Ontology based alarm correlation technology in TD-SCDMA network

Abstract

The faults of some components in the TD-SCDMA network can cause failures of equipments and links. The failures will cause many derivative alarms in equipments. In order to find out the faults immediately and accurately, researches on the methods and mechanisms of the alarms' correlation have been done. In order to reduce the number of alarms and enhance the contents of them, this paper presents a new way to tackle the correlating problem. First, the ontology of the alarms has been constructed, which realized the knowledge sharing, reusing efficiently. Second, the attributes' similarity and semantic distance have be taken into consideration when computing the semantic similarity. Finally the key algorithm has been analyzed and the corresponding testing results have been gotten. The results showed that the similarity algorithm can compute the similarity, compress the repeated alarms and correlate similar alarms accurately and efficiently. Copyright © 2013 Binary Information Press.