A recursive approach for alarm correlation in telecommunication networks

• Published in: Network Operations and Management Symposium (Noms) pp. 985 - 986
• Main Authors: Meira, D.M., Nogueira, J.M.S.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 2000
• Subjects: Bayesian methods; Electric breakdown; Electronic mail; Fault diagnosis; Intelligent networks; NP-complete problem; Parallel processing; Predictive models; Telecommunication switching; Uncertainty
• ISBN: 9780780359284; 0780359283
• DOI: 10.1109/NOMS.2000.830469

A model for alarm correlation in a telecommunication network is proposed, which contributes to the identification of possible causes for the alarms received, as well as the prediction of possible effects of some known faults. The model allows the partition of the telecommunication network into several smaller sub-networks, while preserving some basic information on the propagation of the effect of faults among sub-networks. The breakdown of the problem into smaller sub-problems facilitates its solution and allows the use, in each sub-network, of the correlation technique most suitable to their peculiarities. It also allows the execution of several correlation processes in parallel, increasing the speed of the overall correlation process and reducing the complexity of the problem. This approach named "multi-focal correlation", may be recursively utilized in each sub-network until the network element level is reached.