网络ARP伪装攻击过滤方法研究与仿真

摘要

针对网络攻击优化识别,研究网络ARP攻击过滤问题,提高网络攻击过滤的准确性.由于网络攻击过滤过程,都是以攻击特征分类的结果作为过滤的基础的,但是ARP网络攻击会发生特征伪装行为,形成和正常的数据相似的特征,造成过滤攻击特征时,很难准确分类,过滤ARP攻击准确度不高的问题.为了解决上述问题,提出一种改进的代价敏感决策树的ARP伪装攻击分类方法.以代价减少的多少为衡量标准进行ARP伪装攻击分类的迭代计算.通过结点分裂属件的选择,对伪装攻击进行识别分类.克服传统方法的弊端.实验结果表明,改进方法能够大幅提高ARP欺骗攻击分类结果的准确性,降低分类错误率,取得了很好的效果.%Network attacks are complex and has many levels. This paper put forward a RP camouflage attack classification algorithm based on cost sensitive decision tree to implement the iterative calculation of the ARP attack classification according to the cost reducing. Through the choice of nodes split attributes, camouflage attack classification identification was carried out. The experiment results show that the improved algorithm can improve the accuracy of identification.