Ranging Source Fault Detection Performance for Category III GBAS

M. Brenner, F. Liu

Abstract:	The most stringent goal for guidance applications in aviation is the category III weather minima. Category III conditions split into three sub categories a, b and c. Category a typically means the pilot has visual cues at 50 feet above the threshold, category b means the pilot has visual cues below 50 feet and category c includes guidance throughout touchdown and roll out. The satellite based one-frequency navigation standard for Category III referred to as GAST D was developed by RTCA and ICAO. Although Category I conditions are the most frequently encountered, the airplane manufacturers need to provide category II and III capability as well and are looking for equipment that is capable of providing guidance during all weather conditions. Ground based augmentation systems demonstrated early on that they could provide the required accuracy but as the understanding of the ranging source fault modes developed it became clear that the ILS look alike requirements-methodology used for category I, would be too conservative and therefore too limiting for category III. A different requirements methodology was therefore used in the GAST D standard. GAST D focuses on the touchdown requirements as defined in AC120-28 and CSAWO and develops fault detection requirements that will assure such touchdown requirements are met. As of 2010 the ranging source fault detection requirements applicable to the GBAS ground station have been formulated in the ICAO SARPs draft for category III GBAS. The four ranging faults considered in the SARPs draft are signal deformation, excessive acceleration, excessive carrier code divergence and excessive ephemeris errors. The requirement for such ranging faults is formulated as a misdetection probability curve which is a function of the differential range error in meters and tied to a specific time to alarm. The objective of this paper is to explain how the monitors in the CAT I ground station SLS 4000 SmartPath can be modified to meet this requirement and also to provide supporting validation material. The 4 ranging source fault mechanisms are analyzed and the worst case misdetection curve is calculated for all possible combinations of threats, airborne receiver configurations and detection constraints. The validation is based on experimental data for monitor noise characterization and threshold setting and simulated data for fault impact analysis. The paper focuses on the areas where there are differences versus the CAT I ground station and monitoring functions that are not changed versus CAT I are not described in detail. For the acceleration monitor the main challenge is the shorter time to alert in GAST D and the analysis concentrates on the very large acceleration onsets that would be hazardous to an aircraft below 200 feet. For signal deformation the challenge is to deal with the shorter time constant of 30 sec in GAST D in combination with the reduced time to alert. This will increase the noise levels and also make the fault induced error ramp up faster. The challenge for the carrier code divergence is the quicker fault response. The ephemeris fault impact is not changed much from CAT I and the shorter time to alert does not have any significant impact. The main objective here is to make sure this conclusion holds in all the corner cases. The conclusion reached in the paper is that the ranging source monitors can be updated to meet the GAST D requirements. However, this assumes that stringent siting constraints are applied to assure the specific continuity risk requirement for CAT II-III spelled out in the SARPs draft is satisfied. The paper also formulates an airborne requirement for the minimum time from initialization of the 30 second smoothing filter, to when the satellite is incorporated in the guidance solution. This requirement has been provided to aid the process of establishing a clear interface between ground and air for GAST D. The CAT II-III ranging source fault detection is very demanding given the increased noise levels due to a shorter time constant and the short time to alert. Several innovative steps have been taken as part of the upgrade process to satisfy these new requirements. This is especially the case for the acceleration monitor. The analysis method that builds on an established minimum time delay before inclusion in to the guidance solution is also new. The ground station ranging source fault detection is an essential component of a GAST D capable system and the analysis presented in this paper was performed in support of the GAST D standard technical validation process.
Published in:	Proceedings of the 23rd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2010) September 21 - 24, 2010 Oregon Convention Center, Portland, Oregon Portland, OR
Pages:	2618 - 2632
Cite this article:	Brenner, M., Liu, F., "Ranging Source Fault Detection Performance for Category III GBAS," Proceedings of the 23rd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2010), Portland, OR, September 2010, pp. 2618-2632.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In