A generalized least squares handover algorithm for wireless systems

An adaptive handover algorithm for wireless communications systems is proposed in this paper. The algorithm makes use, among others, of the least squares estimate of path loss parameters and is presented as a generalized version (G-ELS) of the ELS (extended least squares) algorithm. The relevant novelty of the proposed algorithm consists of detecting sudden variations of path loss parameters (a knee in line-of-sight and the street corner effect) in microcellular environments by re-initializing the estimate. A simple hysteresis adaptation rule is also included. It is seen that the G-ELS algorithm yields good performance in the microcellular environment, without determining performance losses with respect to the ELS in macrocellular environments. Furthermore, it may still yield better performance than the classical AVG (averaging) algorithm.