A Mobility-based Prediction Algorithm for Vertical Handover in Hybrid Wireless Networks

This paper proposes a prediction algorithm to choose candidate networks for vertical handover by using the speed of the mobile station and mobility pattern in hybrid wireless networks. If the predicted cell residence time is not long enough compared to the handover delay, then this particular network will be removed from the candidate network list. Since it is very difficult to predict the user mobility exactly, the proposed algorithm measures the mobility pattern of a user and estimates the network residence time by using probability density functions. The main contribution of this algorithm attempts to minimize unnecessary handovers, leading to an effective algorithm for vertical handover. In this paper we conduct simulation studies by OPNet. The simulation results show that the proposed algorithm minimizes the number of handovers in the hybrid wireless network environment.