Shaping opportunistic networks

This paper deals with the problem of identifying and managing overlay topologies in opportunistic networks. To make this possible, we define the concept of “link” in an opportunistic network, which cannot be represented as a simple communication connection between two nodes at a given instant (as in classic MANETs). Rather, links represent contacts among nodes arising during a (non-instantaneous) time interval. With this definition in view, opportunistic networks can be seen as evolving graphs. The interesting thing is that it becomes possible to control the way nodes manage contacts and build overlays based on some desired overlay topology. This is an important result, since the statistical characteristics of the topology allow estimating the ability of the opportunistic network to spread contents. An experimental assessment based on a trace-driven simulation employing different types of datasets (ranging from urban dense networks to very sparse ones) confirms the viability of the proposed approach.