Dear Bruce I think it will be very difficult to achieve synchronization at the level you want using GNSS signals. The BIPM would only claim 2 ns accuracy for calibration of the delays in a GNSS receiver plus antenna. When you process carrier phase data, you typically see jumps at the day boundary of the order of 1 ns, reflecting uncertainties in the model. There's some reduction in ionospheric noise and systematics because you would be operating almost zero-baseline eg two identical receivers in our lab track each other at the level of 100-200 ps, but again, calibration of absolute delays below 1 ns is hard.
If you really can't use cables then you might consider free space signals like a laser beam or microwaves. There is plenty of literature about free space optical links. One example of a microwave timing link is http://www.locata.com/wp-content/uploads/2015/09/USNO-Wide-Area-Wireless-Network-Synchronization-Using-Locata-ION-Publication-9-20-2015.pdf On Mon, Apr 11, 2016 at 3:00 PM, Bruce Griffiths <[email protected] > wrote: > There is a proposal to use multiple light bucket style optical telescopes > to do Intensity stellar Interferometry over short baselines (up to perhaps > 1km or so) by using independent clocks to time tag photon arrivals. store > the time tags and process the data off line. Depending on the time tag > resolution there is a need to measure the time differences between the > independent clocks to an accuracy in the 1ns to 100ps range. Is there a > better way of doing this other than using geodetic grade GPS receivers > capable of GPS carrier phase measurements?Since the local clock flywheel > oscillators will need to not deviate by more than 100ps or so over the > several minutes required to perform the carrier phase averaging what sort > of clock will be suitable apart from a good rubidium standard with a > cleanup oscillator? > NB Running fibres or coax between the telescopes isnt an option. > > The relative positions of the telescopes has to be known to within a cm or > so for this to work. > Bruce > > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
