[LEAPSECS] real-time UT1 realisation

[Poul-Henning Kamp]

In message <20110922150945.GC5501 at lake.fysh.org>, Zefram writes:


>But anyway: how do you imagine applying a Kalman filter?


I would use Bulletin A's as input, but I havn't given it more
thought that that.

First thing, I would probably grab all the recent Bul'As 
(ftp://cddis.gsfc.nasa.gov/reports/iers/bulletin_a/volume_xxiv/)
and see how big the difference between the 7 day prediction
and the subsequent rapid is.

Eyeballing it says around a couple of milliseconds, which, unless
you have done things to your hardware, is lost in the NTP noise.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.