对网络通信异常行为识别方法的相异度分析

Analysis of the Abnormal Degree of Network Communication Distinguishing Behavior

为了对网络通信异常行为进行有效地识别,提出了一种基于相异度分析的识别方法,以该方法的数据基础为Netflow网管数据,并对具体的实现步骤进行了设计。通过实验分析,证明了采用该识别方法能够对网络中通信行为异常、突发流量异常以及异常网络攻击等表征现象进行实时监测,并且取得了良好的效果,使网络运营单位的服务质量和网络安全支撑服务能力取得了有效的提高。

In order to identify the abnormal behavior of network communication effectively, this paper proposes a recognition method based on dissimilarity analysis. The data base of the method is Netflow network management data, and the concrete im- plementation steps are designed. Then, through the experiment analysis and analysis, it is proved that this method can be used to monitor the phenomena such as abnormal communication behavior, sudden traffic anomaly and abnormal network attack in the network, and obtain good results, so that the network service unit＇s service quality and Network security support service ca- pacity has been effectively improved.