On Sun, 18 Jul 2021 15:04:33 -0700 (PDT), Tom Kunich
<
cycl...@gmail.com> wrote:
>On Sunday, July 18, 2021 at 12:42:40 PM UTC-7,
jeff.li...@gmail.com wrote:
>> If GPS birds were in a
>> geosynchronous orbit, where they appear to not be moving in the sky,
>> there would be no doppler shift from which to calculate your location.
Argh. I erred. The above partly wrong. Doppler is used in GPS for
finding velocity, not location. Sorry(tm). With the satellites in
fixed positions in the sky, such as in geosynchronous orbits, we would
still be able to find our location, but have problems calculating
velocity.
>OK, But the doppler shift is just something that needs to be compensated for. In most cases I assume simply by having a wide band receiver.
An over simplified description of how GPS works is by measuring the RF
time of flight between the GPS satellite and the receiver on the
ground. A single time delay forms a circular track on the ground. By
obtaining additional circular ground tracks, a 2D position can be
found by trilateration from 3 satellites. A 3D position requires 4
satellites. The rest of the calculation is compensation for
relativistic effects and atmospheric diffraction. This video covers
this nicely:
"How Does GPS Work?"
<
https://www.youtube.com/watch?v=FU_pY2sTwTA> (4:50)
Doppler comes into the picture when computing velocity. There are
various algorithms, but Doppler is the simplest and most common:
"How does a GNSS receiver estimate velocity"
<
https://insidegnss.com/wp-content/uploads/2018/01/marapr15-SOLUTIONS.pdf>
"Doppler-Aided Positioning"
<
https://www.gpsworld.com/gnss-systemalgorithms-methodsinnovation-doppler-aided-positioning-11601/>
"GPS and other global navigation satellite systems use the Doppler
shift of the received carrier frequencies to determine the velocity of
a moving receiver. Doppler-derived velocity is far more accurate than
that obtained by simply differencing two position estimates."
I don't know what you mean by "simply by having a wide band receiver".
The GPS signal is spread spectrum, which has a wide occupied
bandwidth. There are various signal in the GPS band, which is 24MHz
wide. For a receiver to hear all the various GPS signals, the front
end would have to be 24MHz. Actually, it need to be more than 24MHz
wide so that there are no problems with group delay (variable delays
over the receiver bandwidth) and to handle partly overlapping signals
from other services, such as Galileo. Best shown with a graph:
<
https://www.everythingrf.com/community/gps-frequency-bands>
<
https://www.tallysman.com/gnss-constellations-radio-frequencies-and-signals/>
Yes, it's a mess.