An indoor localization system based on 3D magnetic fingerprints for smart buildings

Abstract

Human behavior modeling and activity interpretation have been of increasing interest in the Information Society for long time. Internet of Things technologies enhance the situational awareness or “smartness” of service providers and consumers alike. On the other hand, user-centric sensing systems are ideal candidates for ubiquitous observation purposes thanks to the indispensable role of mobile phones in everyday life. This paper presents a novel approach for mobile phone centric observation applied to indoor localization for smart buildings. The goal is to provide accurate localization data which can be used for offering customized IoT-based services in buildings. Unlike existing work which uses the intensity of magnetic field for fingerprinting, our approach uses all three components of the measured magnetic field vectors to achieve accurate results of localization. The resulting localization system does not rely on any infrastructure devices and is therefore easy to manage and deploy. Our approach covers, with test, comparison and justified selection, every tools and methods necessary to implement a genuine experiment in a real building which gives order of a few meters precision.