一种基于免疫学的园区网入侵检测模型  

An Immunology-Based Model for Intrusion Detection on the Intranet

作　　者：孙美凤[1] 龚俭[1] 

机构地区：[1]东南大学计算机系网络中心

出　　处：《计算机工程与科学》2004年第10期34-36,40,共4页Computer Engineering & Science

基　　金：国家自然科学基金资助项目(90104031)

摘　　要：现有基于网络的入侵检测系统在许多方面有缺陷,其中两个主要问题是高误报率和缺乏自适应性。本文首先介绍了Hofmeyr的自适应免疫系统模型,然后分析了园区网的流量特征,描述了一个可用于园区网的入侵检测系统模型。本文建议的模型保留了Hofmeyr模型的优点,可望为园区网提供全局性的检测和保护。Existing network-based intrusion detection systems have drawbacks in many aspects, including two obvious problems of high ratio of false alarms and the lack of self adaptation. In this paper, Hofmeyr's model is first introduced briefly, then the traffic features of the Intranet are analyzed. Based on these, a new model is proposed. This model can be expected to detect and protect the overall Intranet with all the advantages of Hofmeyr' s model preserved.

关 键 词：园区网 入侵检测模型 基于网络的入侵检测系统 入侵检测系统模型 自适应性 误报率 描述 免疫学 免疫系统 保留 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]