Spatially inhomogeneous survivable communication networks with directed and non-directed graphs

Natural and Intentional Electromagnetic Interference (IEMI) on a network's communications nodes can cause combinations of short-term upset, long-term upset and permanent damage to equipment. In this paper we provide new theoretical principles based on random graph theory and percolation theory to evaluate the resilience of large random geometric ad-hoc networks formed with non-directed and directed graphs. By combining these graphical representations in a common theoretical framework we provide a theoretical basis for expanding the theory to inhomogeneous networks that are more closely connected to those found in nature and in the military. The proposed metric for survivability is the network's ability to remain radio-frequency connected with a suitable surviving fraction of nodes.