基于概率模糊认知图的Mstream攻击检测方法  

Mstream Attack Detection Approach Based on Probabilistic Fuzzy Cognitive Map

作　　者：杨锋[1] 钟诚[1] 李智[1,2] 

机构地区：[1]广西大学计算机与电子信息学院 [2]广西科技厅网络中心,南宁530012

出　　处：《计算机工程》2006年第10期125-127,共3页Computer Engineering

基　　金：广西科学基金资助项目(桂科自0339008)

摘　　要：提出了一种分布式拒绝服务攻击Mstream的检测方法。设计基于概率模糊认知图(PFCM)的Mstream攻击图,并以该图描述Mstream攻击,利用概率测度表述事件间的时序关系、约束关系和关系的不确定性,结合模糊理论设计基于PFCM的Mstream攻击检测算法。实验表明,该方法不但能准确检测出分布式拒绝服务攻击Mstream,而且还能判断出正在进行中的Mstream攻击的程度。This paper presents an approach detecting the distributed denial of services attack Mstream. Based on the probabilistic fuzzy cognitive map （PFCM）, the Mstream attack map is constructed to describe Mstream attack. This attack map expresses the temporal sequence, the restriction and the uncertainty of the thctor relations by applying probability measure. By applying the fuzzy theory, a PFCM-based Mstream attack detection algorithm is designed. The test results show that the aooroach can detect Mstream attack and judge the degree of Mstream attack.

关 键 词：分布式拒绝服务 入侵检测 概率模糊认知图 Mstream攻击 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]