基于树突状细胞理论的入侵检测模型研究

Research on intrusion detection model based on DCA

人工免疫入侵检测是当前主流的入侵检测技术之一,而危险理论中树突状细胞入侵检测方法是人工免疫研究的最新成果。建立危险理论树突状细胞入侵检测模型的关键是要解决危险信号的定义和表示,文章在免疫危险理论和树突状细胞理论基础上,使用多分类器算法动态提取危险信号,设计MC-DCA入侵检测模型,以提高抗原提呈、抗体识别的效率;并使用KDD CUP 99常用网络入侵检测数据,对构建的MC-DCA入侵检测仿真模型和传统AIS模型、DT模型进行对比和仿真实验,实验结果表明MC-DCA有更好的入侵检测识别能力。

Artificial immune intrusion detection （AIID） is one of the current major technology on intru-sion detection ,and the dendritic cell algorithm（DCA） in danger theory is the latest results of the re-search of AIID .The key to establish the DCA intrusion detection model based on danger theory is to define and express danger signals .In this paper ,based on the immune danger theory and DCA ,a dy-namic danger signals extraction method using the multiple classifier （MC） algorithm is proposed and a MC-DCA intrusion detection model is designed ,so as to improve the efficiency of antigen presentation and antibody recognition .The network intrusion detection data commonly used by KDD CUP 99 is used for the simulation experiments on the proposed MC-DCA intrusion detection model and the tradi-tional AIS model and DT model .The result shows that the recognition ability of MC-DCA intrusion detection system based on danger theory is better .