Telecommunications Network Alarm Correlation Analysis Method

Telecommunication networks are very large and complex, they must work continuously. So, it is difficult to manage such communication networks, especially in local centralized operating and management environment where there are more than twenty kinds of alarms. Usually a technician can only master one equipment and know others two or there equipments. In the face of such many types of alarms, technicians are demanded to rapidly and accurately estimate the source reason for the alarms, and then respond quickly. Nobody can do it by himself. There are two mainly shortages if the alarms are analyzed one by one by hand. On the one hand, a great deal of human resource is wasted because many technicians who have much maintenance experience and know the structure and devices of the communication network well are needed. On the other hand, the technicians will be misdirected when some alarms can not be sent to the network console, and it is difficult to satisfy the demands on such communication network by manual processing.This paper researches the methods of alarm correlation analysis, the techniques of merging relative alarms and translating a group of relative alarms into an alarm. In this way, an alarm can be sent to network monitor terminal instead of lots of relative alarms that network administrators can locate faults and filter alarms quickly and technicians can concentrate on processing filtered alarms.In this paper, the alarms from telecom networks are discussed first and then an association-rule-based model for telecom alarm correlation analysis is introduced. The model proposes an approach to combine the correlated alarms to form an alarm transaction. The approach not only solves the problem of asynchronization of correlated alarm in the network, but also reduces the number of rules in the result. Finally the rules are validated by the resource correlation.