A hierarchical position-prediction algorithm for efficient management of resources in cellular networks

We propose a novel hierarchical position-prediction algorithm which improves the connection reliability and overall system performance by accurately predicting the future movement pattern of the mobile user. Our algorithm adopts a two-level approach-at the top (global) level, approximate pattern matching is applied to determine the mobile's overall inter-cell movement direction, and at the bottom (local) level, an optimum self-learning Kalman estimator is applied that uses real-time signal strength measurements to estimate the mobile's intra-cell movement direction and velocity. Simulation results show that this two-tier prediction algorithm promises to provide a high degree of prediction accuracy as it is robust in the presence of random movement patterns and noise corrupted measurement data.