摘要
提出了一种网络流量异常检测新算法。该算法将线性模型与小波变换相结合,解决了阈值监控无法告警和监测的问题。在实际的网络数据SNMPM IB以及Netflow的应用检测中,性能较好。与GLR算法相比,异常点的判断更加及时、准确、可靠。
This paper proposes a new algorithms which is based on combining the linear model and wave transaction. This algorithm can solve the problem that cannot be detected by threshold detection method, By applying the method to network traffic detection, etc, SNMP MIB data or Netflow data, the performance of this method has been shown. The algorithm has the advantage of accurate, reliable comparing with GLR method.
出处
《计算机应用研究》
CSCD
北大核心
2006年第11期217-219,共3页
Application Research of Computers
基金
国家"973"计划资助项目(2003CB314807)
关键词
网络管理
异常检测
流量监测
Network Management
Anomaly Detection
Traffic Monitor
