[alberg30] GPS Doppler Velocity

[Robert Kirk]

George Dinwiddie wondered:

>So you think the GPS can measure 5 knots based on the doppler shift
>of the relative movement between your boat and the satellite?  I
>don't think so. [...]

It sure does. And the satellites are whizzing around at thousands of miles 
an hour while your boat is doing 5 knots.  Isn't modern technology 
nice?  Your GPS receiver is using a couple of phase lock loops to (1) 
receive the message data, (2) measure the travel time (Pseudo Range), and 
track the phase shift (Pseudo Range Rate or DOPPLER).  All this on up to 12 
satellites at a time. It needs 4 satellites for an x,y,z solution and uses 
any additional ones to improve the solution. It can measure the doppler 
offset to about 1/20 Hertz. It uses the velocity information to update the 
position and the positional changes to help calculate the doppler in a boot 
strappy way. It throws all this into a Kalman filter and outputs position, 
time, & speed. Since your GPS doesn't show negative (vector) speeds, the 
stationary speed won't average out to zero.

Even with SA on, the max probable speed error is less than a knot, and in 
practice is better. The spec for the current Garmins is 1/10 knot RMS.

You can check the GPS speed against wood chips thrown over the side if you 
want real precision.

Cheers,
Bob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alberg30.org/pipermail/public-list-alberg30.org/attachments/20000302/c4c38a1a/attachment-0007.htm>