Workshop on interdisciplinary systems approach in performance
evaluation and design of computer & communication systems

The design and analysis of wireless networks is typically plagued by the so-called scalability problem: as the number of the nodes increases, problems become more complicated, and results come harder. A very promising solution to this problem is the *massively dense* assumption. In particular, we assume that the network has so many nodes, that, it addition to the standard *microscopic* view of the network, a new *macroscopic* view emerges. The macroscopic model of the network can be optimized using mathematics rarely used in networking, for example partial differential equations and calculus of variations, and is related to various branches of physics such as Optics and Electrostatics. In this work, we introduce the massively dense concept, and present a number of very promising results, that show that the massively dense assumption has the potential to shed light and give intuition in the design of large wireless networks.

Talk slides