(Meta-)data modelling

Abstract

GPS has become the reference for outdoor positioning, implementing a direct connection between the GPS satellite and a receiver device. Indoor positioning raises new challenges, locating a target device requires the wireless sensors networks and other technologies. Sensor networks deployed in buildings are commonly used for many applications based on location: surveillance, detection, navigation, etc. These indoor locating sensors generate lot of data related to tracking information. Exploiting this information for investigation issues for example remains a relevant purpose. The context of this paper is related to indoor locations systems based on wireless cell, ICCARD sensors and video surveillance cameras. In this context, as no global reference system similar to GPS is available, the location information issued from various systems, platforms, devices, etc. have neither standards nor common formats, and remain heterogeneous. This heterogeneity is mainly due to the different types of positions (geometric, symbolic, etc.) expressed w.r.t. various reference systems. In order to manage them in a given framework, it is necessary to homogenize the relevant (Meta) data to process the global knowledge they can give. This paper presents a contribution to extend our framework [1] to information generated by location sensor networks deployed in an indoor environment. The use case is illustrated in a forensic application [2].