多层分布式大型差异数据库优化入侵检测仿真

发布时间：2018-05-05 23:50

本文选题：分布式 + 数据库　；参考：《计算机仿真》2013年11期

【摘要】：研究多层分布式大型差异数据库入侵准确检测的问题。与传统的单层数据库不同,多层数据库的层次之间的数据存在较大的差异,分类属性的差异化较小,传统的数据库入侵检测过程只是简单考虑了层与层之间的点对点数据检测,没有考虑层与层之间的相似性,放弃了应用层次间分类对检测属性的优化,导致检测准确性不高。为了避免上述传统算法的缺陷,提出了一种基于粒子群辨别树算法的多层分布式大型差异数据库入侵检测方法。提取计算机数据库的入侵特征,并且将其作为数据库入侵检测的依据。建立粒子群辨别树,对节电进行分层处理,通过运算不同层上的数据库入侵检测的概率,从而实现多层分布式大型差异数据库的入侵检测。实验结果表明,利用上述算法进行多层分布式大型差异数据库的入侵检测,可以提高数据库的安全性,保证数据库的安全运行。
[Abstract]:The problem of accurate intrusion detection in multi-tier distributed large-scale differential database is studied. Different from the traditional single-layer database, there are great differences between the layers of the multi-layer database, and the differentiation of the classification attributes is small. The traditional database intrusion detection process only considers the point-to-point data detection between the layers. Without considering the similarity between the layers, we give up the optimization of the detection attributes by using the classification between the layers, which leads to the low accuracy of the detection. In order to avoid the defects of the traditional algorithms mentioned above, a multi-layer distributed intrusion detection method based on particle swarm discrimination tree (PSO) algorithm is proposed in this paper. The intrusion feature of computer database is extracted and used as the basis of database intrusion detection. A particle swarm discriminant tree is established to deal with the power saving in a hierarchical way. By calculating the probability of database intrusion detection on different layers, the intrusion detection of multi-layer distributed large-scale differential database is realized. The experimental results show that the above algorithm can improve the security of the database and ensure the safe operation of the database.
【作者单位】：长春师范大学网络中心;
【分类号】：TP393.08

【参考文献】