基于机器学习的移动自组织网络入侵检测方法

Intrusion detection method for mobile ad-hoc networks based on machine learning

移动自组织网络是由无线移动节点组成的复杂分布式通信系统。研究了移动自组织网络的入侵检测问题,采用了一种新型的基于机器学习算法的异常入侵检测方法。该方法获取正常事件的内部特征的相互关系模式,并将该模式作为轮廓检测异常事件。在Ad-hoc按需距离向量协议上实现了该方法,并在网络仿真软件QualNet中对其进行了评估。

Mobile ad-hoc networks （MANETs） represent complex distributed communication systems comprised of wireless mobile nodes. Based on the discussion of intrusion detection problem in MANET, a novel anomaly intrusion detection method based on machine learning algorithm was proposed to detect attacks on MANET. The method captured the normal traffic＇s inter-feature correlation pattern which could be used as normal profiles to detect anomalies caused by attacks. The method was implemented on Ad-hoc On-Demand Distance Vector （AODV） protocol and evaluated in QualNet, a leading network simulation software.