Modeling P2P-TV Traffic Using Hidden Markov Models

The increasing success of P2P-TV applications, that may overwhelm the network with their large volume of traffic in the near future, has caught the attention of network operators and service providers. These applications have the potential to completely transform traffic pattern and even push network usage to critical regions; some researchers claim that P2P- TV may overload backbone links. Indeed, their impact may be disruptive due to the large volume of generated traffic combined with huge number of users. Understanding and characterizing P2P-TV traffic is very important. However there is also the need for simple and flexible traffic models that can be effectively introduced in network simulators or complex dimensioning and planning tools. In this study, we proposed a simple traffic model that can be representative of P2P- TV applications. We present our preliminary set of results, showing that a simple Hidden Markov model can capture the P2P-TV traffic behavior.