Robust GNSS phase tracking in case of slow dynamics using variational Bayes inference

For a precise GNSS (Global Navigation Satellite System) positioning, carrier phase measurements are required. However, cycle slipping in classical phase locked loop (PLL) can lead to a local or permanent loss of lock. To address this problem, we propose a robust nonlinear filter for carrier phase tracking based on Variational Bayes (VB) inference. So far, the algorithm is designed only for slow phase dynamics (i.e., first order loop). Interestingly, the estimator update equation can be expressed in closed form. Performance of our algorithm is assessed on synthetic and experimental GNSS data and compared to that of conventional PLL-based techniques. Results show that the proposed method brings significant improvement in terms of cycle slipping.