High-Availability Integrity Monitoring for Multi-Constellation GNSS Navigation with Non-Gaussian Errors

Abstract: Global navigation satellite systems (GNSS) are essential for aviation, requiring strict integrity monitoring to alert users to hazardously misleading information. Conventional receiver autonomous integrity monitoring (RAIM) and advanced RAIM (ARAIM) rely heavily on Gaussian models in bounding nominal errors, which can be overly conservative with real-world non-Gaussian errors with heavy tails, such as the satellite clock and orbit errors. This paper proposes an extended jackknife detector capable of detecting multiple simultaneous faults with non-Gaussian nominal errors. Furthermore, an integrity monitoring algorithm, jackknife ARAIM, is developed by systematically exploiting the properties of the jackknife detector in the range domain. A tight bound of the integrity risk is derived by quantifying the impacts of hypothetical fault vectors on the position solution. The proposed method is examined in worldwide simulations, with the nominal measurement error simulated based on authentic experimental data, which reveals different findings in existing research. In a setting of a single Global Positioning System (GPS) constellation, the proposed method reduces the 99.5 percentile vertical protection level (VPL) 45m, where the VPL of the baseline ARAIM is larger than 50m in most user locations. For dual-constellation (GPS-Galileo) settings, baseline ARAIM suffers VPL inflation over 60m due to the over-conservatism induced by the heavy-tailed Galileo signal-in-space range errors, whereas the proposed jackknife ARAIM retains VPL below 40m, achieving over 92% normal operations for a 35m Vertical Alert Limit. These improvements have promising potential to support localizer performance with vertical guidance (LPV) with a decision height of 200 ft, enhancing integrity and availability for multi-constellation GNSS applications.