Ray-Tracing Based Fingerprinting for Indoor Localization

Abstract

Empirical fingerprinting is currently one of the most efficient localization methods in indoor environments, due to the now ubiquitous deployment of wireless local area networks. A fingerprint is the pattern of received signal strengths from all the access points visible at a particular position. Reference fingerprints are obtained from extensive measurement campaigns and used during an offline phase to construct suitable radio maps of the environment of interest. However, this approach is very site-specific and the radio maps require therefore to be regularly updated in order to take into account changes in the environment, i.e. a labor-intensive and time-consuming task. Hence, ray-tracing simulations are instead used in this paper to construct these radio maps based on deterministic prediction of the radio wave propagation. To validate this approach, field experiments were conducted in an indoor office environment. The results show (i): the sensitivity of fingerprints to small-scale fading and human shadowing, as well as (ii): good agreement between the measured and ray-tracing simulated fingerprints, especially with strong line-of-sight.