AP Selection for Indoor Localization Based on Neighborhood Rough Sets

In this paper, a new AP (access points) selection method is proposed based on neighborhood rough sets (NRS) for indoor localization. Due to the existence of virtual APs in large scale buildings, received signal strength (RSS) may produce similar measurements, leading to biased estimates and redundant computations. Our method utilize neighborhood relations to transform the radio map into an extended rough set model such that offers a more discriminative way to select a small subset of available APs. Experimental results show that our method can be used to further improve the performance of the original AP selection methods.