Modern space probes such as Solar Orbiter employ a SpaceWire network to connect to on-board computer (OBC), solid state mass memory (SSMM), and scientific instruments. Management of SpaceWire links within scientific instruments is typically performed by a data processing module (DPM) featuring a space qualified processor that is executing on-board software. To adapt to changing mission requirements, account for failures and fix possible software bugs, the ability of uploading and patching instrument software is mandatory. However uploading and over-writing of the software's boot image cannot securely be performed by the software itself. If over-writing the boot image fails, the remaining image might be corrupted. So the processor may not be able to reboot successfully and no further upload would be possible. Therefore reception of uploaded patches must be performed by an independent entity. Currently, this is accomplished by a dedicated boot loader in separate memory area, to be qualified according to ECSS criticality category B. This boot loader processes uploading of patches and copies them to the second boot area, where the actual software including the operating system is stored. Due to the opportunity of modern processors to handle SpaceWire RMAP accesses (e.g. SpW-RTC, UT699, GR712RC [1], or upcoming NGMP [2]), it would be possible to perform uploading and patching of the instrument software independent of software execution using RMAP. This would dramatically simplify the development, eliminate the need for a class-B qualified boot loader, and will inherently improve reliability, as reception of patches would entirely be performed by hardware. This paper presents a possible update and patch process for boot images using hardware based RMAP features. Furthermore implications of the standard ECSS services affecting such patching routines are discussed.Index Terms-SpaceWire, RMAP, in-flight update, boot loader, CCSDS PUS.
Abstract-Network management becomes more and more complex with growing networks. Web services are seen as a possible way to deal with the complexity. To exploit the advantages of Web services, a management system must be aware of the management functionality that is offered by the devices as Web services. In a distributed management environment, a centralized registry is not suited. In this paper, we propose a peer-to-peer registry that is based on semantic descriptions of the available management Web services. The idea is to let all devices that offer management Web services take part in the overlay peer-to-peer network, such that no external or central registry server is needed. We show how a policy-based management system that depends on Web services can benefit from such a registry, discuss some of the implementation issues, and give evaluation results.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.