How Does GPS Work?

[Sailor Doug]

How about a quotation from some one who makes them?


Sent from my iPad using Cruisers Sailing Forum

[four winds]

If one googles:

" does GPS use the doppler effect",

three of the first four hits are pdfs on the subject. I have a weak signal here at White City on the ICW or I would download and post.

[colemj]

Originally Posted by Sailor Doug

Gps is quite amazing, at lest three satellites rotating very very high, very fast elliptical orbit. Spinning semi sphere and a boat siting on it in a fluctuating fluid. Device on the boat (gps) is calculating relative direction, speed and position on this rotating sphere. Only a very weak common message is given to all devices trying to solve this problem. It is amazing how accurate and quickly you processor does this. Did I also mention from memory the gps needs to know orbit of satellites, surface of earth at boats position and correct it for surface undulation.

It is even more amazing than that. The satellites are so high and moving so fast, and contain such precise timing devices, that relativistic effects have a huge impact on their error. So engineers use Einstein's equations to pre-build timing corrections into their clocks so that the time one sees down on earth is correct when the signal arrives.

Amazing - all of that was completely figured out by a single person almost 100 years ago before physicist knew of the existence of atoms, and the entire universe was thought to consist only of the milky way (and that was just thought to be half as large as it actually is).

Mark

[Dockhead]

Originally Posted by colemj

It is even more amazing than that. The satellites are so high and moving so fast, and contain such precise timing devices, that relativistic effects have a huge impact on their error. So engineers use Einstein's equations to pre-build timing corrections into their clocks so that the time one sees down on earth is correct when the signal arrives.

Mark

I was staggered by this when I read it a few years ago. Absolutely amazing fact

[transmitterdan]

Originally Posted by Sailor Doug

How about a quotation from some one who makes them?


Sent from my iPad using Cruisers Sailing Forum

Doug,

There is no need for a huge acrimonious debate. Much of the marketing material on GPS is inaccurate. It doesn't matter to 99.9% of the intended audience. If they explained the real internals of GPS 99.9% of readers would have their eyes glaze over in 30 seconds. So let's agree to not use marketing material in an engineering discussion. The reality is that most GPS brands don't actually make the GPS modules. They are made by a handful of far eastern companies. The companies like Apple and Garmin are mainly software companies and they purchase GPS modules from someone else.

But the only thing I hope folks take away from this is that GPS receivers can calculate velocity very fast and very accurately. So fast and so accurately in fact that many trucks and high end cars use GPS speed instead of more traditional cable speedometers.

Accurate speed is a real challenge for slow moving applications like sailing, jogging, walking, and the like. The fact you can get accurate speed at 0.5 knots updated 10 times a second is truly mind boggling.

Fair winds,
Dan

[44'cruisingcat]

I won't claim any expertise, but intuitively, it would seem to make more sense to use position change vs time to calculate speed, rather than using Doppler shift vs the radio signals coming from the satellites.


I mean how much Doppler shift in a radio signal travelling at the speed of light would there be in 0.5 knot of boatspeed? And compared to the relative velocity of the satellites themselves?

[Sailor Doug]

Originally Posted by transmitterdan

Doug,

There is no need for a huge acrimonious debate. Much of the marketing material on GPS is inaccurate. It doesn't matter to 99.9% of the intended audience. If they explained the real internals of GPS 99.9% of readers would have their eyes glaze over in 30 seconds. So let's agree to not use marketing material in an engineering discussion. The reality is that most GPS brands don't actually make the GPS modules. They are made by a handful of far eastern companies. The companies like Apple and Garmin are mainly software companies and they purchase GPS modules from someone else.

But the only thing I hope folks take away from this is that GPS receivers can calculate velocity very fast and very accurately. So fast and so accurately in fact that many trucks and high end cars use GPS speed instead of more traditional cable speedometers.

Accurate speed is a real challenge for slow moving applications like sailing, jogging, walking, and the like. The fact you can get accurate speed at 0.5 knots updated 10 times a second is truly mind boggling.

Fair winds,
Dan

Dan the reason why I ask for manufactures sites is they always try to sell the best technologies. Most of the papers posted on this site refer thesis by phd candidates and are very smart girls and guys. They are giving a lot of proposed new free information away to advance their carers. They are promoting new future technology but not repeating current methods. Companies that know how their systems work are protective of proprietary information.
Doppler shift is a pretty old piece of science. Speed differential without know position is not possible. A triangulation of position of position change is required with Doppler shift assisting in getting accurate timed based position change. (speed)
Atonomas navigation is being advance using multiple antennas, assistance corrections and solid state gyros.
Our conversation needs to be about how consumer marine gps works. Look at the array of antennas and vision sensors on atonamas vehicles we don't have this on our boats.
Our gps gives very accurate speed but a bit of latency is a problems for some of us, and some calculated date also suffers from fractal distortion.



Sent from my iPad using Cruisers Sailing Forum

[dlymn]

Don't forget that the calculation must include the relationship between altitude of the satellite and the force of gravity. If not, the plot could be out by many hundreds of metres.

There is also a lot of recalculation to fine down the triangle of intersection to just a few metres.

I am still amazed at the complexity of the calculation which is all based on the infinitesimal shift in frequency of a signal from a satellite that is moving rapidly some hundreds(?) of kilometres up in the sky

[Sailor Doug]

It is amassing that the time transmitted by the satellites and filter by the Doppler effect can give such amazing results.


Sent from my iPad using Cruisers Sailing Forum

[transmitterdan]

Originally Posted by Sailor Doug

Speed differential without know position is not possible.

Of course it is possible. The radar gun used by your policeman to give you a speeding ticket uses Doppler to measure speed and it has no idea of your position at any specific time. NASA use Doppler shift to measure speed of all sorts of things out in the universe that they can't measure the position of within a few million miles. Yet they can tell us how fast a thing is moving very accurately.

It is well proven that one can measure speed of an object without knowing position. All you need is the frequency of the transmitted radio wave and the frequency observed by the moving object. The difference in frequency tells you the relative speed (i.e. speed differential) between the transmitter and the object precisely.

[Sailor Doug]

How can you possible equate a simple hand gun to a object on earth calculating it's surface from terrestrial satellite. At no time can a liner speed be directly interpreted from a satellite.


Sent from my iPad using Cruisers Sailing Forum

[DeepFrz]

These materials were developed by Peter H. Dana, Department of Geography, University of Texas at Austin, 1994.
Velocity is computed from change in position over time, the SV Doppler frequencies, or both.

You both are partly right.

http://www.colorado.edu/geography/gc...gps/gps_f.html

[socaldmax]

Originally Posted by Wotname

Not really revelant to main discussion but interestly you must be using a different Wikipedia than me

The quoteed part in post 6 was verbatim.

Or perhaps the wiki entry has been ammended in the last few hours!

Anyhow, back to the main discussion....

No, he used the same Wikipedia, bur he's better at cutting and pasting than you, apparently.

Here is your post. Notice the missing word "use." He wasn't disputing you, he was correcting your quote, which is somehow missing a word. This made me question the accuracy of your quote, since a simple cut n paste would not have dropped a word in the middle of the sentence. It honestly made me think it was edited, that's why I looked it up myself.

Wotname

Re: Speed Through GPS Versus Old Fashioned Paddle Log
@ lordgeoff, a short extract from Wikipedia - Global Positioning System: "GPS units can measurements of the Doppler Shift of the signals received to compute velocity accurately".

[socaldmax]

Originally Posted by Dockhead

In another thread, there was a serious debate about how GPS works. I'm not sure why there's such a big debate -- it's not a matter of opinion. But in any case -- here's a place to discuss it. Enjoy.

I hope Goboatingnow will come by and straighten us out -- he's a pro in the field and forgot more yesterday, than the rest of us will ever know about how GPS works.

I understand you're actually trying to establish his credibility in his field, but since you have no way of knowing the extent of anyone else's knowledge of GPS, you really can't speak about that.

You never know, Roger Easton might be a member of cruisers forum.


From what I can see, these guys aren't arguing real differences, but semantics.

To save time, I'm just going to C&P from another website.

In early GPS receivers, four PRs from 4 satellites was converted into a 3-D (XYZ, Lat/Lon/Hgt or whatever) position plus the calibration of the timing bias of your receiver, and 4 PRRs were converted into a 3-D velocity plus a measurement of the frequency error of the oscillator. More modern receivers take all the PR+PRR data from all the N satellites in view for the past T seconds and feeds the 2*N*T PR+PRR samples it into a single mathematical "black box" (BB) (usually a Kalman filter) to produce an over- determined estimate of the same 8 parameters. So in modern receivers, this BB is using both the combination of past & present PRs and PRRs from many satellites to improve the Position, Velocity & Time (PVT) estimate. So the statement about velocities being determined by changes in position is sorta, partially correct, but (when you look at the equations inside the BB), the measured "apparent Doppler" frequencies are even more important.

To be clear, a GPS receiver will not give you a velocity reading just from the Doppler shift of the incoming signal. First it must calculate PVT, then it will use Doppler shift to further refine velocity, so that's why neither side is wrong.

[socaldmax]

Originally Posted by transmitterdan

Of course it is possible. The radar gun used by your policeman to give you a speeding ticket uses Doppler to measure speed and it has no idea of your position at any specific time. NASA use Doppler shift to measure speed of all sorts of things out in the universe that they can't measure the position of within a few million miles. Yet they can tell us how fast a thing is moving very accurately.

It is well proven that one can measure speed of an object without knowing position. All you need is the frequency of the transmitted radio wave and the frequency observed by the moving object. The difference in frequency tells you the relative speed (i.e. speed differential) between the transmitter and the object precisely.

The problem with your analogy is that a stationary radar gun must be held stationary to accurately measure the velocity of the target, movement by the officer can create a huge error.

A mobile mounted radar gun uses calibrated velocity input from the vehicle to add or subtract from the calculated velocity from Doppler shift.

In the case of a GPS mounted on a boat, it doesn't get a speed input from the boat, so the unit really has no idea what it's velocity is in relation to the velocity of the satellite. Yes, Doppler shift can be measured and used to calculate absolute velocity, but again it has no way of knowing how much is satellite velocity and how much is vessel velocity until after it has calculated position, velocity and time from the data in the signal.