Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (30): 92-94.DOI: 10.3778/j.issn.1002-8331.2010.30.027

• 网络、通信、安全 • Previous Articles     Next Articles

Network anomaly detection based on fuzzy clustering algorithm and QPSO algorithm in mobile Ad Hoc

ZHANG Duan1,LIU Yuan1,2,HAO Jian-dong1   

  1. 1.School of Information Engineering Jiangnan University,Wuxi,Jiangsu 214122,China
    2.Digital Media Center,Jiangnan University,Wuxi,Jiangsu 214122,China
  • Received:2009-03-12 Revised:2009-05-07 Online:2010-10-21 Published:2010-10-21
  • Contact: ZHANG Duan1,LIU Yuan1,2,HAO Jian-dong1

模糊聚类和QPSO算法在Ad Hoc异常检测中的应用

张 端1,刘 渊1,2,郝建东1   

  1. 1.江南大学 信息工程学院,江苏 无锡 214122
    2.江南大学 数字媒体中心,江苏 无锡 214122
  • 通讯作者: 张 端

Abstract: Based on fuzzy clustering algorithm and QPSO clustering algorithm,a model of detecting network anomaly based on these two algorithms is presented.In cluster analysis,K-Means algorithm is one of the most widely used methods.The paper uses K-Means clustering algorithm to seed the initial swam.All the process of clustering based on the Euclidean distance among data-vectors.Cluster-center is chosen by QPSO clustering algorithm.Finally,the experiment result shows that this model is effective for wireless network anomaly detection.

Key words: K-Means clustering algorithm, Quantum behaved Paticle Swarm Optimization(QPSO) clustering algorithm, Ad Hoc wireless network, anomaly detection

摘要: 根据模糊聚类算法和量子粒子群算法,提出一种基于以上两种算法的网络异常检测模型,并将该模型应用到Ad Hoc无线网络异常检测中。在聚类分析中,K-Means聚类算法是应用最广泛的方法之一。该模型先利用K-Means聚类算法的结果重新初始化粒子群,聚类过程都是根据数据向量间的欧几里德距离;再通过量子粒子群优化算法寻找聚类中心;最后进行仿真模拟,实验结果表明该模型对Ad Hoc无线网络异常检测是有效的。

关键词: K-Means聚类算法, 量子粒子群算法, Ad Hoc无线网络, 异常检测

CLC Number: