多源网络安全事件的准确评估算法研究  

Research on Multiple Source Network Security Events Accurate Assessment Algorithm

作　　者：刘平[1] 杨新锋[1] 

机构地区：[1]南阳理工学院计算机科学与技术系,河南南阳473000

出　　处：《计算机仿真》2012年第4期126-129,共4页Computer Simulation

摘　　要：研究网络安全事件的评估,由于恶意攻击严重影响网络安全,要求提高报警准确率。针对网络中各类安全设备的增多,产生了大量的安全事件,当安全事件评估结果不一致时,传统的统计方法的评估算法无法处理多源安全事件的不一致,造成漏警率较高的问题。提出了一种加权均值的评估算法,通过模糊聚类方法对多源安全事件融合,设定每个安全事件的权值并计算其加权平均作为评估指标,进行仿真,结果解决了多源安全事件评估不一致的问题。实验证明,改进方法能够有效融合多源安全事件并准确报警,保证了网络的安全,取得了满意的结果。Research network security incident evaluation to improve the alarm accuracy.All kinds of safety equipments are increasing in networks,producing a lot of security events.When these security incidents assessment results do not agree with each other,the traditional evaluation algorithm method based on statistics cannot handle the inconsistencies of multiple sources security incidents,causing higher false alarm rate.This paper put forward a method based on weighted mean of evaluation algorithm.Using the methods of fuzzy clustering for multi-source security incidents fusion,each security incident was set a weight and weighted average was calculated as evaluation index to solve the problem of inconsistent multi-source security incident assessments.Experiments show that the method can effectively fuse multi-source security incidents and give accurate alarm,which can guarantee the network security and obtain satisfactory results.

关 键 词：安全事件 加权均值 模糊聚类 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]