Dynamical Influence Driven Space System Design

Abstract

Complex networks are emerging in low-Earth-orbit, with many thousands of satellites set for launch over the coming decade. These data transfer networks vary based on spacecraft interactions with targets, ground stations, and other spacecraft. While constellations of a few, large, and precisely deployed satellites often produce simple, gridlike, networks. New small-satellite constellations are being deployed on an ad-hoc basis into various orbits, resulting in complex network topologies. By modelling these space systems as flow networks, the dominant eigenvectors of the adjacency matrix identify influential communities of ground stations. This approach provides space system designers with much needed insight into how differing station locations can better achieve alternative mission priorities and how inter-satellite links are set to impact upon constellation design. Maximum flow and consensusbased optimisation methods are used to define system architectures that support the findings of eigenvector-based community detection.