Verification of High Integrity and Continuity Reference Receiver

Massimo Crisci and Herve Guichon

Abstract:	In the past years, THALES has developed reference ground receivers for GNSS systems. Due to their use in an overall ground control system, these receivers have severe requirements in terms of continuity and integrity. During the design phase, this type of statistical performance can be dealt with using a mix of theoretical analysis and simulation. However, these methods limit: - either the confidence in the achieved probability (due to uncertain assumptions in theoretical assumptions), - or the level of achievable probability demonstrated by simulation, due to computation time limitations. In particular, they cannot be used to actually demonstrate the performance by testing, and only long-term experience after deployment will confirm the validity of the assumptions and design. This entails usual troubles incurred in exploiting results of an operational system: environment is not fully controlled and rare events are difficult to analyze, because all necessary information may not be available a posteriori. Whereas these limitations have been accepted in the past as unavoidable, due to practical constraints, the designers of such complex systems now require more confidence in the design of equipments (in particular receivers) of the whole system, and cannot wait for the deployment to guarantee the system implementation. For the final verification of a recent reference receiver, THALES has improved the level of confidence in the demonstration of these low-probability requirements, by applying statistical methods used in other areas. The paper explains successively the context and algorithms for two typical cases: - false-lock on BOC signals - loss-of-lock due to ionospheric scintillation We use these real cases as support for the description of the main statistical methods used to: - estimate low probabilities at a realistic cost: variance reduction methods The focus is on the improvement of the Monte-Carlo method by “Importance sampling” (IS), and the concerns encountered in practical implementation. Two main issues (still subject to active research) are discussed: the choice of the distorted distribution, and the convergence criteria. - estimate the confidence in the value: statistical tests/confidence intervals, resampling methods (bootstrap) For each case, we provide the results of simulations demonstrating the expected probability levels. Then we explain briefly how the integration/verification process allows efficient receiver testing and show how test results agree with simulations on a few examples. The paper concludes by exploring related areas and potential future research directions: - application of the Bayesian framework. An example shows how this point-of-view might be of interest in our technical area - a brief insight into recent statistical methods, based on iterated importance sampling, - extension of these improved low-probability demonstration methods to other areas in GNSS systems, such as Advanced RAIM algorithms.
Published in:	Proceedings of the 25th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2012) September 17 - 21, 2012 Nashville Convention Center, Nashville, Tennessee Nashville, TN
Pages:	2285 - 2300
Cite this article:	Crisci, Massimo, Guichon, Herve, "Verification of High Integrity and Continuity Reference Receiver," Proceedings of the 25th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2012), Nashville, TN, September 2012, pp. 2285-2300.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In