Demo: Indoor geolocation on multi-sensor smartphones
AuthorLi, C. -L
Tsai, Y. -K
Zeinalipour-Yazdi, Constantinos D.
Panayiotou, Christos G.
SourceMobiSys 2013 - Proceedings of the 11th Annual International Conference on Mobile Systems, Applications, and Services
11th Annual International Conference on Mobile Systems, Applications, and Services, MobiSys 2013
Google Scholar check
MetadataShow full item record
In this demo, we present an efficient hybrid indoor positioning solution that uses multi-sensory location-oriented observations, including WiFi, accelerometer, gyroscope and digital compass data, that are widely available on Android smartphones1. Our system mainly comprises three building blocks, namely the WiFi Fingerprinting, the Inertial Measurement Unit (IMU) Positioning and the Location Fusion components. The WiFi Fingerprinting module relies on existing WiFi infrastructure and exploits Received Signal Strength (RSS) values from neighboring Access Points (AP) to infer the unknown user location. Specifically, it utilizes a number of RSS fingerprints collected a priori to build the so-called radiomap. Subsequently, the WiFi-based location is estimated with a state-of-the-art algorithm that exploits the currently measured fingerprint and fingerprints in the radiomap . The IMU Positioning module performs multi-dimensional (i.e., 3-axis accelerometer, gyroscope and digital compass) motion sensor fusion for calculating the user orientation in real-time and implements an in-house pedometer algorithm for pedestrian trajectory tracking. An interesting feature in our implementation is the use of raw magnetic data to detect magnetic anomalies, which are common inside buildings, e.g. due to power cables, electrical appliances or metal surfaces, in order to refine orientation. Moreover, a map-matching submodule performs error correction in order to handle inaccurate IMU location estimates (e.g., showing a user passing through a wall or moving into a restricted area). Finally, the WiFi-based and IMU-based location estimates and associated uncertainties are provided as inputs to the Location Fusion module that implements the hybridization scheme by means of a particle filter. Thus, our prototype system delivers a smooth final location estimate that is consistent with the actual travelled pathsee Fig. 1. We will demonstrate the real-time positioning capabilities of our hybrid system by allowing attendees to carry an Android smartphone running our tracking application and viewing their current location on a floorplan map, while walking around the demo area. In this interactive scenario, the participants will be able to appreciate the potential of our indoor geolocation system, which is reliable and attains a localization error below 3 meters through the integration and optimization of diverse technologies, while our software may run on any commercial Android smartphone.