Research On Alerts Correlation Method Of TMN Based On Bayesian Network

Today, with the booming development of telecommunication industry,the telecommunication network management technology tends to becomprehensive, distributed and intelligent. As a prevailing international standard architecture of telecommunication network, Telecommunications Management Network (TMN) has been widely adopted by more and more network management system. TMN is presented by ITU-T measuring up the M.3010 series standard. In part, it inherited the some ideas of the OSI, such as the concept of Manager and Agent, which reflects the idea of the separation between the management network and the managed network, and the idea of the hierarchical management, and also realizes the effective management on fault, configuration, performance, safety and accounts.Fault management is one of the most basic functions of network management. In the network management of telecommunication, the large amount of alarm is the key factor influencing the capability of the network management and the stability of the system. So the implementation of the fault management system is the main problem in the expanding of the network scale. The introduction of the alarm correlation module solves the problem of the large amount of data, and avoiding the possibility of the fault management system becoming the bottleneck of the capability.On the basis of the implementation necessity of alarm correlation in TMN, in this thesis, several methods that implement the alarm correlation are compared. Based on analyzing the relationship of network alarms and faults, a vertical correlation method utilizing Bayesian network reasoning technique for dealing with the local alarm event including application layer alarm events is proposed. At last, the concrete realization method is given after descripting models of protocol layer entities.