Using phase shift fingerprints and inertial measurements in support of precise localization in urban areas

Abstract

Localization is an important primitive that is utilized in a number of important applications such as location-based mobile services, augmented reality, and autonomous mobile robotics. While the GPS technology is considered the de facto standard for outdoor localization, it is known to suffer from significant accuracy limitation in urban areas. In this work, we present a particle filter-based data fusion technique for localization in urban areas. The proposed localization technique provides more accurate location estimation results due to its ability to efficiently fuse together information collected from diverse sensor technologies. The novelty of our proposed approach stems from its ability to fuse data from diverse sources, namely, phase shift fingerprints collected from Low Power AM Radio (LPAM) towers and inertial measurement sensors. Our simulation results indicate that the proposed approach can achieve an accuracy of 0.5 m using a limited number of LPAM towers as low as 5. Also, the proposed approach requires the collection of a low number of LPAM phase shift fingerprints. Our simulations indicate that 30 fingerprints are enough to provide 0.5 m accuracy in a 100 × 100 m 2 deployment. Keywords Smart cities • Localization in urban areas • Data fusion • Fingerprinting • Channel state information 1 Introduction In recent years, several critical national infrastructures (such as the electric grid, intelligent transportation, water systems) Mohammed Elbes

Special issue on ambient systems, networks and technologies

“…Another paper is by Elbes et al [15] entitled BUsing phase shift fingerprints and inertial measurements in support of precise localization in urban areas^. In this paper the authors present a particle filter-based data fusion technique for localization in urban areas.…”