一种基于代理的分布式入侵检测系统结构设计  

Distributed Agent-based System Design for Intrusion Detection

作　　者：马力[1] 焦李成[1] 卢涛[2] 刘国营[1,2] 

机构地区：[1]西安电子科技大学电子工程学院,西安710071 [2]西安邮电学院网络中心,西安710061

出　　处：《通讯和计算机（中英文版）》2005年第6期55-58,共4页Journal of Communication and Computer

基　　金：本文得到陕西省自然科学基金（项目编号：2002F26）资助.

摘　　要：本文在分析现有入侵检测方法，技术与系统的基础土，基于入侵检测的自治代理模型，给出了一种分布式多代理的入侵检测系统模型。其核心思想是将原来的集中处理分布到网络的观测点，对每类攻击构造相应的代理程序。其优点是处理速度快，布置方便，排除了单点故障。另外，采用CORBA技术，实现了异构环境下的跨平台操作，为入侵检测系统提供了一条有效的技术途径。The available Intrusion Detection System （IDS） can＇t meet the need of the intrusion detection tast in high-speed network. Based on the model of Autonomous Agents for Intrusion Detection, this paper presents a kind of Distributed Multi-Agents Model for intrusion detection in computer network environments. The core idea of the model is that centried detection process at one point in traditional IDS is distributed into many low level node machines to provide. The model is characterized by its high speed, easy expansibility and no single-breakdown Finally, a brie earithative discussion of implementation of the model based on CORBA technique is given. All above provide an efficient technical way for constructing and the system is much better than IDS.

关 键 词：入侵监测系统 分布式处理 代理 CORBA 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]