摘要： 介绍了几种常见的分布式入侵检测系统针对目前入侵检测系统的不足提出了一个基于移动Agent的网络入侵免疫系统——MANIIS。系统借鉴了自然免疫系统的一些特性,如克隆选择、逆向选择和免疫记忆等,具有自主学习和动态适应的能力,智能性较高;采用层次结构的探测和响应机制,各agent相互独立,同时能在网络中自由移动,相互协作,以检测分布式攻击。文中说明了系统的结构框架以及实现细节,并分析了系统的性能。

关键词: 移动Agent, 入侵检测系统, 免疫系统, 网络安全

Abstract: Several common distributed intrusion detection systems are introduced. Considering the shortcoming of the current intrusion detection system, the mobile agent based network intrusion immune system——MANIIS is proposed. The system uses some properties of the nature immune system, such as clone selection, negative selection and immune memory, having the ability of self-learning and dynamic adaptation and the high intelligence. Using the layered construction of sensor and response scheme, the agents are independent but can roam in network freely and collaborate with each other to detect the distributed intrusion. The system's architecture and the realization details are discussed and the system performance is analyzed.

中图分类号: 

• TP393.08