Special Section on Autonomic and Opportunistic Communications

It is our great pleasure to introduce this Special Section of the Journal, focused on Autonomic and Opportunistic Communications. We strongly believe autonomic and opportunistic properties will be a key feature of the Future Mobile Internet. The huge proliferation of mobile devices with wireless networking capabilities makes it possible to foresee a Future Internet environment in which users’ mobile devices will spontaneously network together and build self-organizing wireless networks for enabling users interaction and content exchange. This will be a natural enabler for the take off of User Generated Content and other user-centred networking models in the area of pervasive mobile networks. While this perspective opens up exciting application opportunities, several research challenges have still to be addressed. Spontaneous mobile networks are typically less stable than the conventional Internet, and therefore nodes must embed greater ‘‘intelligence” with respect to Internet core nodes for the network to operate efficiently. More in general, a Future Internet incorporating a huge number of possibly mobile devices is likely to require more ‘‘distributed intelligence”, which should enable localised decision making in order to scale up in size and complexity while remaining efficient. Autonomic and opportunistic communication technologies will play a key role from this standpoint, as they will allow individual nodes to be aware of their networking environment, reason about it, and act accordingly. The four papers in this special section provide outstanding examples of autonomic and opportunistic solutions for Future Internet protocols. The first three papers consider opportunistic networking environments, i.e., self-organising mobile networks in which the existence of continuous multi-hop paths cannot be granted. The paper, ‘‘Distributed Estimation of Global Parameters in Delay-Tolerant Networks” by A. Guerrieri et al. proposes and analyses fully distributed protocols for estimating global parameters, such as the number of nodes in the network, the maximum, minimum or other aggregates of parameters available at individual nodes. Different approaches are proposed and compared in terms of accuracy of estimation, and guidelines are provided to select the most suitable approach in practice. The second paper,