Statistical models to provide meaningful information to GNSS users in the presence of ionospheric scintillation

Abstract

Ionospheric scintillation is one of the most challenging problems in Global Navigation Satellite Systems (GNSS) positioning and navigation. Scintillation occurrence can not only lead to an increase in the probability of losing the GNSS signal lock but also reduce the precision of the pseudorange and carrier phase measurements, thus leading to positioning accuracy degradation. Statistical models developed to estimate the probability of loss of lock and Geometric Dilution of Precision normalized 3D positioning errors as a function of scintillation levels are presented. The models were developed following the statistical approach of nonlinear regression on data recorded by Ionospheric Scintillation Monitoring Receivers operational at high and low latitudes. The validation of the probability of loss of lock models indicated average correlation coefficient values above 0.7 and average Root Mean Squared Error (RMSE) values below 0.35. The validation of the positioning error models indicated average RMSE values below 10 cm. The good performance of the developed models indicates that these can provide GNSS users with information on the satellite loss of lock probability and the error in the 3D position under scintillation.