基于免疫的多通道入侵防御模型  

Immunity-based multi-channel intrusion prevention model

作　　者：刘才铭[1] 赵辉[1] 张雁[1] 曾友州[2] 胡莹[2] 彭凌西[1] 曾金全[1] 

机构地区：[1]四川大学计算机学院,成都610065 [2]成都航空职业技术学院电子工程系,成都610021

出　　处：《计算机应用研究》2008年第6期1846-1849,共4页Application Research of Computers

基　　金：四川大学计算机学院青年教师基金资助项目(JS20070411506428)

摘　　要：提出了一种基于免疫的多通道入侵防御模型,按网络协议类型对数据包进行分类,利用多过滤器技术优化网络流的实时分析处理,模拟生物免疫系统发现和杀死病原体的原理,采用免疫细胞的特殊机理实现入侵防御功能,建立受保护网络或主机的指纹特征库,减少过滤器分析和处理的数据量,提高网络入侵防御的效率。实验结果表明,本系统提高了入侵防御的智能化程度,能发现未知攻击,并能对未知攻击作出一定的处理。This paper proposed a multi-channel intrusion prevention model based on biological immune system.Network flow was classified according to classification of protocol.The captured network packets were properly analyzed and disposed by multi-filters in real-time.The principles by which biological immune system detected and killed pathogens were simulated to realize the function of intrusion prevention.Then canstructed finger print characteristic library to protect interior networks and reduce the data to be dealt with by filters.Advanced efficiency of networks intrusion prevention.Experimental results show that the proposed system improves the intelligence of intrusion prevention.Unknown network attacks can be discovered and treated appropriately.

关 键 词：生物免疫系统 入侵检测 入侵防御 指纹特征 

分 类 号：TP309.2[自动化与计算机技术—计算机系统结构]