Network-Based Ionospheric Gradient Monitoring to Support Ground Based Augmentation Systems

The Ground Based Augmentation System (GBAS) is a local-area, airport-based augmentation of Global Navigation Satellite Systems (GNSS) that provides precision approach guidance for aircraft. It enhances GNSS performance in terms of integrity, continuity, accuracy, and availability by providing differential corrections and integrity information to aircraft users. Differential corrections enable the aircraft to correct spatially correlated errors, improving its position estimation. Integrity parameters enable it to bound the residual position errors, ensuring safety of the operation. Additionally, a GBAS ground station continuously monitors and excludes the satellites affected by any system failure to guarantee system integrity and safety. Among the error sources of GNSS positioning, the ionosphere is the largest and most unpredictable. Under abnormal ionospheric conditions, large ionospheric gradients may produce a significant difference between the ionospheric delay observed by the GBAS reference station and the aircraft on approach. Such a spatially decorrelated ionosphere could lead to hazardous unbounded position errors if undetected. Conventional GBAS solutions to mitigate this threat assume that the “worst-case” ionospheric gradient ever observed in the relevant region is always present, which is a very conservative assumption. This approach, which relies on the conservative ionospheric threat models derived for GBAS, maximizes integrity, often at the expense of availability and continuity, especially in geographic areas with highly active ionosphere. As opposed to assuming a permanent “worst-case” gradient, I propose the Network-GBAS concept, in which several reference stations collaborate to monitor for actual ionospheric gradients. This concept consists of two main steps. First, the network detects the anomalous ionospheric gradients, estimates the gradient parameters, and transmits this information to the GBAS stations installed in its coverage area. Then, the GBAS stations replace the “worst-case” gradient used to mitigate the ionospheric threat in current algorithms with the gradient information provided by the network. This approach reduces conservatism and leads to an improvement of the system availability without compromising user integrity. This thesis validated the performance of the detection and estimation algorithms with simulated and real ionospheric gradients from two different locations known for their high levels of ionospheric activity. One location was Alaska, where the analyzed real anomalous gradients were small in size but fast-moving; the other location was Brazil, dominated by large-but-slow anomalous gradients. This analysis led to the adaptation of the algorithms to work in challenging scenarios. The evaluation of the Network-GBAS concept compared in simulations the availability of a Category I (CAT I) GBAS station at the Brazil location in two cases: assuming the conservative ionospheric threat model, and using the gradient information provided by the network. On a selected nominal day (i.e., with no significant ionospheric activity availability improved from 79.5% to 94.6% during the nighttime. On a selected active day, availability improved from 68.7% to 89.5% during the nighttime. During the daytime, availability achieved 100% on both days. Results demonstrate that the Network-GBAS concept can significantly enhance CAT I GBAS availability in active ionospheric regions without compromising user integrity. Furthermore, by incorporating the information provided by the network into existing solutions, the Network-GBAS is compatible with existing algorithms and hardware, and thus should be certifiable if adapted to the characteristics of each region where GBAS is fielded.