采用多播依赖树模型推测链路时延分布

网络断层成像(network tomography)技术研究基于端—端测量推测链路级的性能参数,对于评估网络性能十分重要。文章提出多播依赖树模型,并采用上向—下向算法来推测网络链路的时延分布。该方法通过引入树的马氏性减少了估计时的自由参数,仿真实验证明了这种方法的正确性,能精确地推算出网络链路的时延性能。

Using Dependence Tree Model for Network Delay Tomography

It is essential to have link-level performance data for understanding the features of a network.Network tomography is based on end-to-end measurement to reason the network internal information.In the paper we propose a method using the multicast dependence tree model and upward-downward algorithm to infer link delay distribution,which requires a relatively small number of free parameters by making Markov-like assumptions on the tree.The simulation result shows the method works efficiently and produces reasonable accurate results.