Dynamic Optimization of Generalized Least Squares Handover Algorithms

Abstract

Efficient handover algorithms are essential for highly performing mobile wireless communications.These algorithms depend on numerous parameters, whose settings must be appropriately optimized to offer a seamless connectivity. Nevertheless, such an optimization is difficult in a time varying context, unless adaptive strategies are used. In this paper, a new approach for the handover optimization is proposed. First, a new modeling of the handover process by a hybrid system that takes as input the handover parameters is established. Then, this hybrid system is used to pose some dynamical optimization approaches where the probability of outage and the probability of handover are considered.Since it is shown that these probabilities are difficult to compute, simple approximations of adequate accuracy are developed. Based on these approximations, a new approach to the solution of the handover optimizations is proposed by the use of a trellis diagram. A distributed optimization algorithm is then developed to maximize handover performance. From an extensive set of results obtained by numerical computations and simulations, it is shown that the proposed algorithm allows to improve performance of the handover considerably when compared to more traditional approaches.C. Fischione and G. Athanasiou are with the Automatic Control Lab,

A Multi-Layer Parametric Approach to Maximize the Access Probability of Mobile Networks

“…In [13], a mechanism based on trellis (so, somewhat similar to the above two approaches) has been used by considering closed-form performance evaluation. These are even though, based on statistical modeling and not tied to real measurements.…”

“…These are even though, based on statistical modeling and not tied to real measurements. More, [13] takes into account the signal strength at the receiver only on the basis of the user speed and distance. Last, the work in [14] incorporates the use of parameterized utility functions in order to model several Quality of Service (QoS) elasticities of different applications.…”