The article represents a practical approach to the problem of modeling stochastic features of Internet traffic. We summarize the main methods to determine the Hurst parameter and to include it in traffic models. In particular we show that we are able to construct efficiently Markov models of traffic with LRD, also with the use of Hidden Markov Chains. These Markov models may be a part of a computer network models aiming to evaluate its performance. Of course, the complexity of traffic models enlarges the size of the entire state space to be considered and hence the number of equations to be solved numerically. Therefore we are developing a software tool able to cope with models having hundreds of millions states.
On Stochastic Models of Internet Traffic
PAGANO, MICHELE
2015-01-01
Abstract
The article represents a practical approach to the problem of modeling stochastic features of Internet traffic. We summarize the main methods to determine the Hurst parameter and to include it in traffic models. In particular we show that we are able to construct efficiently Markov models of traffic with LRD, also with the use of Hidden Markov Chains. These Markov models may be a part of a computer network models aiming to evaluate its performance. Of course, the complexity of traffic models enlarges the size of the entire state space to be considered and hence the number of equations to be solved numerically. Therefore we are developing a software tool able to cope with models having hundreds of millions states.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.