Model-based diagnosis for fault management in ATM networks

ATM networks require powerful management capabilities for configuration, performance, security and particularly fault management. International standard organizations and particularly the ATM Forum and ITU-T organization have proposed several recommendations to manage faults. However, the large size and the high level of complexity of such networks requires new intelligent approaches. This article gives some examples of the existing formalisms and proposes an intelligent approach to build rule-based systems used to detect and to localize faults in ATM networks.