|
Cellular forums Home > Archive > Garmin GPS > February 2008 > Estimate the accuracy of your GPS
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Estimate the accuracy of your GPS
|
|
| Rupert Brooke 2008-02-14, 10:33 am |
| Estimate the accuracy of your GPS receiver by following these four steps. A
set of measurements is worth a thousand expert opinions! Forget about EPEs!
Trust your own plot.
1. Find any convenient unobstructed place.
2. Record the UTM coordinates for that place. Don't throw out ANY data
points!
3. Make a graphic plot of Easting's and Northing's (pencil and paper works
really well for this).
Repeat steps 2,3 (at different times of day and night... the more the
better)
© Copyright 2007 - Samuel J. Wormley
Updated 08/11/2007 11:03:00 by swormley1@mchsi.com
Easting's and Northing's (info for laymen)
.Instead of degrees of latitude or longitude UTM uses easting's and
nothings. (info for laymen)
. Easting's are given as the number of meter east of an artificial
reference meridian (0East).
. The central meridian of the zone is always 500,000East.
. Nothings are the number of meters north of the equator.
| |
| Dale Atkin 2008-02-16, 4:33 am |
|
"Rupert Brooke" <yl617@yahoo.com> wrote in message
news:4KZsj.539$dh.236@trnddc01...
> Estimate the accuracy of your GPS receiver by following these four steps.
This will only estimate precision, not accuracy. Precision being essentially
internal consistency of the data, accuracy being how well the data
corresponds to the real values.
If a GPS is consistently reading 100m north of the 'real' location, it
doesn't matter how many times you measure the location, you'll still get
100m north of the real location, and have no way of knowing.
The only way to measure accuracy is with a known benchmark location.
Dale
| |
| nhopton@gmail.com 2008-02-20, 4:33 am |
| On 16 Feb, 06:14, "Dale Atkin" <labrad...@ibycus.com> wrote:
> ...The only way to measure accuracy is with a known benchmark location.
The problem being, of course, that any results you get by measuring
'accuracy' in this way are only good for that time and that place.
Because you get an accuracy of, say, 3 metres on this occasion it
doesn't follow that this is what you'll get if you repeat the
exercise, the variability of the system is itself variable. It's the
most serious problem with GPS, one can *never* know how much
confidence can be placed in what it tells you.
Regards, Nick.
| |
| Dale Atkin 2008-02-20, 10:33 pm |
|
<nhopton@gmail.com> wrote in message
news:5df54f68-c3fb-4d7f-8d5a- 29784155fd8b@q70g200
0hsb.googlegroups.com...
> On 16 Feb, 06:14, "Dale Atkin" <labrad...@ibycus.com> wrote:
>
>
> The problem being, of course, that any results you get by measuring
> 'accuracy' in this way are only good for that time and that place.
You're missing out that most GPSrs also give you a measure of what they
think the accuracy is. Comparing these accuracy measures to the real
coordinates, should give you an estimate of the real accuracy (the GPSrs
estimate of accuracy isn't going to be totally arbitrary, so you should,
given time patitience and some statistics, be able to determine a rough
conversion between the accuracy displayed and the 'real' accuracy). I'm
betting though roughly doubling the error reported by the GPS will give you
about the 'right' estimate of error.
Dale
| |
|
| > You're missing out that most GPSrs also give you a measure of what they
> think the accuracy is. Comparing these accuracy measures to the real
> coordinates, should give you an estimate of the real accuracy (the GPSrs
> estimate of accuracy isn't going to be totally arbitrary, so you should,
> given time patitience and some statistics, be able to determine a rough
> conversion between the accuracy displayed and the 'real' accuracy). I'm
> betting though roughly doubling the error reported by the GPS will give you
> about the 'right' estimate of error.
I'd like to think you were right, Dale. Just to try and make sense of
what the 'accuracy' values that receivers report, this is (I think) an
indication of how tightly data from the receiver groups around the
mean. As you say above, it's an estimate of precision. Also, as you
mention, accuracy is something else. The instrument is capable of
giving you some sort of indication of precision, but not of accuracy,
and I don't think it's necessarily the case that you can draw hard and
fast conclusions about accuracy from precision.
We all have fun pushing the limits when we try to use cheap, hand-held
GPS receivers for Gonzo surveying work and long may it continue, but
as I said above, *the* big problem is that you can never know what
confidence can be placed in what they tell you.
Regards, Nick.
| |
| Tom H. 2008-02-23, 4:33 am |
|
"Nick" <nhopton@gmail.com> wrote in message
news:1814a1ad-b87b-43d4-88fd- 504730e02400@h25g200
0hsf.googlegroups.com...
>
> I'd like to think you were right, Dale. Just to try and make sense of
> what the 'accuracy' values that receivers report, this is (I think) an
> indication of how tightly data from the receiver groups around the
> mean.
From my reading on this issue, the accuracy displayed by the gpsr is related
to the satellite geometry. If the satellites are closely bunched for
example, the accuracy will be degraded. There used to be (maybe still
exists) a website where you could adjust satellite position and see the
resulting DOP (dilution of precision). One interesting thing I found was
that if you are getting signals from only four satellites, each is 45° above
the horizon, and they are each 90° apart in azimuth, the postion finding
equations blow up and there is no solution. That probably would never
happen in real life, but it is interesting as a non-intuitive result.
Tom
|
|
|
|
|