Geomagnetic field holds much promise for indoor localization due to its pervasive spatial presence, high signal stability and wide availability of magnetometers already embedded in mobile devices. Previous work in the area often fuses it with a pedometer (step counter) via particles. These approaches are computationally intensive and require strong assumptions on user behavior. To overcome that, we propose Magil, a pedometer-free approach which makes use of magnetic field for indoor localization. To the best of our knowledge, this is the first piece of work using geomagnetism for smartphone localization without need of a pedometer. Magil is applicable to any open or complex indoor environment. In the offline phase, Magil continuously collects and stores geomagnetic fingerprints...[ Read more ]

Geomagnetic field holds much promise for indoor localization due to its pervasive spatial presence, high signal stability and wide availability of magnetometers already embedded in mobile devices. Previous work in the area often fuses it with a pedometer (step counter) via particles. These approaches are computationally intensive and require strong assumptions on user behavior. To overcome that, we propose Magil, a pedometer-free approach which makes use of magnetic field for indoor localization. To the best of our knowledge, this is the first piece of work using geomagnetism for smartphone localization without need of a pedometer. Magil is applicable to any open or complex indoor environment. In the offline phase, Magil continuously collects and stores geomagnetic fingerprints while a surveyor is walking in the indoor area. In the online phase, it employs a fast algorithm to match the geomagnetic segments whose fingerprint variations best match the target observations. Given the closely matched segments, Magil constructs the user path efficiently with a modified shortest path formulation by selecting and connecting these matched segments, hence obtaining the target locations over time.

To further increase localization performance, we propose MagFi, which extends Magil by fusing it with Wi-Fi signals. In the offline phase, MagFi further collects opportunistic Wi-Fi RSSI for RSSI fingerprint construction. In the online phase, MagFi leverages RSSI to further enhance the results. We have implemented both Magil and MagFi, and conducted extensive experiments at our university campus. Our results show that both systems outperform state-of-the-art schemes by a wide margin (often cutting localization error by more than 30%).