Precise GNSS time and frequency transfer

GPS measurements are used since the eighties for the remote clock comparisons needed for TAI. The technique is based on “Common View”, i.e., the simultaneous observation of the same satellite by two ground stations in order to deduce the synchronization error between the ground clocks and the satellite clock; from that one deduces the synchronization error between the two ground clocks. On a GPS signal, two types of measurements can be performed: the pseudorange (code), but also the phase of the carrier wave. Thanks to the shorter wavelength, the noise level on the GPS phase observable is about 100 times smaller than the corresponding noise level on the code observable. For this reason, the carrier phase measurements have been used for a long time for geodetic applications requiring very high precision, and were later introduced in time transfer studies. When analyzed together with a consistent modeling of the measurements similar to GPS data analysis dedicated to precise positioning, the carrier phases provide a very powerful tool for frequency comparisons between remote clocks, with a precision at the 100 ps level at each epoch. However, since the carrier frequency cycles are not time-tagged, they cannot provide any information about the absolute offset between these clocks. The carrier phase measurements can thus only be used for frequency transfer, i.e. for the determination of the evolution of the phase differences between the clocks. The absolute value of the clock offset will have to be determined using the code information. A combined analysis using both carrier phase and code observations is necessary to get both the absolute value of the clock offset and a very precise determination of the frequency transfer, and the precision of the absolute clock offset determination is still limited by the code noise level (some ns). The tutorial will present the way of introducing code and phase measurements in a combined analysis for precise time and frequency transfer, and the results that can be expected using different approaches.