分布拒绝服务攻击(DDoS)通过很多代理产生大量的数据包,在很短的时间内就能耗尽受害者的计算和通信资源.通过研究和分析几种基于对DDoS攻击阶段分类的检测办法,得出基于聚类分析的算法是比较有效的,然而这种算法存在冗余.根据熵的特性对这种基于聚类分析的早期检测算法做了优化,对相关变量进行了关键变量的提取,并通过实验对其进行了分析,实验结果表明,对该算法的优化有效的提高了基于聚类分析的DDoS攻击检测方法的效率.%Distributed denial of service (DDoS) attacks generate enormous packets by a large number of agents and can easily exhaust the computing and communication resources of a victim within a short period of time. The conclusion is drawn that a method for the proactive detection based on cluster analysis is more effective after some methods for detection based on the classification of each phase of DDoS attack are researched and analyzed. However the algorithm has some redundancy. The algorithm based on cluster analysis is improved according to the property of entropy by extracting the key variables from the related variables. The result of the analysis on the experiment shows that the efficiency of the algorithm for detecting DDoS attacks based on cluster analysis is raised by improving the original algorithm.
展开▼
