Continuous-time Hidden Markov models in Network Simulation

Abstract

The use of continuous-time hidden Markov models for network protocol and application performance evaluation has been validated to simulate network environments. In this paper, we develop a better algorithm to infer the continuous-time hidden Markov model from a series of end-to-end delay and loss observation of probing packets. We prove the algorithm's feasibility by theory deduction and realize numerable validation by comparing the probability of the observed sequence produced by the model inferred by different methods. The algorithm complexity is lower.