The History of TWSTT at USNO (1962 to Present)

Early history (1962-1964)

In 1962, scientists from the U.S. Naval Observatory (USNO) Time Service in the United States and the National Physical Laboratory (NPL) in the United Kingdom) got together to use the first active-mode communication satellite, Telstar, to perform the very first transatlantic two-way clock comparisons [1]. The large silvered mylar balloon known as Echo 1 was used during August 1960 for one-way time transfer attempts [2]. The one-way time transfer method is not ideal, since the one-way propagation delays are not known.


Left Image: ECHO 1 First Passive Communication Satellite (NASA Image - GIF - 79 KB)
Right Image: TELSTAR First Active Communication Satellite (NASA Image - GIF - 88 KB) Click on each to enlarge.

In 1965, scientists from USNO got together with researchers from the Radio Research Laboratory (RRL) in Japan to perform the first transpacific clock comparisons using the communication satellite Relay II [3].

RELAY Communication Satellite (NASA Image - GIF - 99 KB)

Compared to today the early signal transfer methods and measurements were primitive. In most of these early experiments, a pulsed signal generated every second, known as a one-pulse-per-second (1pps), was modulated onto the carrier wave of a video signal for transmission. Oscilloscopes, polaroid cameras, and other primitive methods of measurement were employed. Precisions reached during the Telstar and Relay experiments were near the 10-to-100-nanosecond level. The accuracies were near the 100-to-1000-nanosecond level.


Middle Ages (1974-1983)

Ten years passed before the next major effort took place. This effort formed the basis of the current technologies and methods applied to TWSTT. USNO, RRL, and the National Aeronautics and Space Administration (NASA) in the United States performed remote clock comparisons using two-way methods, spread-spectrum, and random-access communication methods using the Applications Technology Satellite 1 (ATS-1). Precision now reached the 1-nanosecond level and accuracies the 10-nanosecond level [4]. The major reasons TWSTT was not implemented as a routine high-precision and accurate time transfer method was the very high cost of the required equipment and custom hardware required. These basic problems were not to be solved for about another decade.

The ATS-1 Communication Satellite with engineer before launch (Unknown credit - GIF - 144 KB)

The next experiments with a significant impact on the development of TWSTT at USNO was a series of measures made using the Communication Technology Satellite (CTS/Hermes) during the late 1970s. The National Research Council (NRC) in Canada, National Bureau of Standards (NBS) in the United States, and USNO performed TWSTT experiments that reached the 0.2-nanosecond-precision levels. Accuracies at the 50 nanosecond level were attained [5,6]. This experiment tested phase-shift-keyed signals using normal COMSAT satellite modems, rather than pseudo-random-noise sequence modems, for transfer of the time ticks.

An experiment between the Deutsche Forschungs und Versuchsanstalt fur Luft und Raumfahrt (DFVLR), COMSAT, USNO and Institut fur Luft und Raumfahrt der Technischen Universitat Berlin (TU) was performed during 1983 using an Intelsat-V satellite [7]. Precisions at the 300-picosecond level were attained in one of the first applications of the Mircrowave Time and Ranging Experiment (MITREX) modem.

CTS/Hermes Communication Satellite (NASA Image - GIF - 72 KB)


Recent History (1987 to Present)

By this time, the cost and physical size of satellite dishes and associated equipment had become small enough that more timing centers and laboratories could invest in the required equipment to apply the information learned during the early R&D efforts into a more automated and operational timing transfer system.

In 1987, routine (three times per week) experiments were begun using commercial Ku-band satellites. These experiments were begun by the National Institute of Standards and Technology (NIST, formerly known as NBS), USNO, and NRC. These routine experiments continue.

In 1993, eight timing centers and laboratories began routine (three times per week) transatlantic TWSTT experiments. Those participating include USNO, NIST, the Technical University of Graz, Austria (TUG), the National Physical Laboratory, United Kingdom (NPL), the NMi van Swinden Laboratory, The Netherlands (VSL), Deutsche Telekom AG, Germany (DTAG), Physikalisch-Technische Bundesandstalt, Germany (PTB), and the Observatoire de la Cote d'Azur, France (OCA) [8,9]. These experiments were restarted during January 1997.

Since 1996, we have used 24/7 a channel on the CONUS satellite located at 103.0W for TWSTT operations. Initially the satellite was GSTAR-1. Currently the satellite is GE-1.

Left image: The A2100 type communications satellite artists view
(Lockheed Martin Missiles & Space - GIF - 99 KB)
Right image: GE-1 an A2100 Hybrid (C- Ku-band) Communications Satellite
(Russ Underwood: Lockheed Martin Missiles & Space - GIF - 125 KB)


The Future

During the next few years, the U.S. Naval Observatory will be implementing a method of characterizing and implementation of real-time corrections for systematic noise contributions originating, for example, from environmental perturbations and the ionosphere. Operational real-time steering of a commercial cesium-beam frequency standard using the linear quadratic Gaussian (LQG)/Kalman filter algorithm via TWSTT is producing long-term (months) syncronization at the 4-nanosecond RMS levels. Work is currently in progress to characterize, automate, and operationally implement the newest generation of time transfer modems, the SAtellite Time and Range Equipment (SATRE). Time deviation errors (TDEV sigmas) are indicating 20 to 30 picoseconds at a sampling time of 900 seconds operationally.


  1. Steele, J. M., Markowitz, W., and Lidback, C. A. (1964), "Telstar Time Synchronization," IEEE Transactions on Instrumentation and Measurement, Vol. IM-13, pp. 164-170.

  2. Jakes, W. C., "Participation of Bell Telephone Laboratories in Project Echo," Bell System Tech. J., Vol. 40, p. 975.

  3. Markowitz, W., Lidback, C. A., Uyeda, H., and Muramatsu, K. (1966), "Clock Synchronization via Relay II Satellite," IEEE Transactions on Instrumentation and Measurement, Vol. IM-16, pp. 177-184.

  4. Saburi, Y., and Yamamoto, M., and Harada, K. (1976), "High-Precision Time Comparison via Satellite and Observed Discrepancy of Synchronization," IEEE Transactions on Instrumentation and Measurement, Vol. IM-25, pp. 473-477.

  5. Costain, C. C., Daams, H., Boulanger, J. S., Hanson, D. W., and Klepczynski, W. J. (1978), "Two-Way Time Transfers between NRC/NBS and NRC/USNO via the Hermes (CTS) Satellite," Proceedings of the 10th Annual Precise Time and Time Interval (PTTI) Applications and Planning Meeting, November 1978, Greenbelt, Maryland, USA, pp. 585-600.

  6. Veenstra, L., Kaiser, J., Costain, C., Klepczynski, W., and Allan, D. (1981), "Frequency and Time Coordination via Satellite," COMSAT Technical Review, Vol. 11, No. 2, pp. 369-402.

  7. Hartl, P., Veenstra, L., Gieschen, N., Mussener, K. -M., Schafer, W., Wende, C.-M., Klepczynski, W., Nau, H.-H., and Stoiber, R. (1984), "Spread Spectrum Time Transfer Experiment via INTELSAT," in Proceedings of the 15th Annual Precise Time and Time Interval (PTTI) Applications and Planning Meeting, 6-8 December 1983, Washington, DC, USA, pp. 331-356.

  8. DeYoung, J. A., Klepczynski, W. J., McKinley, A. D., Powell, W., Mai, P., Hetzel, P., Bauch, A., Davis, J. A., Pearce, P. R., Baumont, F., Claudon, P., Grudler, P., de Jong, G., Kirchner, D., Ressler, H., Soring, A., Hackman, C., and Veenstra, L. (1995), "The 1994 International Transatlantic Two-Way Satellite Time and Frequency Transfer Experiment: Preliminary Results," in Proceedings of the 26th Annual Precise Time and Time Interval (PTTI) Applications and Planning Meeting: 6-8 December 1994, Reston, Virginia, USA, pp. 39-49.

  9. DeYoung, J. A., McKinley, A., Davis, J. A., Hetzel, P., and Bauch, A. (1996), "Some Operational Aspects of the International Two-Way Satellite Time and Frequency Transfer (TWSTFT) Experiment using INTELSAT Satellites at 307 Degrees East," in Proceedings of the 27th Annual Precise Time and Time Interval (PTTI) Applications and Planning Meeting, 29 November-1 December 1995, San Diego, California, USA, pp. 335-345.