To clarify a couple of points:
Quote:
The GPS unit determines your location including altitude, then in a preset time frame it takes another bearing and compares it to the timne that has elapsed. A simple calculation by the unit displays your speed through the given distance.
Sort of. There is a position component to the velocity calculation that is performed, but I’ll get to that in a minute. GPS channels, each assigned to a particular satellite, take two measurements called pseudorange and deltarange (also called delta-pseudorange). You can thing of pseudorange as a measure of position + time, while deltarange is a measure of (position + time) rate over the deltarange summation interval. A GPS cannot directly measure true range, because it cannot account for it’s own internal clock error. This is why to determine true 3D position, a GPS needs 4 measurements (3 degrees of position freedom, plus user clock).
Now, even though pseudorange data contains an unknown quantity (and tends to be very noisy as it is generated from the code-chipping on the signal), deltarange is very accurrately measured by using the GPS carrier, which can give relative accuracy from second to second in the millimeter range. Because the common error (clock rate) in deltarange applies to all measurements, satellite geometry allows excision of this quantity with relatively high accuracy as well (i.e. an SV straight East has v+c, one in the west has -v+c). The relatively low noise on the deltarange measurements allows the navigation filter to really clamp down and respond rapidly to changes in velocity, so you always get a good measure.
The position component that I referred to earlier can affect velocity measurement in that it is used to determine relative geometry of the user to the satellites. The impact of ‘normal’ position errors on velocity is minimal, as it is de-weighted by the cosine of the angle from the errant position to a satellite which is 11,000,000 meters away.
Now:
Quote:
A GPS measures everything in a straight line. I do not believe they are capable of measuring speed up or down in altitude. I belive if you were to drop a GPS off a cliff it would read 0 MPH.
It is true that 90% of the GPS radios that we (on this site) have are set up to measure Speed Over Ground (SOG). This is because almost all of us are using handhelds or marine units, which are expected to be operated in relatively low dynamic environments with little or no Z component in velocity. The unit is constantly computing the Z velocity, but does not output it because the odds of you “going ballistic” in your bassboat (okay, bad example… in your walleye boat??) are pretty minimal. Airborne units output 3-axis velocity data, because to a pilot, this is critical information. Even though GPS is not approved as a primary navigation aid in aerial transport, these units have to go through a rigorous approval process which translates to a lot of cash for the end unit.
All this is to say: They output SOG because when you’re glued by gravity to the lake surface, that’s what you really need to know. This also generates a better solution, as it is not impacted by the poor vertical determination inherent to the GPS system. Horizontally, you’re overdetermined because you’re tracking satellites in every direction. Vertically, all of the signals are coming from above (unless you’ve a satellite buried in the bottom of your boat)!
Whew!! Beer me. 
Hope he’s doing well!
Steve 
