LEVERAGING HUMAN MOBILITY AND PERVASIVE SMARTPHONE MEASUREMENTS-BASED CROWDSOURCING FOR DEVELOPING SELF-DEPLOYABLE AND UBIQUITOUS INDOOR POSITIONING SYSTEMS

Abstract. Nowadays, it's hard to imagine life without smartphones. These devices surround us and have become an inseparable part of our daily lives, since they provide end-users with unlimited access to information, services, and communication through their multitude of sensors and computational capabilities. Through sensor-rich smartphones, it is possible to collect pervasive crowdsourced signatures. Manual training approaches to localize these signatures and construct offline radio and magnetic maps required for fingerprinting-based positioning techniques are time-consuming, labour-exhausted, and restrict the scalability of ubiquitous signals, such as Wi-Fi and magnetic field. As an alternative, autonomous localization of pervasive signatures can eliminate human supervision and scalability barriers, prompting to development of self-deployable systems. The development of these systems offers unparalleled potential for extending the ubiquity of indoor positioning systems (IPS) and makes realizing ubiquitous IPSs one step closer. This study proposes a scheme to leverage pervasive crowdsourced data to develop a ubiquitous IPS without the need for external resources.