Localization Using Radial Basis Function Networks and Signal Strength Fingerprints in WLAN

Abstract

Fingerprinting localization techniques provide reliable location estimates and enable the development of location aware applications especially for indoor environments, where satellite based positioning is infeasible. In our approach we utilize received signal strength (RSS) fingerprints collected in known locations and employ a radial basis function (RBF) neural network to approximate the function that maps fingerprints to location coordinates. We present a clustering scheme to reduce the size and computational complexity of the RBF architecture and demonstrate the applicability of this approach in a real-world WLAN setup. Experimental results indicate that the RBF based method is an efficient approach to the location determination problem that outperforms existing techniques in terms of the positioning error.