This research contributes to a better understanding of how reconfigurable\udField Programmable Gate Array (FPGA) devices can safely be\udused as part of satellite payload data processing systems that are exposed\udto the harsh radiation environment in space. Despite a growing\udnumber of publications about low-level mitigation techniques, only\udfew studies are concerned with high-level Fault Detection, Isolation\udand Recovery (FDIR) methods, which are applied to FPGAs in a similar\udway as they are applied to other systems on board spacecraft.\udThis PhD thesis contains several original contributions to knowledge\udin this field. First, a novel Distributed Failure Detection method\udis proposed, which applies FDIR techniques to multi-FPGA systems\udby shifting failure detection mechanisms to a higher intercommunication\udnetwork level. By doing so, the proposed approach scales better\udthan other approaches with larger and complex systems since data\udprocessing hardware blocks, to which FDIR is applied, can easily be\uddistributed over the intercommunication network. Secondly, an innovative\udAvailability Analysis method is proposed that allows a comparison\udof these FDIR techniques in terms of their reliability performance.\udFurthermore, it can be used to predict the reliability of a specific\udhardware block in a particular radiation environment. Finally,\udthe proposed methods were implemented as part of a proof of concept\udsystem: On the one hand, this system enabled a fair comparison\udof different FDIR configurations in terms of power, area and performance\udoverhead. On the other hand, the proposed methods were all\udsuccessfully validated by conducting an accelerated proton irradiation\udtest campaign, in which parts of this system were exposed to\udthe proton beam while the proof of concept application was actively\udrunning.
展开▼
