In traditional TV delivery systems, all the channels are pushed toward the subscribers. Therefore, the user activities like channel switching have no effect on the provider network. Consequently, TV providers have felt no strong need to analyze and investigate the behavior of their subscribers. In IPTV networks, however, the situation is quite different. Channel switching activity of users imposes extra load on the network and therefore is worth to be investigated and analyzed. In this paper we elaborate the requirements of IPTV user modeling and propose a model taking into account the requirements. A formal description of the model is given and after parameterization, the model is implemented using the LoadSpec tool. Our proposed model can produce, in a realistic manner, the workload of switching events of a typical IPTV client during an active session. Understanding and modeling user activities is essential for many design and engineering tasks such as evaluation of various design options, optimal system parameter tuning, admission control mechanisms and network capacity planning. The application of the proposed model for the IPTV analysis, specifically, for the evaluation of the call blocking probability is shown by two case studies.
