Handoff algorithm based on radial-basis function networks in high altitude platform station cellular systems

In this paper, we propose a handoff algorithm, based on a neural network, in a joint system of terrestrial and high altitude platform station (HAPS) cellular systems. A radial-basis function (RBF) network is used for making a handoff decision to the neighbor base station. The set of training patterns consisted of averaged signal strength received from the serving and nearby base stations (BSs), directions of users estimated by the MUSIC algorithm on an antenna array, and traffic intensities. This combined mobile-cell related information improves the handoff algorithm, yielding both low number of unnecessary handoffs and decision delay. As a revolutionary wireless system, the HAPS base station can supply services for uncovered areas, improving total capacity of areas service-limited by a terrestrial BS. Performance comparisons of the presented method and the conventional hysteresis rule are given in terms of handoff rate, blocking rate and dropping rate. Simulation results demonstrate that employing the presented algorithm can reduce unnecessary handoffs, call blocking rate and call dropping rate as well.