Victor S. Zhang

Characterizing the performance of GPS disciplined oscillators with respect to UTC(NIT)

Global Positioning System Disciplined Oscillators (GPSDOs) are now the primary standard of time and frequency at many laboratories and calibration facilities. They are typically accepted as self-calibrating standards, and their users generally assume that they meet the manufacturers specifications. To gain a better understanding of the actual performance of GPSDOs, this paper presents a method of characterizing both their long and short-term performance that uses the UTC(NIST) time scale as a reference. It then describes how this method is used to characterize four GPSDOs, including two that use an oven controlled quartz oscillator (OCXO) as their time base, and two equipped with a rubidium oscillator. All four devices were simultaneously tested using the same antenna over two 60 d measurement intervals. During the first 60 d measurement, a previously surveyed antenna position was used and the same coordinates were applied to all four devices. During the second 60 d measurement, each GPSDO performed an independent survey of the antennas position and applied its own coordinates. Both the timing output (1 pulse per second) and the frequency output (10 MHz) of each GPSDO was measured during both 60 d intervals. A low-noise dual mixer time difference system was used to characterize the short-term frequency stability of each devices 10 MHz output, and all measurement results are presented and summarized.

Effects of antenna cables on GPS timing receivers

We report efforts to minimize the effect of reflected signals in GPS antenna cables on the delay instability of GPS common-view time transfer receivers. The delay instability of interest is at long term time intervals out to about one year. Our measurements of various cables indicate a range of power for multi-path signals from the cables arriving at the receiver. For NBS-type receivers with traditional cables we find significant variations of the reflected power with frequency changes around the 75 MHz IF. The levels range from -16 dBc to -27 dBc. For a commercial receiver with the signal transmitted from the antenna to the receiver at the L-band frequency, the best result is for a special cable system we have built with a reflection loss of -30 dBc. With new cabling systems, any effects due to reflected signals in the cables for either the NBS-type or the commercial receivers should be below 1 ns. We include several weeks of data from receivers using these cables. A full evaluation of the results will require many months, and perhaps years, of data.

Long-term uncertainty in time transfer using GPS and TWSTFT techniques

The techniques of GPS time and frequency transfer (code based and carrier phase) and TWSTFT are widely used in remote clock comparison and in the computation of TAI and UTC. Many timing laboratories in the world utilize both techniques (GPS and TWSTFT transfer links) to compare each others clocks. A time link must be calibrated to assure the time transfer accuracy. In many cases, calibration campaigns have been very infrequent due to the expense and lack of suitable equipment. In lieu of repeated calibrations, some information regarding the long-term stability of these links can be obtained through comparisons between the two links (a so called double difference). Without frequent calibrations it is impossible to tell where the instabilities originate, but information regarding the magnitude of the instabilities can be obtained from double difference data. We have been investigating the combined variations of GPS and TWSTFT links for a number of laboratory pairs, including both long and short baselines. Our results show that the relative change between GPS and TWSTFT transfer links can be as large as 6 to 7 ns over a few years and that all of the laboratory pairs that have been investigated show similar magnitudes in the double difference data. Currently the longest set of good double difference data is about 7 years. The study results point out the need for frequent calibration campaigns if accuracies at the nanosecond level are required.

Coordinating GPS calibrations among NIST, NRL, USNO, PTB, and OP

Reviewing calibration results over the history since early 1980s among several labs shows very mixed results. The best stabilities of GPS receivers, as given by calibrations, are of the order of a few nanoseconds or better over a year, though many results are quite a bit worse. Absolute calibrations show similar potential, though there are problems. We conclude that more calibrations and standard methods are needed.

Ionospheric models and measurements for common-view time transfer

We discuss technologies for estimating ionospheric delays in Global Positioning System (GPS) common-view time transfer. Such technologies include the broadcast Klobuchar ionospheric model, the use of ionospheric maps from the International GPS Service (IGS), codeless techniques for the measurement of the ionospheric delay, and using the pseudo-random code on two frequencies. We find that the code receivers we use in this study are excellent references for comparing other ionospheric measurements. Receivers using the Klobuchar model exhibit some with correct values, others with problems. A receiver using a codeless technique appears to have problems. We also find that, because of potential uncertainties in calibrating code receivers, the IGS maps are particularly useful for time transfer.

Multi-channel GPS/GLONASS common-view between NIST and USNO

Both National Institute of Standards and Technology (NIST) and United States Naval Observatory (USNO) now operate a commercial 12-channel common-view receiver. The receiver tracks the GPS (Global Positioning System) satellites and the GLONASS (Global Navigation Satellite System) satellites. In this paper, we evaluate the receivers performance using the common-clock, common-view calibration data. We also study the performance of multichannel GPS/GLONASS common-view between NIST and USNO. The common-clock, common-views using multichannel GLONASS C/A (Coarse/Acquisition) code and single-channel GLONASS P-code (precise code) contained very little transfer noise. A diurnal effect was noticed in the multi-channel GPS C/A code common-clock, common-view calibration. For the common-view comparison between NIST and USNO, the increased number of daily common-view tracks and the more precise measurements with GLONASS P-code reduced the transfer noise in short term. The multi-channel GPS/GLONASS common-view was influenced by some periodic systematic errors in long term.

The 2015 TWSTFT calibration for UTC and related time links

Two-Way Satellite Time and Frequency Transfer (TWSTFT or TW) is one of the primary time transfer techniques for UTC generation. In this framework the Triangle Closure Calibration (TCC) method has been used to calibrate links between certain laboratories whose links to the pivot lab are directly calibrated. TCC is based upon requiring the closure of three co-joined links to be zero. In this paper, the results of the 2015 calibration computation are presented. The uncertainties are usually below 2 ns.

Transatlantic 2.5 MChip/s two-way satellite time and frequency transfer with surface acoustic wave filters

From August 2010 to April 2011, NIST and OP conducted an experiment of using surface acoustic wave (SAW) filters in transatlantic two-way satellite time and frequency transfer (TWSTFT) with 2.5 MChip/s pseudo-random noise (PRN) codes. The SAW filters used in the experiment are 2.5 MHz band-pass filters with a center frequency of 70 MHz. Instead of using 3.5 MHz bandwidth required for the 2.5 MChip/s signal, the use of SAW filters allows the 2.5 MChip/s TWSTFT to use only 2.5 MHz bandwidth on a satellite transponder. We evaluated the SAW filters. We compared the time transfer instability of the filtered 2.5 MChip/s TWSTFT to the 1 MChip/s TWSTFT with and without the SAW filters. This paper presents the experiment results.

A study on reducing the diurnal in the Europe-to-Europe TWSTFT links

Most of the Two-Way Satellite Time and Frequency Transfer (TWSTFT or TW) links exhibit a daily variation (diurnal) on the order of 1 ns in the differences. The stability of TW is degraded by the diurnal. Many studies on the sources of diurnal have been carried out, but no dominating cause of the diurnal has been found. In this study, we examine the diurnal in several Europe-to-Europe TW links and report that the diurnal and the short-term transfer noise can be reduced by using the triangle difference of the transatlantic TW. We will also analyze where the improvement comes from.

A Low-Cost Time Transfer Receiver for Contributions to Coordinated Universal Time

This paper describes a low-cost time transfer receiver that allows timing laboratories, including national metrology institutes and other designated institutions, to contribute data to the computation of Coordinated Universal Time (UTC). The time transfer receiver compares a laboratory’s local realization of UTC, to signals broadcast by Global Positioning System (GPS) satellites. It stores the measurement results in a format compatible with international standards, and sends data via the Internet to the Bureau International des Poids et Mesures (BIPM) for inclusion in the UTC computation. In addition to being inexpensive, the receiver was designed to be easy to use, allowing recently established timing laboratories to begin contributing to UTC with a minimal investment in training.