Improved GNSS receiver bias estimation using a neural-network based total electron content model

Abstract

With the availability of fast computing machines, as well as the advancement of machine learning techniques and Big Data algorithms, the development of a more sophisticated total electron content (TEC) model featuring large scale ionospheric irregularities and anomalies is possible. We recently developed a fully connected neural network model trained with Global Ionospheric Maps (GIMs) data from the last two solar cycles. The model can successfully reconstruct ionospheric features that are not always visible such as Nighttime Winter Anomaly (NWA) which is only visible in the Northern Hemisphere at the American sector and in the Southern Hemisphere at the Asian longitude sector during low solar activity, winter and local night-time conditions. The NN based TEC model inherits also other features such as the distribution of Mid-latitude Ionospheric Trough (MIT) and the longitudinal variation of the Equatorial Ionization Anomaly (EIA) features. Being motivated from the performance of the NN based TEC model in ionosphere reconstruction we applied the model for differential code bias (DCB) estimation for a network of ground GNSS receivers. The investigation shows that the receiver DCBs can be accurately computed by the NN-based TEC model. The obtained accuracies are comparable to those obtained by the conventional method of DCB estimation by fitting GNSS TEC data to the ionospheric basis function represented by spherical harmonics or other approaches. It is expected that the application of NN based TEC model for GNSS receiver bias estimation will simplify the operational procedures for near real-time ionosphere monitoring without losing its accuracy.