How do GPS Satellites work exactly?

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
5 replies · 2K views
James_Space
Messages
2
Reaction score
0
How do GPS Satellites work exactly, with emphasis on exactly.

I know that GPS satellites use electromagnetic waves to send out a signal of their current location and time for GPS devices to pick up. Using at least 3 GPS satellites, GPS devices can pin point their location by knowing the distance they are at from each satellite. 3 spherical intersections will pin point the location of a satellite down to a single point on a 2D surface.

What I want to know is how the GPS device knows how far away it is from each satellite. Do they use v=d/t to figure out the distance? That's what I initially thought, but I don't think this is correct. If all you get from a satellite is the time and location at which the signal was sent, then you should be able to compare that to the time that you received the signal. Unfortunately, this doesn't sound plausible to me as this would require almost every device to have an atomic clock that is set very accurately, or the calculations will be far off.

Could anyone explain this to me? It's for a school project I'm working on. Thanks in advance!
 
Physics news on Phys.org
James_Space said:
What I want to know is how the GPS device knows how far away it is from each satellite. Do they use v=d/t to figure out the distance? That's what I initially thought, but I don't think this is correct. If all you get from a satellite is the time and location at which the signal was sent, then you should be able to compare that to the time that you received the signal. Unfortunately, this doesn't sound plausible to me as this would require almost every device to have an atomic clock that is set very accurately, or the calculations will be far off.

Could anyone explain this to me? It's for a school project I'm working on. Thanks in advance!
Welcome to PF!

A GPS reciever doesn't need to calculate the distance to each satellite directly. It uses the difference in the time signals to calculate where it is between them. For a 1 dimensional example, consider if you were directly between two objects and each sent you a signal at the same time. The difference between the signals can be used to calculate how far you are from exactly in between them since if you were exactly between them, you'd receive the signals at exactly the same time.
 
russ_watters said:
Welcome to PF!

A GPS reciever doesn't need to calculate the distance to each satellite directly. It uses the difference in the time signals to calculate where it is between them. For a 1 dimensional example, consider if you were directly between two objects and each sent you a signal at the same time. The difference between the signals can be used to calculate how far you are from exactly in between them since if you were exactly between them, you'd receive the signals at exactly the same time.

Thank you for the quick response, and for welcoming me to the forums.

So you're saying that when a GPS device receives those three multiple signals, it simply compares the time at which they have been sent? Or the time that they have been received? Or are these two negligible since they travel at the speed of light?

So let's consider your 1D example. You have GPS device x, and Satellite A and B. x is somewhere in between A and B. If both A & B send out their location at 12:00 at noon, how would you calculate the location of x? Because I'm assuming x does not have an atomic clock on board, so comparing the times at which the signals were received are probably very small differences not detected by a normal clock on your ordinary smartphone.

If it's no problem, could you go in more depth on how this works. I'm a high school student btw, so I won't be able to understand things that are too complex. So if possible, simple terminology please. Thank you once again.
 
Let me briefly add, that while the basic principle behind GPS [1,2] is simple the GPS signal structure [3] broadcast by the satellites is not. Among other things these signals contains the actual information needed by the receiver to calculate the position of all the satellites as a function of time. Only once the receiver has this information it will be able to calculate its own position from the measured pseudoranges [4]. To speed things up, a receiver can use Assisted GPS [5] to "cheat" a bit and load the satellite positioning information from a source other than the satellites.

[1] http://en.wikipedia.org/wiki/Global_Positioning_System
[2] http://www.montana.edu/gps/understd.html
[3] http://en.wikipedia.org/wiki/GPS_signals
[4] http://en.wikipedia.org/wiki/GNSS_positioning_calculation
[5] http://en.wikipedia.org/wiki/Assisted_GPS
 
Last edited by a moderator:
James_Space said:
So you're saying that when a GPS device receives those three multiple signals, it simply compares the time at which they have been sent? Or the time that they have been received? Or are these two negligible since they travel at the speed of light?
it is subtly different from the way I described:
The time they were sent is coded into the signal. The receiver compares signals that it receives at the same time, but have different time codes (as opposed to timing the difference between the same signal received at different times). Remember, the satellites are continuously emitting time signals. Sorry for the confusion.

There are a lot of sites out there you can utilize as sources. Just Google "GPS theory".
 
It's probably easier to start in 1-dimension. If you are on a line, between two satellites, and each satellite continually broadcasts the time, do you see how you could use the time difference to determine where you are between them? If not, we should start there.

With two satellites, you get two numbers, which can be converted into x (the 1-d position) and t. In 3D-space, you need four measurements, to convert to x, y,z, and t. A fifth measurement let's you place a single uncertainty on this number, so the system can determine if this measurement is reliable or not. And so on.