Access Point Selection Using Particle Swarm Optimization in Indoor Positioning Systems

In recent years, the usage of existing WLAN infrastructure of large buildings have been suggested for indoor localization and tracking systems. In indoor positioning systems (IPS), WLAN fingerprinting-based methods require recording received signal strength (RSS) of surrounding access points (APs). These are usually more than the necessary number of APs needed by the IPS system. Therefore, eliminating redundant or non-informative APs does not only reduce computational cost, which is necessary for the real-time system, but also improves the accuracy of the indoor positioning system. In this paper, we present a particle swarm optimization (PSO)-based technique to select the most informative APs at each clustered area, where K-means clustering method is utilized to confine location of the user into a small area. At each cluster, PSO is applied to select the best joint combination of APs decided by the minimum mean of distance error. Simulation results show that the proposed system outperforms the behavior of the other commonly proposed selection methods such as random, strongest APs, and Fisher criterion.