Internet loss-delay modeling by use of input/output Hidden Markov Models

Performance of real-time applications on end-to-end packet channels are strongly related to losses and temporal delays. Several studies showed that these network features may be correlated and present a certain degree of memory such as bursty losses and delays. The memory and the statistical dependence between losses and temporal delays suggest that the channel may be well modeled by a Hidden Markov Model with appropriate hidden variables that capture the current state of the network. In this paper we propose an Input/Output Hidden Markov Model that, trained with a modified version of the Expectation-Maximization algorithm, shows excellent performance in modeling typical channel behaviors in a set of real packet links. The work extends to case of variable inter-departure time the previous proposed Hidden Markov Model that well characterizes losses and delays of packets from a periodic source.