Social based throwbox placement schemes for large-scale mobile social delay tolerant networks

• Study how to deploy throwboxes in a large-scale mobile social DTN.
• Propose a set of social-based throwbox placement algorithms.
• Simulation results from real-life traces confirm the efficiency of proposed methods.

Mobile phone sensing is a new paradigm which takes advantage of smart phones to collect and analyze data at large scale but with a low cost. Supporting pervasive communications among mobile devices in such a large-scale mobile social network becomes a key challenge for this new mobile sensing system. One possible solution is allowing packet delivery among mobile devices via opportunistic communications during intermittent contacts. However, the lack of rich contact opportunities still causes poor delivery ratio and long delay, especially for large-scale networks. Deployment of additional stationary throwboxes can create a greater number of contact opportunities, thus improve the performance of routing. However, the locations of deployed throwboxes are critical to such improvement. In this paper, we investigate where to deploy throwboxes in a large-scale throwbox-assisted mobile social DTN. By leveraging the social properties discovered from real-life tracing data, we propose a set of social-based throwbox placement algorithms which smartly pick the location of each throwbox. Extensive simulations are conducted with a real-life wireless tracing dataset and a wide range of existing DTN routing methods. The results confirm the efficiency of the proposed methods.