Ad hoc networks are expected to be an essential ingredient in tomorrow's communication sys- tems. Important applications include highly exible personal communication systems, rescue and disaster recovery operations. Wireless sensor networks that are envisioned to be indispensable in future infrastructure: from manufacturing to medical { can also be considered a special kind of ad hoc networks. This research studies the problem of overall performance (eciency, quality of service, and robustness) of large realistic ad hoc networks which are, in particular, characterized by variable node density and constant changes in the network due to node mobility, node failure, changing node membership, fading etc. While some progress has been made in understanding the main factors a ecting the performance of ad hoc networks, theoretical exploration of the fundamental performance characteristics of realistic networks that possess essential space- and time-inhomogeneity has been largely absent. This research attempts to close this gap and develop a theoretical framework for the analysis of fundamental limitations and performance of such complex large scale ad hoc networks. On a larger scale, this research contributes to the analytical methods of study of higher network layers by developing general analytical machinery for the analysis of the network layer in the presence of spatial and temporal inhomogeneity.
