Performance Bounds for Mobile Cellular Networks with Handover Prediction

Abstract. We determine the gain that can be achieved by incorporat-ing movement prediction information in the session admission control process in mobile cellular networks. The gain is obtained by evaluat-ing the performance of optimal policies achieved with and without the predictive information, while taking into account possible prediction er-rors. We evaluate the impact of predicting only incoming handovers, only outgoing or both types together. The prediction agent is able to de-termine the handover instants both stochastically and deterministically. Two different approaches to compute the optimal admission policy were studied: dynamic programming and reinforcement learning. Numerical results show significant performance gains when the predictive informa-tion is used in the admission process, and that higher gains are obtained when deterministic handover instants can be determined.