Inferring content-centric traffic for opportunistic networking from geo-location social networks

Opportunistic networking has been proposed to support a number of novel applications, like content sharing or mobile data offloading, that follow a content-centric communication model, i.e., many users are interested in the same content. Users' traffic demand patterns can crucially affect the performance of such applications, but our knowledge about the characteristics of content demand is limited. Nevertheless, opportunistic networking is known to exhibit strong locality and social characteristics. For this reason, in this paper we argue that some initial insights about opportunistic traffic patterns could be inferred from geo-social network data. In particular, we study the check-in patterns of users in datasets of two real Location-Based Social Networks, towards understanding potential traffic characteristics and implications for opportunistic networking.