AI-Based Ground Station-as-a-Service for Optimal Cost-Latency Satellite Data Downloading

Recent advances in satellite technology and the introduction of the Space-Data-as-a-Service business models are attracting interest for using the platform in a wide range of applications that include earth observations, 5G integration, aeronautical and maritime tracking and communications, asset tracking, sensor data collection, among others. In this new scenario, it becomes of interest not only selecting a ground station provider based on the monetary cost, but also the download features with prevailing atmospheric conditions to achieve high reliability and low data delivery latency. To this end, the Cognitive Space Gateway approach is used to route data bundles in a Ground Station-as-a-Service context with multiple providers to find a suitable balance between data latency and the ground station monetary costs, exploiting as possible site-diversity when available. The approach is tested experimentally using NASA's High-Rate Delay-Tolerant Networking (HDTN).