[ntp:questions] $GPGGA fix indicator 6 estimated/dead reckoning

Richard B. Gilbert rgilbert88 at comcast.net
Wed Feb 4 01:45:41 UTC 2009

Dave Hart wrote:
> $GPGGA fix indicator field 6 has at least the following possible
> vaues:
> 0  no fix
> 1  GPS fix
> 2  DGPS fix
> One reference [1] mentioned
> 3  GPS PPS fix
> and several reference
> 6  estimated or dead reckoning
> ntpd/refclock_nmea.c when using $GPGGA tests for 0 and declares the
> clock not in sync.  When I read "estimated" or "dead reckoning" wonder
> if ntpd should consider the clock not of sync when it sees fix
> indicator 6.  The gotcha that comes to mind is what if there are too
> few satellites in view for a good position fix but still enough for
> the time to be stable and accurate?

If ntpd knows where you are a single satellite is all that's needed.

At last reports there were something like 27 Navstar (GPS) satellites 
aloft. Eight or nine of them can be expected to be above the horizon at 
any given time.  In order to be really useful, a satellite should be at 
least ten degrees above the horizon.

When you first set up a GPS timing receiver, you should do a "site 
survey"; e.g. let it run for at least 24 hours and until you have your 
latitude, longitude, and elevation accurately pinpointed.  If you are 
REALLY serious, you might spend as long as thirty days using a least 
squares calculation to get your EXACT location.  The more accurately you 
  know your location, the more accurately you know the time!

> I'm just reading code and inferring from what little I've found
> regarding fix indicator 6.  GPS wisdom/reality check solicited.
> Cheers,
> Dave Hart
> [1]
> http://www.whtechtop.com/accessory/0183.pdf

More information about the questions mailing list