Reduction of call drop-outs during handoffs using efficient signal strength prediction algorithms for personal communication systems

In cellular mobile networks, the handoff decision is usually taken based on the past signal and interference level measurements taken at regular time intervals. Quality degradation can be minimized if a mobile can always be handed over to the base station which would provide highest quality during the next measurement cycle. For this purpose, a good prediction algorithm is needed to predict the condition of the signals received from various base stations during the next measurement cycle. One of the major objectives of this paper is to discuss whether a considerable quality improvement can be obtained using prediction. A lower-bound for the drop-out performance is obtained using a hypothetical scheme which assumes the best possible prediction. The performance of several prediction algorithms are considered under various shadowing conditions. It is found that the performance of prediction schemes greatly depends on the speed of the mobile and the shadowing correlation environment.