A Simple and Accurate Procedure for the Absolute Calibration of GNSS Receivers

P. Waller, D. Schultz, R. Prieto-Cerdeira

Abstract:	Among the Laboratories participating to UTC, it is well known that the accuracy of GNSS time transfer strongly depends on the uncertainty with which the hardware delays in the GNSS receiver chain are known. To address this challenge, the BIPM has established guidelines for the relative calibration of GNSS equipment contributing to UTC, based on GPS only [1]. With the advent of new GNSS observables that could potentially increase the accuracy of GNSS time transfer, an accurate and consistent calibration of hardware delays is required. The same is valid for the receivers chains that are operating within the Galileo system, and that shall provide an accurate estimate of the biases between the Galileo System Time and UTC (to determine the GST-UTC conversion coefficients) and the biases between the Galileo System Time and the GPS Time (to determine the GGTO coefficients) [2]. It is therefore essential to have the means to determine these biases in a consistent and accurate manner, with methods that are simple to implement, allowing for both a good control of the uncertainty budget and the possibility to execute regular calibration campaigns with minimum impact on operation. Based on prior seminal work reported in particular by CNES [3] and DLR [4], we have developed a simple procedure for accurate absolute calibration of GNSS receivers. It is simple because it relies on a minimum set of equipment and measurements, it is accurate because we have minimized the number of contributors to the uncertainty budget. The procedure makes use of a GNSS simulator, whose differential delays between the rising code transition of each individual GNSS signal component and the reference 1pps signals are estimated using cross-correlation technique. These calibrated GNSS simulator signals are subsequently fed into the GNSS receiver whose hardware delays are estimated for each signal component in a single and consistent process. Using a GNSS simulator (Spirent GSS9000), we have applied this procedure to one of our reference receivers (a Septentrio PolaRx4TR) and have obtained a consistent set of individual hardware delays for GPS signals (L1 C/A, L1 P and L2 P) and for Galileo (E1, E5a and E5b), with an estimated uncertainty at the ns level. All in all, the whole procedure can be executed in about one hour, thereby limiting the interruption of the operational receiver. The details of the procedure, together with its validation will be presented at the conference, followed by the results of the calibration campaign, including uncertainty budget and temperature sensitivity. An outlook to future work will conclude the presentation.
Published in:	Proceedings of the 48th Annual Precise Time and Time Interval Systems and Applications Meeting January 30 - 2, 2017 Hyatt Regency Monterey Monterey, California
Pages:	104 - 104
Cite this article:	Waller, P., Schultz, D., Prieto-Cerdeira, R., "A Simple and Accurate Procedure for the Absolute Calibration of GNSS Receivers," Proceedings of the 48th Annual Precise Time and Time Interval Systems and Applications Meeting, Monterey, California, January 2017, pp. 104-104. https://doi.org/10.33012/2017.14984
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In