DISCO: Ultra-Lightweight Mobility Discovery

Abstract

Capturing individual mobility patterns has become a crucial issue for a tremendous number of applications, often requiring the use of privacy-invasive or energy-consuming sensors and online services. In parallel to this, the proliferation of wireless network access points (APs), scattered in a very dense manner in many geographical areas, is now opening up new technological opportunities. In this work, we demonstrate the use of network discovery data passively collected from Wi-Fi APs to infer mobility indicators. This local approach can potentially be implemented on any device with a communication interface, and allows for continuous and long-term data collection. The demo showcases a multi-platform mobile app (DISCO) and is presented alongside an extended desktop analysis toolbox. Additional material can be found online1.