Experimental Validation of an Ionospheric Monitoring Scheme for Dual Frequency GBAS

In this paper we describe a method to monitor for a difference in the ionospheric delay observed by a ground station of a Ground Based Augmentation System (GBAS) and an airborne user. In case of detection the affected satellite can be excluded or a switch to an ionospheric free processing mode can be triggered. As it is not possible to estimate the absolute ionospheric delay at the ground station from the transmitted corrections directly, we compare a pseudo-ionospheric delay estimate from the corrections with an ionospheric estimate after removing biases. We then show the performance of the proposed monitoring architecture in flight trials from our GBAS test environment for different scenarios. In order to obtain results for a full constellation we considered an L1/L2 dual frequency combination where the expected noise and multipath is larger than in the L1/L5 case which will be used in an operational GBAS. Results show that even with the larger test statistic and in the single constellation case the monitor is feasible and provides good results. Furthermore, we also show results using the available L5/E5a signals collected during flight tests. Also in this case the performance is good and the threshold was not exceeded despite the low threshold of the monitor due to the marginal geometry. In order to test the reaction of the monitor to ionospheric gradients we injected a simulated error into the airborne measurements. As the monitoring threshold depends on the satellite geometry we show some exemplary results of the impact of different gradient slopes on the test statistic of the affected satellites.