Handover prediction for cellular systems in a multi-services context

Abstract

Handover prediction can become an excellent way to optimize the resource allocation scheme in a cellular system. Indeed, the base station can modify in real time the distribution of its allocation in order to free resources for incoming users. This allocation is applied in different ways depending on the user's service (voice, data, etc.). For example, a Web service does not occupy a channel in the same way of a voice user. We provide a user profile based upon a cell or user approach. A resource allocation scheme is then introduced in order to compare these two approaches. We also provide a simulation model that integrates user's movements in a cellular system, multiple services, and impacts on system occupancy.