Indoor positioning using WLAN coverage area estimates

This paper introduces a novel method for positioning using coverage area estimates of wireless communication nodes. The coverage areas are estimated in a Bayesian inference framework using location fingerprints that are collected in an offline calibration phase, and the estimated coverage areas are stored in a database. In the online positioning phase the coverage areas of the heard communication nodes are used to infer the position of the mobile terminal. Floor plan information is used to enhance the positioning accuracy. In a field study comparing Kalman Filter, Box Filter and Particle Filter using real WLAN measurement data, it is found that Kalman Filter achieves almost the same accuracy as Box Filter and Particle Filter but with smaller computational load.