A Map-Assisted WiFi AP Placement Algorithm Enabling Mobile Device’s Indoor Positioning

Abstract

Location information and positioning technology are important to some Internet of Things (IoT) applications. The accuracy of indoor positioning using WiFi can be substantially enhanced by appropriate access point (AP) placement strategies, i.e., in a given indoor environment to deploy the WiFi APs at the locations where the mobile devices can work out their location more precisely. The plan of AP placement needs to be generated automatically by algorithms, especially for large-scale indoor environment. This paper presents an indoor map system that provides coordinate system and graphic representation. The detailed map information such as walls can be explicitly expressed and used to assist the AP placement algorithm. In this paper, AP placement is formulated into an optimization problem in which the sum of Euclidean distance of fingerprints among all the reference points (RPs) is maximized. The fingerprint at RP is predicted by an indoor radio propagation model which takes the attenuation of walls into consideration with the assistance of our indoor map. The optimization problem is solved by particle swarm optimization (PSO) and evaluated by k-nearest neighbors positioning algorithm in a real-world environment. The experimental results show that our map-assisted AP placement can provide higher positioning accuracy.