基于不同时间粒度的端到端网络流量重构算法

End-to-end network traffic reconstruction algorithm based on different time granularity

针对间接测量网络流量的问题,提出一种基于不同时间粒度的新的端到端网络流量重构算法。根据网络流量分形和自相似特征,对粗时间粒度下的链路流量测量进行分形插值,得到细时间粒度下的链路流量;利用压缩感知理论,构造恰当的稀疏变换矩阵和测量矩阵,重构细时间粒度下的端到端网络流量。仿真结果表明,该算法有效可行。

This paper proposed a new end-to-end network traffic reconstruction algorithm basing on different time granularity for the indirect measurement of network traffic. According to the characteristics of fractal and self-similarity of the network traffic, firstly this paper applied fractal interpolation on link traffic of coarse time measurement granularity in order to get link traffic of fine time granularity. Then through the theory of compressive sensing, this paper constructed the appropriate sparse transformation matrix and measurement matrix to reconstruct end-to-end network traffic of fine time granularity. And the simulation results show that the method proposed in this paper is effective and feasible.