入侵检测系统响应策略的进化博弈论分析  

An Evolutionary Game Analysis on The Response policy of The Intrusion Detection System

作　　者：夏子超[1,2] 银鹰[1,2] 陈晓桦[1,2] 

机构地区：[1]上海交通大学信息安全工程学院,上海200240 [2]上海市信息安全综合管理技术研究重点实验室,上海200240

出　　处：《微计算机信息》2009年第33期60-61,47,共3页Control & Automation

基　　金：基金申请人:陈秀真;银鹰;项目名称:计算机网络系统安全评估优化的研究;基金颁发部门:国家自然科学基金委(60605019)

摘　　要：起源于生物进化理论的进化博弈论更关注博弈方的策略选择的长期变化趋势和稳定状态分析,而在对入侵检测系统的研究中,响应策略变化的趋势和状态一直是关注的重点。本文将进化博弈论的研究方法运用到入侵检测系统中去,提出了一个基于进化博弈论的入侵检测系统学习模型,运用复制动态理论对模型进行分析,体现了入侵检测系统的响应过程中策略的调整以及变化趋势。本文模型的进一步研究将在入侵检测系统的学习机制方面带来更多的讨论。Originating from biological evolutionary theory, evolutionary game theory focuses on long-term trend and stable state analysis of strategy selection, and trend and status of response policy are always treated as key points in analysis of intrusion detection system （IDS）. Research methods for evolutionary game theory are applied to IDS in this paper, and a model of IDS, which is based on evolutionary game theory, is also introduced. By means of replicator dynamics theory, the paper analyses the new model, and research results embodies strategy adjustment and development trend during IDS＇s response process. Further research on the model introduced in paper will bring on more discussion on learning system of IDS.

关 键 词：入侵检测系统 进化博弈论 复制动态 进化稳定策略 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]