|
Abstract:
|
As a technique to investigate link-level properties of a computer network with low operational cost, network tomography has received considerable attentions in recent years. A number of methods have been proposed to estimate link-level loss rate or delay distribution for networks with a tree or general structure. However, these methods suffer from either high computational cost or insufficient use of information in the data. In this talk, I will report our recent theoretical results and practical algorithms for parameter estimation in loss and delay tomography. By introducing a group of novel statistics and alternative parameter systems, we find that the likelihood function of the observed data from loss tomography keeps exactly the same mathematical formulation for tree and general topologies, and can both be converted into the standard exponential family. The spirit of this approach can also be applied to more challenging delay tomography problem to greatly simplify the computation. Simulation studies show that the algorithms based on our theoretical finding can speedup traditional methods for more than 1000 times for a large network.
|
