Observability of path loss parameters in WLAN-based Simultaneous Localization and Mapping

Indoor positioning by means of received signal strengths has been gathering much interest since the massive presence of wireless local area networks (WLANs) in buildings. Theoretical approaches rely on the perfect knowledge of the APs' positions and propagation conditions; since this is unrealistic in real world, we estimate such knowledge as well as the building map from data by applying Simultaneous Localization and Mapping (SLAM). In this paper we address the joint estimation of the path loss parameters, namely the transmitted power and the path loss exponent, this latter being usually approximated in the literature by the free space value. We provide examples that show the relevance of estimating both parameters and analyze observability issues from the point of view of estimation theory. The integration of the parameter estimation in a WLAN based SLAM algorithm - WiSLAM - has been carried out and the results are discussed.