Cluster-based location report for person tracking in wireless sensor networks (original) (raw)

Discovering Co-Located Walking Groups of People Using iBeacon Technology

Co-located mobile users have found several useful and real-world applications in proximity-based services. Aiming at unleashing the potential of these proximity-based services, it is essential to devise robust techniques enabling smart devices to know their proximity close neighbors and be able to communicate with each other. To this end, we propose, design, and evaluate a robust framework capable to successfully co-localize walking groups of mobile users, in real-time and in a centralized manner. It leverages Bluetooth low energy technology to achieve a high degree of co-location accuracy. From the collected radio signals, we construct a graph network in which the distance between pairwise vertices represents the connection strength between mobile users. Then, we propose a modified version of edge betweenness techniques, with an average path length, as a key enabler for a high clustering accuracy. We analyze the performance in terms of clustering accuracy of the proposed scheme. First, we assess its performance numerically. Then, we conduct analysis on the experimental data set to demonstrate the feasibility and the efficiency of our method. Through obtained results, we have shown that our method can be successfully applied to co-localize people walking as part of the same group.