GNSS Carrier Tracking via a Variational Bayesian Adaptive Kalman Filter for High Dynamic Conditions

Under high dynamic conditions, a robust carrier tracking approach is essential for global navigation satellite system (GNSS) receivers. In this paper, the powerful Kalman filter (KF) technique is adopted in GNSS carrier tracking. The correlation signals are used as system measurements to discard the discriminator restricted by linear region. Then, a linear measurement equation is established based on the error-state, so that the system model is linear. Hence the KF can be used instead of the nonlinear KF which requires more computational costs. Furthermore, to exploit the potential of the KF under different conditions, an adaptive KF (AKF) based on the variational Bayesian approach is proposed. The proposed filter has accurate and robust estimation performance, especially under high dynamic conditions. Simulation results verify the applicability of the proposed system model and the superiority of the proposed filter compared with the traditional KF and existing AKFs.