CAREER: Random Networks for Wireless Communication

University Of California-San Diego, La Jolla CA

Investigators

Abstract

Motivated by the need to understand the basic properties of next generation large-scale wireless systems, this project introduces new paradigms for their analysis and design that lie at the frontier of statistical physics and network information theory. Leveraging state of the art tools from random graphs and percolation theory, as well as developing new methods, the objective is to set the foundations for a new theory of large-scale communication systems. In wireless networks there is no a priori notion of static links, but only spatial configurations of nodes that can share the medium in arbitrarily complex ways. Hence, a new dynamic definition of network link is introduced. This depends on the notion of interference and of information rate. Exploiting geometric properties of node configurations, a series of fundamental problems that range from achievable throughput to energy efficiency, fault tolerance, and distributed network coding are then considered. The research effort is complemented by an educational effort to train new engineers through curriculum development, plans to introduce a new course on random networks to be taught at the graduate level, and through the writing of a book devoted to random networks for communication. Expected results include the foundations of a new mathematical theory that will lead to basic advancements in both the current understanding and the predicted behavior of large-scale networks of wireless devices. Results are also expected to have implications well beyond the field of wireless communication networks, both in terms of the introduction of new mathematical methods, and of their applicability to other fields of science.