A Probabilistic Approach to Modeling Users´ Network Selection in the Presence of Heterogeneous Wireless Networks

In the presence of heterogeneous service provider (SP) networks such as wireless local area networks and cellular networks, users have to make a choice. The quality of service provided and the price charged by an SP play a crucial role in this decision process. In this paper, we study the network selection behavior of mobile users. A novel probabilistic framework is presented, with explicit analytical expressions for the transition and steady state network selection probabilities of the users. The proposed framework utilizes the price charged and the utility obtained at the SPs in obtaining the transition probabilities, even in the presence of multi-homing users. Furthermore, network-level parameters such as call blocking probabilities and throughput are obtained in terms of the steady state probabilities. Numerical results indicate that the prices and utilities at the heterogeneous SPs have considerable impact on the overall network performance.