The Airplace Indoor Positioning Platform for Android Smartphones

Summary

A short demonstration video (90 seconds) can be viewed online here (800x450 MP4). This video is also available to download here (800x450 MP4, ~13.5MB).

A narrated video tutorial on how the RSS Logger and Find Me applications work can be viewed online here (640x480 MP4). This video is also available to download here (1024x768 MP4, ~9MB).

Screenshots of the Airplace system are available here.

 

Nowadays, the vast majority of mobile devices, feature ubiquitous Internet connectivity through wireless or cellular networks, while according to recent statistics people tend to spend 80-90% of their time in indoors environments. These two facts have revitalized the interest in indoor location-aware applications that require alternative solutions for the provision of accurate and reliable location estimates because satellite-based positioning, e.g. GPS, is unavailable inside buildings. This has motivated the development of positioning algorithms that rely on existing WLAN infrastructure and exploit Received Signal Strength (RSS) fingerprints to determine location, owing to the wide availability of WLAN Access Points (AP) and the ease of collecting RSS samples without specialized equipment.

So far, the focus has been on improving accuracy and the proposed methods are usually evaluated only in terms of the accuracy in small-scale setups. However, the time required to estimate user location is equally important as it may violate the real-time constraints of an application. Moreover, low power consumption is another critical requirement to preserve valuable energy. Yet, both the estimation time and the battery depletion during positioning have received little attention, especially in the context of mobile devices.

Our Airplace system is an end-to-end mobile positioning platform developed on Android smartphones that facilitates the construction of the radiomap and features real-time positioning, while offering the additional option to assess the performance of different fingerprint-based algorithms with respect to the positioning time, accuracy and power consumption.