Research And Application Of Alarm Relativity Analysis Mechanism In Network Management Systems

With the continual expansion of the scale of network, and the complexity of the network increasing, requirements for high quality service from customer is increasing,. Traditional div-network management mode can't accommodate new requirements. Telecommunication service providers require a integrated, centralized, intelligent management system to be constructed.Local telecommunication network is connected by quite a lot of equipment and link, any default of a component may communicate with several other equipments, and cause many children alarm. Therefore, it requires that using advanced computer tech to research and develop alarm relativity analysis mechanism for finding , locating and deposing default quickly and accurately. It's one of the core tasks of constructing local communication network integrated management that finding root alarm and alarms derive from root alarm by building alarm tree via alarm relativity analysis.The article firstly presents basic conceptions, correlated knowledge and actuality of communication management network. And the article analysis the TMN standard and TMN-based integrated network management deep, introduces the conception of data mining briefly, illustrates association rule mining algorithm and basic methods used in correlation analysis of communications net in detail.Secondly, it labors relativity characters of alarm data, compares usable relative analysis method on alarm, and probes into some means on disposing correlative alarm.Lastly based on the research results, this article puts forward relative analysis model of alarm based on association rule mining algorithm, and model-based correlative alarm disposal mechanism. Further more, it illuminates implement of alarm relative mechanism and its application effects.In our country , it is not a long time to start the research and exploitation of local telecommunication integrated network management . And it is shorter to apply data mining and knowledge base technology to alarm relativity analysis. So the research content and results have fine academic, technical and practical value. Here's the innovations of this article:· To put forward relativity analysis model of alarm based on association rule database.· To brings forward a relativity- rule creating algorithm which combining Apriori algorithm with event slide window. The algorithm can root out relating alarm set and build relativity analysis rules automatically.· To adopts model-based alarm relativity disposal mechanism, to realize convergence of alarm effectively, also flexibly configuring and implement of relative alarm disposal functions.