基于从网络流量的大小和流量曲线的形状两个方面的研究,建立了一种基于统计的网络流量模型,提出了计算正常情况网络流量曲线的算法。通过对比正常网络流量曲线和异常网络流量曲线之间的差距,实现了对异常数据流的自动检测。实验表明,该模型不仅可以模拟与网络实测数据相似的网络流量,而且具有一定的异常流量发现能力。
This paper builds a statistics-based network traffic model by analyzing the size of network traffic and the shape of the traffic curve. It also presents an algorithm to calculate the network traffic curve under normal condition. Automatic detection of anomaly network traffic is realized by calculating the difference between the normal network traffic curve and the abnormal one. Experiments show that the simulation results are in good agreement with the real network traffic and the model can, to some extent, detect anomaly traffic.