Network Behavior based Mobile Virus Detection

摘要：

　　Fast spreading mobile viruses have caused major damage on both mobile users and carriers networks.Polymorphism and metamorphism make the detection even more difficult.Traditional signatures-based anti-virus systems yields excellent detection rates for existing and previously encountered viruses,but they lack the capacity to efficiently detect new unknown variants.In this paper,we study network behavior based mobile virus detection to figure out unknown viruses.We presented a network behavior feature selection method to find out generic features for different types of viruses.Real traffic from the GPRS (General Packet Radio Service) network is analyzed and plenty of tests are conducted to prove the effectiveness.The experiments show that many viruses can be classified into common virus families based on the similarities in network behaviors,and our feature selection method can detect unknown viruses with high possibility.

关键词： mobile virus detection network behavior feature selection data mining

作者: Ai-Fen Sui Dai-Fei Guo Tao Guo Ming-zhu Li

作者单位: Corporate Technology, Siemens Ltd. China Wangjing Zhonghuan Nanlu, Chao yang District,P.O.Box 8543, Beijing 100102, P.R.China

会议类型: 国际会议

会议名称: 2012 IEEE 14th International Conference on Communication Technology(2012年第十四届通信技术国际会议(ICCT 2012))

会议地点: 成都

会议语种:英文

页码: 1605-1609

在线出版日期: 2012-11-09（万方平台首次上网日期，不代表论文的发表时间）