Optimized GNSS Station Selection to Support Long-Term Monitoring of Ionospheric Anomalies for Aircraft Landing Systems

Abstract

Differential global navigation satellite systems (GNSS)-based aircraft precision approach and landing systems require the development of ionospheric threat models to insure that users are sufficiently protected against ionospheric anomalies. The long-term ionospheric anomaly monitor (LTIAM) is being used to build ionospheric threat models for ground-based augmentation systems (GBAS) and to continuously monitor ionospheric behavior over the life cycle of GBAS. While LTAIM exhaustively detects all potential anomalies, the use of poor-quality GNSS data degrades the accuracy of ionospheric delay estimates and produces many faulty anomaly candidates, thus adding a great burden to LTIAM processing. To select GNSS reference stations with high-quality data, an optimized set of thresholds for data quality metrics are established. The high-quality station selection method maximizes the elimination of spurious gradients while minimizing unnecessary station removals. When applied to the continuously operating reference stations (CORS) network in the Conterminous U.S. (CONUS), this method discards 90% of faulty candidates while only excluding 14% of the over 1600 CORS stations. The well-distributed subnetwork selection method is also proposed to remove geographically redundant stations in dense regions. The number of CORS stations in CONUS is reduced to 48% of total stations when a desired baseline constraint is 100 km. The results demonstrate that the optimal GNSS station section methods are applicable to a wide range of GNSS station networks that will be used for ionospheric monitoring.