System Level Diagnosis Applied to GNSS

Chad C. Lamb, John Fagan, Anindya Das, Ralph Sexton

Abstract:	The comectivity of a system ia an important measure of fault toleranm. ~ical interconnection networks are sparse and symmetrical. The classical System-Level Diagnosis (SLD) theories, such as t-diagnosability, are not appropriate for these systems because the number of faulty elements is limited by the minimum node degree, which is very small for sparse systems. Another &advantage of the classical approach is the need for a supervisory procesaor. This processor is a bottleneck in the systeu unneceasady burdening the communication paths. The supervisor processor is also a single-point fhilure in the system. The SLD theory oft-in- L2 diagnosability overcomes these limitations. We investigate the application of SLD to the GNSS to assess the diagnosability and fault tolerance of the system. We show, in a determinktic manner, the number of faulty elements that can exist in the system and still allow for the accurate diagnosis of the system. Our deterministic approach is in contrast to other known probabilistic approach. In this research we investigate the local diagnosability of the GNSS for permanently faulty elements. In t-in-L2 diagnosabi[ity theory, if there are at most tXtlndty elements in the local neighborhood set of x (the aircmft), and at most t(m) fimlty elements in the extended-local neighborhood set of A then we can uniquely diagnose the GNSS. The extended-local 12mlt constrai@ t(m), is a fimction of the local fimlt constraint k, and the degree of node x.
Published in:	Proceedings of the 54th Annual Meeting of The Institute of Navigation (1998) June 1 - 3, 1998 The Adams Mark Hotel Denver, CO
Pages:	647 - 655
Cite this article:	Lamb, Chad C., Fagan, John, Das, Anindya, Sexton, Ralph, "System Level Diagnosis Applied to GNSS," Proceedings of the 54th Annual Meeting of The Institute of Navigation (1998), Denver, CO, June 1998, pp. 647-655.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In