End-to-end System Performance Simulation: A Data-Centric Approach

摘要

In the early times of space industry, the feasibility of Earth observation missions was directly driven bywhat could be achieved by the satellite. It was clear to everyone that the ground segment would beable to deal with the small amount of data sent by the payload. Over the years, the amounts of dataprocessed by the spacecrafts have been increasing drastically, leading to put more and moreconstraints on the ground segment performances - and in particular on timeliness. Nowadays, manyspace systems require high data throughputs and short response times, with information coming frommultiple sources and involving complex algorithms. It has become necessary to perform thorough endto-end analyses of the full system in order to optimise its cost and efficiency, but even sometimes toassess the feasibility of the mission.This paper presents a novel framework developed by Astrium Satellites in order to meet these needsof timeliness evaluation and optimisation. This framework, named ETOS (for “End-to-end TimelinessOptimisation of Space systems”), provides a modelling process with associated tools, models andGUIs. These are integrated thanks to a common data model and suitable adapters, with the aim ofbuilding suitable space systems simulators of the full end-to-end chain. A big challenge of suchenvironment is to integrate heterogeneous tools (each one being well-adapted to part of the chain) intoa relevant timeliness simulation.