Author Login Editor-in-Chief Peer Review Editor Work Office Work

Computer Engineering ›› 2012, Vol. 38 ›› Issue (08): 123-124. doi: 10.3969/j.issn.1000-3428.2012.08.040

• Networks and Communications • Previous Articles     Next Articles

Social Network Analysis Method for Detecting Spam over Internet Telephony

CHEN Jia 1, LI Min 1, XU Lei 2   

  1. (1. School of Mathematics and Computer Science, Wuhan Textile University, Wuhan 430073, China; 2. School of Computer, Wuhan University, Wuhan 430070, China)
  • Received:2011-08-08 Online:2012-04-20 Published:2012-04-20

一种检测SPIT的社会网络分析方法

陈 佳1,李 敏1,徐 蕾2   

  1. (1. 武汉纺织大学数学与计算机学院,武汉 430073;2. 武汉大学计算机学院,武汉 430070)
  • 作者简介:陈 佳(1982-),女,讲师、博士,主研方向:数据挖掘,演化计算;李 敏,讲师、博士;徐 蕾,硕士
  • 基金资助:

    湖北省教育厅中青年科技基金资助项目(20111613)

Abstract: Existing VoIP security technology can not effectively detect Spam over Internet Telephony(SPIT) attack. To resolve the problem, this paper uses social network method to analyze the history of communication behavior data of the user node. Average call length, call frequency, call does not exist the user number, hang up the number/passive hang up the number and one-way outgoing/two-way communication nodes are extracted as characteristic properties to establish a Bayesian model. Experimental results verify that the method is effective.

Key words: Voice over Internet Protocol(VoIP) service, Spam over Internet Telephony(SPIT) attack, social network analysis, user communication behavior, Bayesian model

摘要: 已有的VoIP安全技术无法有效检测SPIT攻击。针对该问题,利用社会网络分析方法,通过分析用户节点的历史通信行为数据,提取平均通话时长、主动呼叫频率、呼叫不存在用户次数、主动挂断次数/被动挂断次数以及单向呼出的节点数/双向通信节点数作为特征属性,建立贝叶斯模型,实现SPIT节点的识别与检测。实验结果证明了该方法的有效性。

关键词: VoIP业务, SPIT攻击, 社会网络分析, 用户通信行为, 贝叶斯模型

CLC Number: