Capacity Scaling of Sparse Mobile Ad Hoc Networks

We provide the scaling laws for the transport capacity of a wide class of mobile wireless ad hoc networks. Our analysis generalizes previous results obtained under restrictive assumptions on the node mobility process and overall node density over the network area. The broader family of mobile networks that we consider is able to account for many important characteristics usually recognized in real traces of both human and vehicular mobility. In particular, we consider clustered, sparse networks of heterogeneous nodes, in which the shape of the spatial distribution of each node around one or more home-points plays a fundamental role in determining the overall transport capacity. We identify different operational regimes that arise within our general class of mobile networks, and for each regime we propose optimal scheduling and routing strategies achieving the maximum asymptotic capacity.