计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (13): 114-118.DOI: 10.3778/j.issn.1002-8331.2009.13.034

• 网络、通信、安全 • 上一篇    下一篇

移动Ad Hoc网络联合路由入侵检测模型研究

杨 清1,2,李方敏1   

  1. 1.武汉理工大学 信息工程学院,武汉 430072
    2.湖南科技大学 计算机科学与工程学院,湖南 湘潭 411201
  • 收稿日期:2008-03-07 修回日期:2008-05-26 出版日期:2009-05-01 发布日期:2009-05-01
  • 通讯作者: 杨 清

Study on model of intrusion detection for wireless Ad Hoc network based on SVM and FSM

YANG Qing1,2,LI Fang-min1   

  1. 1.School of Information Engineering,Wuhan University of Technology,Wuhan 430072,China
    2.School of Computer Science & Engineering,Hunan University of Science and Technology,Xiangtan,Hunan 411201,China
  • Received:2008-03-07 Revised:2008-05-26 Online:2009-05-01 Published:2009-05-01
  • Contact: YANG Qing

摘要: 无线移动Ad Hoc网络是一种新型的无线移动通信网络,由于其动态拓扑、无线信道以及资源有限等特点,容易遭受各种攻击。特别地,由于网络中的每个节点都参与路由,Ad Hoc网络路由协议自身的安全性尤为重要。以入侵检测技术为基础,分析针对AODV路由协议的各种攻击,提出了一种新的、有效的将有限状态机(FSM)协议分析和支持向量机(SVM)统计学习方法相结合的无线移动Ad Hoc网络路由入侵检测模型,该模型通过检测针对路由协议的各种攻击来实现安全路由。通过NS-2网络仿真实验表明:基于FSM和SVM相结合的混合入侵检测机制具有较高的检测精度和检测性能。

关键词: 支持向量机, 有限状态机, 无线Ad Hoc网络, 入侵检测

Abstract: Wireless mobile Ad hoc network is a novel wireless mobile communication network.Because of dynamic structure,wireless channel and limited resource,the nodes of mobile Ad hoc network are susceptible compromise.It is particularly vulnerable to all kinds of network attacks.In particular,every node joins in network route,so it is especial important for the security of route protocol itself.After analyzing all kinds of attacks to AODV protocol,a novel and effective intrusion detection model are presented for route protocol in wireless mobile Ad Hoc network,which combining Finite State Machine(FSM) with statistical learning method in Support Vector Machine(SVM) based on intrusion detection technology.This model can realize secure route through attacks detection of route protocol itself in wireless mobile Ad Hoc network.Simulation results in NS-2 shows this approach based on FSM and SVM method can achieve desired precision and performance and meet the security requirement of mobile Ad Hoc network.

Key words: Support Vector Machine(SVM), Finite State Machine(FSM), wireless mobile Ad Hoc network, intrusion detection