Performance Evaluation of PPP Integrity Monitoring Based on Multiple Hypothesis Solution Separation for BDS/GPS/Galileo
Integrity monitoring is a crucial technique for ensuring the reliability of Global Navigation Satellite System (GNSS) navigation and positioning. While Precise Point Positioning (PPP) currently meets the high-precision requirements of most applications, safety-critical applications such as UAVs and autonomous driving, impose even stricter demands on the reliability of PPP. How to maintain both high precision and high integrity simultaneously remains a challenge. To achieve this goal, conducting research on integrity monitoring for PPP is of paramount importance. This study systematically evaluated the PPP integrity monitoring performance of BDS, GPS, and Galileo constellations respectively, using the Multiple Hypothesis Solution Separation (MHSS) algorithm as the core framework. This study selected observation data collected on July 1, 2025, from 18 globally distributed MGEX (Multi-GNSS Experiment) stations, along with precise ephemeris and clock products provided by an analysis center at Wuhan University. By simulating faults, we assessed the capability of the MHSS algorithm in calculating the protection level (PL) and performing fault detection and exclusion (FDE) for BDS, GPS, and Galileo constellations. The results indicated that the average horizontal protection level (HPL) and vertical protection level (VPL) for the BDS, GPS, and Galileo constellations was 3.009 m, 2.305 m, 5.259 m and 2.143 m, 1.594 m, 3.863 m, respectively. The global average availability of integrity monitoring for BDS, GPS, and Galileo in the horizontal direction reached 82.41%, 93.07%, 68.30%, and in the vertical direction was 81.69%, 89.80%, 63.62%, respectively. Moreover, BDS demonstrated superior integrity monitoring service performance in the mid-latitude Asia-Pacific region. Under simulated step faults in pseudorange and carrier-phase observations, the average fault detection rates for BDS, GPS, and Galileo reached 89.86%, 83.71%, 92.41% and 96.02%, 94.21%, 95.66%, respectively. At fault-free epochs, the PL consistently bounded the positioning error (PE). However, due to the re-initialization of the covariance matrix during fault exclusion, the PL exhibited a re-convergence behavior.
Authors
- Zhen Jiang (ORCID: https://orcid.org/0009-0000-2161-3699)
- Junqiang Han (ORCID: https://orcid.org/0000-0002-8707-9297)
- Xiaochun Lu (ORCID: https://orcid.org/0009-0004-0919-8815)
- Lihong Fan
- Yunfei Ling
- Rui Zhang
Institutions
- Chinese Academy of Sciences (CN)
- National Time Service Center (CN)
Publication Details
- Journal
- Research Square
- Published
- 2026-11-18
- DOI
- https://doi.org/10.21203/rs.3.rs-10662093/v1
- Primary Topic
- GNSS positioning and interference
- Type
- preprint
Funders
- National Natural Science Foundation of China
- Youth Innovation Promotion Association of the Chinese Academy of Sciences
- Wuhan University