Multi-objective optimization of WLAN associations with improved handover costs

Abstract

The deployment of a large number of access points in public areas presents a new paradigm in which a given station may be served by more than one access point at a time. Recently, a network model has been proposed to derive the optimal association strategy for the stations by solving a linear optimization problem. According to that model, a station cannot download data while performing a handover. In this letter, we demonstrate that, even in static conditions (i.e., no mobility and constant load in the cell), an association strategy can force a handover. We thus propose a multi-objective optimization function that maximizes the download rates and minimizes the number of handovers. As a handover implies an extra cost due to necessary signaling to be exchanged, we propose a new handover cost function that takes into account the extra bandwidth. Using the new strategy, the system throughput slightly decreases, and the number of handovers is reduced.