Automated Data Retrieval from Large-Scale Distributed Satellite Systems

We consider a central challenge that is mission critical for the successful operation of large-scale satellite constellations in Low-Earth Orbit: How can we coordinate the short-term download operations for the enormous amounts of generated data, based on wireless line-of-sight connections to a limited number of stationary ground station? These issues are critical for the future growth of space systems, with multiple commercial space operators competing for downloading their commercial data in a timely fashion, relying on the services of a scarce set of ground stations that is subject to numerous strong constraints, so it cannot simply be expanded. We present a distributed auction-based scheduling approach for maximizing the value of the downloaded data. Our method allows competing satellite operators to bid for contact times and has a fair and transparent price estimation based on the competition. On its own, it can also be used with a simple bidding strategy to obtain good schedules; this is demonstrated on benchmark simulation with up to 1080 satellites. As a consequence, we are able to achieve values rates of 74% of available data, compared to 28% for standard greedy strategies.