一种分布式拒绝服务攻击的检测方法

A Novel Method to Detect DDoS Attack on Internet

该文提出和实现了一种基于网络异常流量特征的检测模型。通过将多条链路或多个流的流量信号作为一个整体进行研究．构建了网络异常流量监控的系统模型。该模型包括了数据采集、分析、异常判断和警告等功能，并综合了异常信息融合及警告信息关联性分析技术。通过实验证明，该套系统模型对短时间段内的突爱流量能提供有效的检测和报警服务。

A novel model is proposed for detecting DDoS attack, one of the most popular attacks on Internet. The model can recognize abnormal traffic based on integrating multi-routes or multi-traffics into a whole and includes three sub-models： data collection, data analy- sis, and abnormality detection. From the experiments, it can detect outburst traffic and afford alarm service effectively.