Poster: Detecting if a Smartphone is Indoors or Outdoors with Ultrasounds

Abstract

Context information is fundamental for mobile application. A system able to detect if a smartphone is indoors or outdoors can give useful information to upper-level applications, permitting to improve their performance or reduce the computational load and consequently the lifetime of the smartphone battery. For example, GPS provides an accurate location reference in the outdoor environment while it performs poorly inside buildings. The proposed Indoor/Outdoor (IO) detector can provide a useful essential information to a localization application that can check whether the user is outdoors before turning on the GPS interface and decide not to turn it on and use other localization methods if the user is detected indoors. In mobile data services, mobile phones normally observe more WiFi access points (APs) with strong signals inside buildings, whereas it is unlikely to have good WiFi connections in outdoor environments. Therefore, knowing whether the smartphone is indoors or outdoors can help to make smarter decisions regarding whether to perform or not AP scans. Although it is clear that various applications may benefit from accurate and prompt indoor/outdoor information, the research work on indoor/outdoor detection of mobile devices is still lacking. There are mainly two techniques to perform such detection. One is to use GPS and its signal quality as a cue to infer if the user is indoor. This technique is proven to be highly power consuming. Another technique is to leverage the sensors which the smartphone is equipped with. IO detection can be done by exploiting lightweight sensors such as the light sensor, the radio interface and the magnetism sensor [1]. The proposed IO detector is an active system. The phone periodically emits an ultrasonic ping using its in-built speakers and continuously listens for the echoes through its microphone. It is impossible to identify the direction of the echoes being the microphone (as well as the phone speakers) non-directional. The idea is that indoors the number and the intensity of the echoes should be higher than outdoors due to the higher number of obstacles. Translating such an idea into a practical Indoor/Outdoor detector means finding those features that model such behaviour. The proposed