Towards worldwide GNSS-based landing systems : modeling and probabilistic integrity assessment of low-latitude ionospheric threat = GNSS 기반 착륙 시스템의 전 지구적 운용을 위한 저위도 지역의 전리층 위협 모델링 및 확률적 무결성 평가 기법 연구 modeling and probabilistic integrity assessment of low-latitude ionospheric threat

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 15
  • Download : 0
The Ground-Based Augmentation System (GBAS) supports Global Navigation Satellite System (GNSS)-based aircraft approach and landings by providing differential corrections and integrity information to aircraft users. A GBAS ground facility continuously monitors and excludes the satellites affected by any system failure to guarantee the system integrity and safety. Among the error sources of GNSS positioning, the ionosphere is the largest and most unpredictable. Under unusual ionospheric conditions, the electron density varies dramatically and unpredictably, even across small regions. Such a spatially decorrelated ionosphere, if undetected, may cause an extremely large time delay difference in GNSS signals over short distances of just a few tens of kilometers, leading to unacceptably large residual errors even after applying differential corrections. The ionospheric threat mitigation is performed by utilizing the ionospheric anomaly threat model to assess the maximum undetected position errors that users should be protected from and to identify potential ionospheric threats. However, the ionospheric behavior differs from region to region. The current ionospheric threat model was developed based on historical GNSS observation data obtained from the Conterminous United States (CONUS) at mid-latitudes and thus cannot be applied any other regions such as those in low-latitude regions. Hence, it is required to develop new ionospheric threat model and mitigation method fitted to each region for worldwide GBAS operations. This thesis presents new ionospheric anomaly threat model, validation methodology, and mitigation technique for GBAS operations in low-latitude regions where the ionospheric activity is known to be the most intense in the world. The thesis initially develops new ionospheric anomaly threat model that captures the low-latitude ionospheric behavior which is significantly different from what encountered in mid-latitude regions such as CONUS. Data processing from the GNSS for 123 active ionospheric days identifies 1017 anomalous ionospheric gradients caused by nighttime Equatorial Plasma Bubbles (EPBs). A significant number of gradients, including the largest verified gradient of 850.7 mm/km, exceed the upper bound (375–425 mm/km) of the CONUS threat model. This thesis defines a series of parameters to model the geometry of EPBs. A maximum ionospheric delay drop of 35 m and a transition zone between 20 and 450 km are estimated for EPBs that move roughly eastward and parallel to the geomagnetic equator with speeds of between 40 and 250 m/s. While GNSS remote sensing techniques have been used to measure ionospheric gradients, the measurements can easily be corrupted during ionospheric disturbances. Extremely large ionospheric gradients are required to be validated before being declared to be real ionospheric events as opposed to artifacts of erroneous measurements. The use of existing methods is however limited due to the small size of EPBs compared to the baseline distances between GNSS network stations in low-latitude regions. Thus, this thesis proposes a multi-dimensional validation procedure which provides a comprehensive analysis of EPB gradient impacts in the spatial and temporal domains. This methodology utilizes a time-step method to estimate gradients over any short distance. Equatorial anomaly events are visualized in multiple time series by combining all available sources, including severe gradients observed from multiple widely spread stations, the estimated EPB and known satellite motions, and the known station locations. A similar ionospheric pattern over multiple station-satellite pairs supports the fact that they are impacted by the same EPB at different times and locations. An extreme ionospheric gradient of 501.2 mm/km observed in the Brazilian region during the 31 December 2013 EPB event is validated to be real using this methodology. The results demonstrate the effectiveness of the methodology for validating EPB-induced ionospheric spatial decorrelation. Lastly, we assess EPB impacts on GBAS performance and develop new mitigation technique that improves the system availability. We first apply Position-Domain Geometry Screening (PDGS) which has been previously developed to mitigate anomalous ionospheric gradients induced by mid-latitude ionospheric storms. This method utilizes the threat model to estimate the largest position errors under the worst-case situations in which an extreme gradient is presumed to always exist. PDGS with a higher gradient bound for the EPB model brings excessive inflations of integrity parameters to eliminate a large number of potentially unsafe satellite geometries, decreasing the availability to 58.3%. A new probabilistic mitigation method, Monte Carlo-based PDGS (MC-PDGS), randomizes ionospheric scenarios using randomly generated parameter combinations within the threat model and assesses the ensemble impacts. By taking credit for a prior probability of an extreme EPB, this algorithm determines the inflated integrity parameters to meet the safety requirement in the probabilistic definition. This thesis shows that with this method, the system availability for category I precision approaches dramatically improved to 89.6% when a data-driven prior probability of $10^{-5}$ was applied. This performance improvement can be achieved without critical changes in the GBAS ground facility and avionics.
Lee, Jiyunresearcher이지윤researcher
한국과학기술원 :항공우주공학과,
Issue Date

학위논문(박사) - 한국과학기술원 : 항공우주공학과, 2018.8,[vi, 119 p. :]


Ground-based augmentation system (GBAS)▼aionospheric spatial gradient▼aionospheric anomaly threat model▼aionospheric threat mitigation▼aequatorial plasma bubble; 지상기반 보강시스템(GBAS)▼a전리층 공간 기울기▼a전리층 위협 모델▼a전리층 위협 완화▼aequatorial plasma bubble

Appears in Collection
Files in This Item
There are no files associated with this item.


  • mendeley


rss_1.0 rss_2.0 atom_1.0