[aprssig] How gps works

Stephen H. Smith wa8lmf2 at aol.com
Sun May 6 02:22:23 EDT 2012


On 5/6/2012 1:44 AM, Andrew Rich wrote:
> Cool
>
> And some other thoughts
>
> Radio waves travel at different speeds in a vaccum to air as we know it.

The difference between air and vacuum is very minor.   The diffference between 
air/vacuum and anything else is major.

>
> I am sure what they did was get a known point and work out what the system is 
> telling them and adjust the satellites to send the correct signals to give 
> the correct answer.
>
> //
>
> Does that mean each GPS has a satellite tracking program running inside ?

If by "GPS" you mean the satellites, no.

If by "GPS" you mean the gadget on the ground, YES.  It's a multi-channel 
spread-sprectrum receiver at 1575 MHz, a bunch of fancy timing logic and a 
fairly powerful CPU.

>
> If you loose your battery or move alot, you will need to spend 12 minutes 
> downloading your "elements" again ?

Sometime more than twelve minutes.     If you move a GPS receiver more than 300 
miles or so while it is off, it can stumble around for quite a while trying to 
figure out where it is.   Older units are far worse in this respect than ones 
made in the last 10 years or so.

Since all the satellites broadcast all the almanac and ephemeris information, 
from a cold start, the receiver just blindly starts trying all the de-spreading 
codes until it starts getting a data stream.    (The orbtital mechanics of the 
24-satellite constellation is such that at any given point on the earth's 
surface, you will always be able to receive two or three satellites.)

As it starts downloading the data for all the satellites, the receiver is also 
looking for other receivable signals.   Once it's got all the data for all the 
satellites (or at least all the ones currently in view), and one is designated 
the reference to which all the others are compared, it can start solving 
simultaneous trigonometric equations to determine it's location.

The almanac and ephemeris are normally stored in battery-backed static RAM when 
the GPS is turned off. If the data is not too "stale" when the unit is turned 
back on in the same place (say within 6-12 hours), the GPS will usually re-lock 
nearly instantly.  The satellite orbits shift significantly over a fairly short 
time, due to the interaction of solar and lunar gravity and the influence of 
the solar wind.  so data more than a few days old is nearly useless.  .  
Because of this, some devices now use a super-capacitor instead of batteries to 
back the CMOS RAM since there is no value in retaining the data more than a few 
days.   The receiver will have to go through the cold-start drill again anyway.

About a decade ago, Radio Shack offered a cheap faceless GPS unit for use with 
laptops called the "DigiTraveller".   This unit which looked like a small 
flying saucer sitting on top of your car's dashboard was notorious for slow 
acquisition times.  It turned out that a) it had no memory backup - every start 
was a cold start, and b) the unit was made by Sony, and was cold-starting 
assuming it was in Tokyo!  It would stumble around for 5-10 mins sometimes 
before it figured out it was in North America instead!

Many US-made (or at least designed) units cold-start at the geographic center 
of the continental US, which ironically is very close to Garmin's world 
headquarters.      The early Delorme Earthmates (intended to be used with 
laptops running the Delorme Street Atlas program) allowed you to assist the 
cold start by clicking on your approximate location on a map of the US (or 
specify what state you were in from a pull-down list) which would push 
approximate starting coordinates into the unit to speed the startup.

>
> Cold start ?
>
>





More information about the aprssig mailing list