[time-nuts] Relationship between "fixes" and time duration?
mkperrett at gmail.com
Sun May 5 21:28:14 EDT 2013
There are multiple variables that effect the "minimum" number of fixes that
will yield a particular position accuracy (and hence a time accuracy when
using that position as truth).
(1) The ability of the antenna to see the entire sky. If a mountain, tall
building or tree is between you and a satellite that signal may be lost. In
this case your position fix will be degraded due to poor trilateralization
(2) If you are in a place where you get a "bounced" GPS satellite signal
then multipath occurs and will degrade that satellites signal and because
it is a bounce the signal path will be longer, resulting in an increase in
time for the signal to reach the receiver. This results in a time error and
is usually the predominate error when it occurs.
Although repetitive, the resultant geometry changes with time of day. There
are 'good' times and 'bad' times. So if you take a one hour fix (3600
points) then it is the luck of the draw whether you get a good position
fix. There are programs on the net which provide plots of HDOP versus time
(or GDOP, TDOP, PDOP etc). One such product is available from ARINC, Inc.
My recommendation is to do the longest survey you can, with a 48 hour goal,
in the end there is no downside
Michael / K7HIL
On Sun, May 5, 2013 at 11:18 AM, Magnus Danielson <
magnus at rubidium.dyndns.org> wrote:
> On 05/05/2013 04:15 PM, James Robbins wrote:
>> What is the relationship between the T'Bolt survey "fixes" number (e.g.
>> 2000) and number of hours duration of the survey? Is it one "fix" per
>> second or something? What number of fixes should be input to accomplish a
>> 24 hour or a 48 hour survey?
> If there is too few sats to get a fix, then it waits until it has
> sufficients sats for the next fix. So, rather than a fixed time, you get a
> fixed number of fixes, and the reception situation will decide how long
> time it takes to get the same minimum quality fix.
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/**
> and follow the instructions there.
More information about the time-nuts