Performance assessment of indoor location technologies

Location based services (LBS) are currently enjoying a strong success as a result of well-proven GNSS positioning technology (GPS, assisted GPS). However, the future generation of LBS will have to address the challenge of accurate and reliable indoor localization. In fact, this need has already been clearly expressed by various communities of professional end users (firemen, security forces, etc.) in the context of LIAISON and WearlTork projects funded by the European community's sixth framework program. In order to assess the performance of the location technologies most suited to cope with the stringent constraints of indoor LBSs oriented towards the needs of professional users, in particular those of the firemen, a "location trial" composed of several test scenarios was organized, putting face to face in a common systematic reference the following positioning technologies: - Inertial MEMS coupled with GNSS, using two different algorithmic approaches for the inertial component: signal pattern recognition associated with human biomechanical walking models and conventional inertial navigation using zero velocity updates at footfalls; - UWB radio-based localization prototype system. The results from this "location trial" show that inertial technology achieves interesting performances (stand-alone positioning accuracy better than 3 meters RMS after 4 minutes and less than 6 meters RMS after 8 minutes of continuous pedestrian walk), but still lacks robustness against specific environmental conditions (in particular magnetic disturbances affecting orientation estimation) and users' walking behavior. As for the UWB prototype system, it provides good positioning accuracy (less than 3 meters RMS) for nominal operational conditions, but but it can experience severe degradation under certain circumstances. By highlighting the pros and cons of each technology under a common framework, this "location trial" has provided a clearer understanding how their seamless combination can realistically address all users' needs: accuracy, reliability, robustness, coverage, deployability and wearability.