The Vernotte, et al. paper is at: https://iopscience.iop.org/article/10.1088/0026-1394/38/4/6 On Fri, 2 May 2025 17:03:03 +0200, Javier Serrano via time-nuts <time-nuts@lists.febo.com> wrote: Hello Abdul, François Vernotte is giving a series of four lectures in the FEMTO-ST institute in Besançon (France). Three of them are already online at https://www.youtube.com/@femtost4271. The fourth one will be on "Holdover and long-time extrapolation", and is scheduled in principle for May 22. If you keep an eye on the YouTube channel of FEMTO-ST in the days after the lecture, you should see the recording appear there. Also, an illustrious member of this list recently gave an excellent presentation on holdover ( https://www.ion.org/publications/abstract.cfm?articleID=19958), but I do not think the material is public. If you are comfortable with (lots of) math, here is a classic reference, even if the word "holdover" is not in the title: Vernotte, F., Delporte, J., Brunet, M. and Tournier, T., Metrologia, (2001). Uncertainties of drift coefficients and extrapolation errors: Application to clock error prediction, E1319 Cheers, Javier On Fri, May 2, 2025 at 1:48 PM ABDUL HAI via time-nuts < time-nuts@lists.febo.com> wrote: > Dear all, > > I am reading Time from GPS receiver (from nmea) using FPGA. when GPS is > lost, the time should be updated and continue (I have OCXO) and when GPS > signal returns, my logic should switch back to GPS. > What are available Holdover mechanism available? like.. is there any > standard procedure?. It would be great help if some one suggest me methods > and literature available. Thanks very much in advance. > > > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe send an email to time-nuts-leave@lists.febo.com > _______________________________________________ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe send an email to time-nuts-leave@lists.febo.com  

Hi all, (Demetrios already knows this by heart, I'm just following the thread as the occation presents itself) The Vernotte paper is such a gem, but it was only in the most recent update of IEEE Std 1139 that it was included. François presented it for me an Attila as we visited as guest researchers in the most huble way. He humbly presented it if it would possibly be of any use, while it is one of the most fundamental papers I have encountered in the field of holdover. The work was done for the french space organisation CNES in order to prepare for the GALILEO constellation. The problem is similar to holdover but in technical details can be said to be a bit different. If you have a GNSS satellite with an oscillator onboard, you can measure that while it is in view, but then as it settles below the horizon you will not be able to measure it for the roughly 8 hours it is gone after about 4 hours of measurements. What is the confidence interval of the phase, and how does it change from last calibration/measurement. Depending on noise-type François and colleagues come with polynomial values, using an approach which I did not expect, but appreciated for its robustness in thinking. While very math-heavy for some, it's a recommended read. As result of visiting François and Attila, I wrote a noise simulator that measured how the clock ensemble behaviour of 4 milion clocks change over time (I chose a logarithmic scale). It aligns up to thie Vernotte et al paper. Oh, I should mention what I mean by ensemble behaviour here. Consider that you have a number of clocks, all being completely equal in noise properties. Depending on the details of the noise state of the clock, the phase for each individual clock will deviate away at some path. As we now look at the spread of phase for these clocks, at some time after last calibration / going into holdover, the spread of these will first of all be gaussian. The histogram of any of those non-convergent noise traces will not be gaussian, it will be dominated by non-convergerence, but looking at these clocks together, as an ensamble, they have this behaviour. The statistical spread of these grow with some polynomial depending on the noise-type, which is what the Vernotte paper predicts. We can thus use these to build confidence intervals etc the usual way. David Allan have written a number of papers on the same topic, and these are also reflected in the holdover part of IEEE Std 1139-2022. So what does all this say about a particular clock? Well, this clock ensemble performance predict the behaviour of all possible future variants of this clock, as well as we have characterized it's random noise. OK, so what confidence interval do we use for specifications then? Well, ITU-T have not yet written this down, but in practice they have been using 95% or 2-sigma confidence intervals in the specifications. We are sketching up a contribution to put that into writing. Prof. François Vernotte is a fantastic researcher and teacher. Cheers, Magnus On 2025-05-05 15:49, Demetrios Matsakis via time-nuts wrote: > Flattery works. I’ve attached my ION-PTTI paper, although in truth it is just a minor extension of the Vernotte et al. masterpiece. (ION does not hold the copyright to individual papers, only to them as a whole.) > > >> On May 2, 2025, at 11:03, Javier Serrano via time-nuts <time-nuts@lists.febo.com> wrote: >> >> Hello Abdul, >> >> François Vernotte is giving a series of four lectures in the FEMTO-ST >> institute in Besançon (France). Three of them are already online at >> https://www.youtube.com/@femtost4271. The fourth one will be on "Holdover >> and long-time extrapolation", and is scheduled in principle for May 22. If >> you keep an eye on the YouTube channel of FEMTO-ST in the days after the >> lecture, you should see the recording appear there. Also, an illustrious >> member of this list recently gave an excellent presentation on holdover ( >> https://www.ion.org/publications/abstract.cfm?articleID=19958), but I do >> not think the material is public. >> >> If you are comfortable with (lots of) math, here is a classic reference, >> even if the word "holdover" is not in the title: >> Vernotte, F., Delporte, J., Brunet, M. and Tournier, T., Metrologia, >> (2001). Uncertainties of drift coefficients and extrapolation errors: >> Application to clock error prediction, E1319 >> >> Cheers, >> >> Javier >> >> >> On Fri, May 2, 2025 at 1:48 PM ABDUL HAI via time-nuts < >> time-nuts@lists.febo.com> wrote: >> >>> Dear all, >>> >>> I am reading Time from GPS receiver (from nmea) using FPGA. when GPS is >>> lost, the time should be updated and continue (I have OCXO) and when GPS >>> signal returns, my logic should switch back to GPS. >>> What are available Holdover mechanism available? like.. is there any >>> standard procedure?. It would be great help if some one suggest me methods >>> and literature available. Thanks very much in advance. >>> >>> >>> _______________________________________________ >>> time-nuts mailing list -- time-nuts@lists.febo.com >>> To unsubscribe send an email to time-nuts-leave@lists.febo.com >>> >> _______________________________________________ >> time-nuts mailing list -- time-nuts@lists.febo.com >> To unsubscribe send an email to time-nuts-leave@lists.febo.com > > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe send an email to time-nuts-leave@lists.febo.com