MetroTrack: Predictive Tracking of Mobile Events Using Mobile Phones

We propose to use mobile phones carried by people in their everyday lives as mobile sensors to track mobile events. We argue that sensor-enabled mobile phones are best suited to deliver sensing services (e.g., tracking in urban areas) than more traditional solutions, such as static sensor networks, which are limited in scale, performance, and cost. There are a number of challenges in developing a mobile event tracking system using mobile phones. First, mobile sensors need to be tasked before sensing can begin, and only those mobile sensors near the target event should be tasked for the system to scale effectively. Second, there is no guarantee of a sufficient density of mobile sensors around any given event of interest because the mobility of people is uncontrolled. This results in time-varying sensor coverage and disruptive tracking of events, i.e., targets will be lost and must be efficiently recovered. To address these challenges, we propose MetroTrack, a mobile-event tracking system based on off-the-shelf mobile phones. MetroTrack is capable of tracking mobile targets through collaboration among local sensing devices that track and predict the future location of a target using a distributed Kalman-Consensus filtering algorithm. We present a proof-of-concept implementation of MetroTrack using Nokia N80 and N95 phones. Large scale simulation results indicate that MetroTrack prolongs the tracking duration in the presence of varying mobile sensor density.