基于加权映射模型的主干网流量特性分析

网络测量表明,IP主干网上流量在次秒级的时间尺度下呈现出非平稳的泊松特性,而不是近10年大量研究的自相似现象。从分析因特网的结构出发,提出了一种加权映射模型来解释这种特性。该模型通过引入混沌变化因子和流量特征间的关系,对接入网中的具有自相似特性的流量进行了变换得到具有趋泊松性的流量。最后,通过在NS2上的仿真实验证实了该模型的合理性。

Analysis of Network Traffic Characteristic in IP Network Backbone Based on Weighted Map Model

Network measurement has proven that traffic present the non-stationary Poisson characteristic at sub-second time scales in IP backbone,which does not accord with self-similarity phenomena for last decade.By analyzing Internet structure,a novel weighted map model was put forward,which gave a reasonable explanation for the new traffic characteristic.The model related chaos impact factors with traffic characteristic,and transferred the traffic with self-similarity in access network to that with Poisson characteristic.Finally,the validity of the model was verified by simulation experiments on NS2.