摘要
将小波理论和统计学习运用到网络入侵检测中,使用小波核支持向量机(WSVM)对网络连接信息进行攻击检测和异常发现。仿真试验结果表明,与RBF核相比,小波核支持向量机在泛化能力和检测能力方面都有所提高。
Wavelet theory and statistical learning are combined to apply in network intrusion detection field.Through the analysis of current intrusion detection /nethods and characteristic of wavelet support vector machine （WSVM）,this paper tries to apply WSVM as classifying means to deal with network connecting data.The results indicate that the classifying performance of wavelet kernel support vector machine shows better than RBF kernel SVM at extending ability and precision.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第17期143-145,共3页
Computer Engineering and Applications
基金
国家自然科学基金资助项目(编号:10471036)
湖南省教育厅青年科研基金资助项目(编号:05B055)
关键词
小波核函数
支持向量机
入侵检测
wavelet kernel function,wavelet supporting vector machine ,intrusion detection
