Zadeh (L. A.). Outline of a new approach to the analysis of complex systems and decision process. IEEE Trans. SMC (1973),3, pp. 28–44.
Boutleux (E.), Dubuisson (B.). Fuzzy pattern recognition to characterize evolutionary complex systems, application to the French telephone network. In Proc. Fifth IEEE Int. Conf. on Fuzzy Systems, New Orleans, USA (Sep. 1996).
Boutleux (E.), Dubuisson (B.). Learning membership functions to follow a system state evolution, application to a model of the French telephone network. In Proc. Conf. on Computational Engineering in Systems Applications, Lille, France (July 1996).
Author information
Authors and Affiliations
Additional information
This work has been supported by France Télécom cnet (Centre national d’études des télécommunications, France Telecom’s research centre paa/atr) under grant 93 1B 142, project n° 505.
Rights and permissions
About this article
Cite this article
Boutleux, E., Dubuisson, B. Fuzzy pattern recognition to characterise a system evolution. Application to a model of the French telephone network. Ann. Télécommun. 51, 509–520 (1996). https://doi.org/10.1007/BF02997713
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02997713