Inconsistent pseudorange biases in time links and their effect on BDS and GPS PPP frequency transfer

Abstract

Abstract Precise point positioning (PPP) using global navigation satellite system (GNSS) phase and code measurements has recently been the primary technique for time and frequency comparisons. Several scholars have studied multi-GNSS PPP clock comparison, but the inconsistent pseudorange bias from receivers with different correlator spaces and front-end designs in pseudorange observations has not been considered. In this work, we analyze the characteristics of inconsistent pseudorange biases of the Global Positioning System (GPS), BeiDou Navigation Satellite System (BDS) and their effects on PPP frequency transfer. The biases are confirmed to exist in receivers from manufacturers Septentrio, Trimble and Leica and differ by manufacturers. To explicitly investigate the pseudorange biases effects of GPS and BDS on PPP time and frequency comparisons, nine stations with receivers from three different manufacturers that can track BDS-2 and BDS-3 signals are selected. Regarding PPP frequency transfer with inhomogeneous receivers, the modified Allan variances (MDEVs) of GPS and BDS PPP frequency stability are significantly optimized. According to the receiver-type classification strategy, the overall improvements of frequency transfer with Trimble-Septentrio and Trimble-Leica are 33%, 7% and 39% and 23%, 14% and 23% for GPS, BDS-2 and BDS-3, respectively. Moreover, the convergence time of clock comparisons is obviously shortened after using the bias corrections. For GOPE-BRUX and GOPE-MATE links, the corrected cases yield average clock difference stability gains of 38%, 35% and 53% and 35%, 35% and 23% for GPS, BDS-2, and BDS-3, respectively. The results show that the bias corrections are vital and allow more stable time links for PPP frequency transfer.