Optimized Intrusion Detection Simulation on Multi- tier Distributed and Large Differences Database
Yandong Zheng · Jisuanji fangzhen · 2013
In this paper,the accurately intrusion detection problems in multi- tier distributed and large differences database were researched. Different from the traditional single- database,there exists a big difference between multi- level data in the multi- tier database,and the difference is small categorical attributes. In the traditional database intrusion detection process,it only simply considers the point- to- point data detection between the layers,without considering the similarities between the layers,and gives up the optimization for detected attribute by application level classification,resulting in accurate detection is not high. To avoid these shortcomings of traditional algorithm,this paper presented an intrusion detection method based on particle swarm identification tree algorithm for multi- tier distributed and large differences database. We extracted the features of a computer database intrusion and used it as a basis for the database,then created a PSO to identify the tree,treated the point by layer,and through the probability operation of the database on the different layers in intrusion detection,achieved intrusion detection in multi- tier distributed and larger differences database. The experimental results show that the algorithm applied in intrusion detection of multilayer distributed and large difference database can improve the security of the database and ensure the safe operation of database.