Distributed Network Alarm Weighted Association Rules Mining System Research And Design

Fault diagnosis and localization is the vital core of the network fault management. When the faults take place in the networks, it is necessary to find the locations and the causations of the faults accurately as soon as possible in order to get rid of the faults and recover the networks'function in time. The alarm correlation analysis is an important approach of fault diagnosis, which plays a crucial role in network fault management. Data mining provides a new approach of the knowledge updating during the alarm correlation analyzing. The modern communication networks and its network management is a typical distributed system. The realization of management fuction is done by different management components which in different levels to cooprate with each other. And some of the alarm's attributes has different levels, and services with different QoS requests need different treatment of alarms. With the background of the project: the alarm correlation in telecommunication networks based on data mining, which is supported by National Natural Science Foundation of China, this thesis focuses on mining weighted alarm association rules in distributed communication networks, including the alarm's pretreatment, mining weighted association rules, algorithm of distributed weighted association rules and the simulation and validation of the system of alarm weighted association rules distributed mining in communication networks.The alarm's pretreatment is carried out by expert system. The analytic hierarchy process is applied to deciding the weight of the alarm. The problem of alarm's synchronization is settled by setting time window and the slip length. The attributes of the alarm that reflect the faults were picked out to form an item of an alarm transaction and the redundant alarms are got rid of by alarm compressing. By the alarm's pretreatment expert system, the alarm database could be transformed into alarm transaction database, which is suitable for distributed weighted association rules mining.Based on the existent algorithms of weighted association rules and distributed association rules, aimed at the characters of communication networks: the vast amount of alarm data which could burst out suddenly and distributed alarm databases, a new algorithm of distributed weighted association rules named DWAP is proposed, which could mine alarm global weighted association rules effectively. The shared pattern distributed structure is introduced, so the alarm global weighted association rules mining are achieved by the cooperation between local stations and global station. An improved algorithm WAP based on weighted association pattern tree runs on the local stations for mining local weighted association rules. It is unnecessary to traverse the database repeatedly and construct the conditional pattern tree recursively. Algorithm DWAP assigns the weight of an alarm transaction according to cumulation weighted method and ratio weighted method, which could reflect the importance of the alarm transaction perfectly and make the result with a high dipartite degree. Moreover, this algorithm adopts an effective iterative and pruning strategy that could compress the scale of candidate patterns and reduce the communication cost. The performance test of the algorithm DWAP indicates that it has high time efficiency, low communication cost and strong retractility. It is also valuable and useful for the alarm correlation analysis and fault diagnosis and localization in communication networks.