基于遗传算法的免疫网络设计  

Design of immune network based on evolutionary algorithm

作　　者：李军[1] 常征[2] 

机构地区：[1]电子科技大学英特尔互联网交换架构研究实验室,四川成都610054 [2]成都信息工程学院,四川成都610225

出　　处：《成都信息工程学院学报》2004年第3期392-394,共3页Journal of Chengdu University of Information Technology

摘　　要：从人体免疫网络抵御抗原与计算机网络抵御入侵的相似性,提出一个基于遗传算法的免疫网络并应用于分布式入侵检测系统。将一个基于每秒网络数据量为参数的时间序列作为检测子的抗体表达形式,在单个网段检测网络数据,通过各个网段之间的抗体交流,实现抗体的优化。在抗体的进化过程中,提出了较好地评价抗体适应度的评估函数,为优化抗体的遗传算法的实现打下了基础。A distributed intrusion detection system based on the immune network theory is presented by the inspiration of the similarity between the defense against agtigen taken by the human immune network and the defense against intrusion taken by the computer network.The detector is represented by a time series taking parameter of network data traffic per second. A portion of network detected by only one sniffer for all traffic passing through it is detected while all sniffers exchange detections for the optimization. Some fitness functions to improve the detector quality is also presented which is a part of the evolutionary algorithm.

关 键 词：免疫系统 免疫网络 遗传算法 分布式入侵检测 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]