GNSS receiver clock modeling when using high-precision oscillators and its impact on PPP

摘要

Processing data from Global Navigation Satellite Systems (GNSS) always requires time synchronization between transmitter and receiver clocks. Due to the limited stability of the receiver's internal oscillator, the offset of the receiver clock with respect to the system time has to be estimated for every observation epoch or eliminated by processing differences between simultaneous observations. If, in contrast, the internal oscillator of the receiver is replaced by a stable atomic clock one can try to model the receiver clock offset, instead of estimating it on an epoch-by-epoch basis. In view of the progress made in the field of high-precision frequency standards we will investigate the technical requirements for GNSS receiver clock modeling at the carrier phase level and analyze its impact on the precision of the position estimates. If we want to take advantage of the frequency stability provided by a high-performance oscillator in combination with a GNSS receiver we have to ensure that the signal delays inside the receiver hardware remain constant. Therefore, we have analyzed the relative receiver clock offsets for a number of GNSS receivers that derive their frequency reference from a common oscillator. Based on experimental data of an exemplary pair of geodetic receivers we show that the noise from variations of the hardware delays in the receiver electronics does not exceed the receiver clock noise (5 ps RMS) when all environmental effects are carefully controlled. By analyzing the elements of the parameter covariance matrix for a simple case of point positioning, the impact of GNSS receiver clock modeling on kinematic and static solutions is studied. Furthermore, we demonstrate the suitability of a single quadratic polynomial to model a receiver clock that is locked to the frequency of an active hydrogen maser for periods up to 24 h. Based on simulated and real GNSS data it is shown that receiver clock modeling improves the RMS of the height component of a kinematic Precise Point Positioning (PPP) by up to 70%, whereas for the static case the gain is almost negligible.