RE: [pct-l] The PCT GPS Mapping Project

Carl wrote:
>
 At the time, I
attributed it to one of the following:
* Echos off the canyon walls (is this a possible source of location errors?)
* Wrong settings on GPS (perhaps wrong datum)
* Operator error
>

There are all sorts of possible error sources when using a GPS.  I've listed
a few below, from most likely or serious to least likely/serious.

1.  "Selective availability", or purposeful signal degradation.  The
satellites can actually lie to you, giving you a position that could be off
by as much as 1/4 mile.  Fortunately, SA was turned off last January so the
satellite signals are now much more reliable, though the US military
reserves the right to turn it back on at any time.  With SA off, I can get
accuracy to within +/- 30 feet about 99% of the time, and within +/- 5 feet
about 50% of the time, if I can get a good signal.

2.  Multi-path problems, where you get echoes of the same signal bouncing
from different locations.  The bouncing signal indicates a different
distance than the direct signal does, so the GPS has to decide which signal
to believe.  Sometimes it chooses wrong.

3.  Poor satellite configurations.  Sometimes you can only receive signals
from a few satellites for whatever reason.  Most GPSes require at least four
satellites in order to give you a 3D fix, but the accuracy goes up with even
more satellites.  If you've only got the minimum, accuracy will be poor.  In
addition, the relative positions of the satellites in the sky make a
difference.  If you get four satellites from different parts of the sky,
you'll get a decent fix, but if they're all close together in the same part
of the sky, accuracy will suffer.  Most GPSes will display an "estimated
precision error" or something like that which indicates how reliable the GPS
thinks its reading is.

3.  Mismatching map datums.  There are different ways of calculating
lat/long.  If your GPS is using one datum and your map is using another, the
numbers won't match up.  Most GPSes allow you to select which datum to use.
This is a pretty common error among novices.  Note that (I believe) even the
USGS maps use different datums depending on when they were produced, so you
have to check each map to make sure you're in sync.

4.  Inaccurate maps with local or systemic errors.  Either the map is
incorrect in its placement of a single feature like a river, peak, etc, or
the entire map is accurate relative to itself, but the whole thing is
shifted over so that the lat/long readings are off by a certain fixed amount
everywhere.  Either way, it's simply an error on the part of the USGS.  This
is becoming more rare over time.

5.  Ionospheric interference, where the atmosphere bends the radio signals,
making them take a longer path than they should.  Fortunately, this error
source is usually small, on the order of a couple of feet or less.

These problems are why mapping with a consumer-grade GPS is harder than it
sounds.  Fortunately, SA is no longer an issue, but multi-path and bad
satellite configurations are still difficult to deal with in mountainous
areas.  That's why whenever you take a waypoint reading, you have to spend
time fiddling with things to make sure you're getting a good reading.  It's
certainly doable, but it takes time and patience.  Survey-quality equipment
has features that can overcome all of these problems, but it's heavy and
expensive.

*** Free Addendum - How GPS Works, Grievously Oversimplified ***

A lot of people are surprised when they learn how GPS actually works.  The
signals broadcast by the satellites don't directly tell you where you are.
They don't contain any distance or location information for the GPS
receiver; all they broadcast is an extremely accurate time signal.  Each
satellite broadcasts each clock tick at exactly the same time as all the
other satellites.  That's the primary device the GPS uses to determine
position.

The satellites also broadcast information about where each satellite is.
(Not where the receiver is!)  When you first turn on your GPS and it takes
several seconds to lock onto the satellites, part of what it's doing is
downloading the precise satellite position information.

Since radio waves travel at a known fixed speed, and since each satellite is
a different distance away from your GPS receiver, the clock ticks from each
satellite arrives at the receiver at slightly different times, even though
they were all broadcast at the exact same instant.  The GPS records the
differences in the arrival times of each signal, which tells it the
difference in distance from each of the satellites.  In other words, based
on the arrival times, the GPS knows that satellite A is the closest (but it
doesn't yet know _how_ close), and satellite B is 2,896 meters farther away
than satellite A, and satellite C is 50,390 meters father away than
satellite B.

So given the satellite positions and the differences of the distances from
them, it goes to work with some seriously heavy-duty geometry and
trigonometry to turn that data into an absolute distance from each
satellite, and from there into a lat/long/alt position.

Once you understand exactly how GPS works, it's easier to understand what
can go wrong with it, and how to fix it.

Eric
* From the PCT-L |  Need help? http://www.backcountry.net/faq.html  *

==============================================================================
To:            <Undisclosed-Recipient:@snipe.prod.itd.earthlink.net;>