Cell stay time prediction for mobility independent predictive services in wireless networks

In the past few years, there has been much research in wireless adaptive networking and resource management, in order to examine how to efficiently handle wireless communications between mobile hosts and base stations. In this paper, we focus the attention on the mobile independent predictive (MIP) service class, which provides a passive reservation policy, with the goal of QoS guarantees during hand-off events. We analyzed MIP users´ mobility along coverage areas, to obtain a system utilization improvement, by reducing passive resource reservations on cells that users will probably visit. A prediction technique is also proposed and calls are managed in a wireless system by a utility-based algorithm that takes into account the radio-link quality variability by a Markov channel model, based on the IEEE802.11b standard. System performances are evaluated by varying users´ speed and system utilization with prediction error percentages are shown. It is observed that the proposed prediction technique causes an increase in system utilization.