Is GPS better at midday than it is at 7am?

G

GHA

Well-Known Member
Joined
26 Jun 2013
Messages
12,552
Location
Hopefully somewhere warm
Visit site
This is interesting, number of visible satellites from a usb GPS dongle on a stationary boat, with a little bit of averaging which makes it easier to see what's going on >

A8X7haS.png


Only couple days data , but seems pretty consistent. 12 satellites visible around midday & 7 at 7am BST.

Fascinating!! :cool:
 
GHA said:
This is interesting, number of visible satellites from a usb GPS dongle on a stationary boat, with a little bit of averaging which makes it easier to see what's going on >

A8X7haS.png


Only couple days data , but seems pretty consistent. 12 satellites visible around midday & 7 at 7am BST.

Fascinating!! :cool:
Click to expand...
I think they take 5 down every night for servicing.
 
You only need three for a fix, but most systems use four and we all know that the sun needs taken down every night for servicing.
 
All those fractional satellites must be a hazard.
It's a good GPS that can track half a satellite.

Is 'number of satellites' actually a useful measure?
Does it correlate to more accurate fixes?
 
TernVI said:
All those fractional satellites must be a hazard.
It's a good GPS that can track half a satellite.
Click to expand...
As said in the first post there's some averaging going on which makes it easier to read.

Is 'number of satellites' actually a useful measure?
Does it correlate to more accurate fixes?
Click to expand...
Dunno - there seems a bit of correlation between horizontal dilation and number of satellites though maybe not massively conclusive though more satellites than 10 looks like the horizontal dilation will be zero. >
FweRHMd.png
 
If you look at the data, You'll see that it's fairly cyclic... Basically there are more satellites above you during the day... This is designed so at high useage times, There are more available for the demand of Gps systems (sat-nav etc etc)....
Regards numbers required...3 will give you a fix... But 4 will lock that fix... then accuracy will improve with every additional satellite added, Most systems will see signal strength and switch between various satellites...
We use this system for track positioning, Generally we hope for a minimum 0f 8 gps locks, So we have 2x gpr triangulation and 2 Gps position check locks..

Regards

Tim
 
The number of satellites is not a factor for accuracy. You simply need 3 minimum for fix and forth one for a positive confirmation. As long as you have this, anything more is redundant. What does affect horizontal accuracy is the height of the satellites used for calculation above horizon. The lower, the better. As this doesn't change over the time of a day, the accuracy doesn't change either. The question is of a rather academic significance anyway, because in general you are getting degree of accuracy more than sufficient virtually all the time.
 
Jiris, Yes you are correct. Accuracy doesn't improve after 3/4 with pos lock... However, In our application... Fast moving motorcycle with high lean angles, We need more spare GPS signals so the internal system can switch quickly if they lose one of the original gps lock satellites
 
Morepower said:
If you look at the data, You'll see that it's fairly cyclic... Basically there are more satellites above you during the day... This is designed so at high useage times, There are more available for the demand of Gps systems (sat-nav etc etc)....
Regards numbers required...3 will give you a fix... But 4 will lock that fix... then accuracy will improve with every additional satellite added, Most systems will see signal strength and switch between various satellites...
We use this system for track positioning, Generally we hope for a minimum 0f 8 gps locks, So we have 2x gpr triangulation and 2 Gps position check locks..

Regards

Tim
Click to expand...
There's the same constellation of satellites up there 24/7.
There are not 'more above you during the day'.
How a fix improves with the number of satellites depends a lot on the receiver.
Your receiver should work its fix on the best satellite signals it receives. Tracking a few extra weak ones probably won't improve that fix. What matters is the quality of the signal from 2 or 3 satellites and where they are in the sky. Ideally you want them spaced around the horizon, so the position lines are at good angles to one another.

Around dawn, atmospheric effects are likely degrading the signals of satellites to your West. That might not affect the quality of your fix, but it might if you've got say buildings or a mountain to your East.

I doubt your receiver is really processing 8 satellites as two groups of four, that would be prone to losing lock with signals from one direction being interrupted.
 
jiris said:
The number of satellites is not a factor for accuracy. You simply need 3 minimum for fix and forth one for a positive confirmation. As long as you have this, anything more is redundant. What does affect horizontal accuracy is the height of the satellites used for calculation above horizon. The lower, the better. As this doesn't change over the time of a day, the accuracy doesn't change either. The question is of a rather academic significance anyway, because in general you are getting degree of accuracy more than sufficient virtually all the time.
Click to expand...
The lower the better up to a point. A lower satellite makes a better angle to the others, but its signal travels through more atmosphere and suffers more delay uncertainty as well as attenuation.
Actually most receivers would fix on two satellites. Back in the old days this would be quite common. But these days you are unlikely to receive only two usable signals, if the other dozen satellites which should be in view are not seen, then the signals from the two are likely very poor quality too.

The other tracked satellites are not really redundant, their signals can be used to improve the correlation of the others, improving accuracy.
 
dont more sats improve vertical accuracy (greatly indifferent in a boat)?
do we have any idea how mainstream systems use more sats (or how many?) I mean Raymarine, Garmin, etc

V.
 
4 give you RAIM (receiver autonomous integrity monitoring) so the receiver itself can work out bad data and disregard it. However if your receiver is SBAS i.e. it also uses EGNOS/WAAS then it doesn't need to use RAIM as the SBAS signal contains an integrity code for each PRN along with a correction value to reduce ionospheric delay.
The values above of 1/2 a meter and 1 meter are not correct. 13m for raw GPS and 3m for d-gps are a better average.
 
TernVI said:
The lower the better up to a point. A lower satellite makes a better angle to the others, but its signal travels through more atmosphere and suffers more delay uncertainty as well as attenuation.
Actually most receivers would fix on two satellites. Back in the old days this would be quite common. But these days you are unlikely to receive only two usable signals, if the other dozen satellites which should be in view are not seen, then the signals from the two are likely very poor quality too.

The other tracked satellites are not really redundant, their signals can be used to improve the correlation of the others, improving accuracy.
Click to expand...
Yes +1 to both your comments, rather like astro fixes the elevation of the star and the angles of the position lines are very important.
 
And where the satellites are at any given time >
Live World Map of Satellite Positions

IMHO anyone who doesn't have an occasional breathless sense of awe at how a tiny little cheap computer chip running on next to no power can listen to radio waves from satellites and turn the data into an almost exact position on the planet is missing out on some very pleasant emotions :)
Plus time goes a bit quicker and slower for the sats because of the gravity of earth and the speed the satellites are going. Just astounding! Just like Einstein said! :cool:
 
A range gives a position line in 2D.
For sure it's really a bit of a sphere, just as in astro position lines are short chunks of circle.
When you only allow 'fixes' to be solutions that are somewhere near the earth's surface and aren't doing hundreds of mph, it's a lot easier....
 
You must log in or register to reply here.

Share this page

Top