A predictive bandwidth reservation strategy based on mobile station´s weekly histories for macrocellular wireless networks

One of the major challenges in the cellular networks is to guarantee the quality of service of the handoff calls by prioritizing them over the new calls. An operative approach to prioritize the handoff calls over the new calls is by reserving bandwidth for the handoff calls in the potential next cell that a mobile station may visit. Improving the prediction of potential next cell that a mobile station may visit will cause better bandwidth utilization. In this paper, we propose a predictive bandwidth reservation strategy that improves prediction by means of storing weekly movement probabilities of mobile station based on Markov modeling techniques. In order to decreasing the storage space that is needed for storing the mobile station´s movement probabilities, we adopted a dynamic hashing approach. Simulation results show that the weekly prediction in this strategy significantly improves bandwidth utilization, and the adopted dynamic hashing approach caused the overhead of storage space to be acceptable.