its long and there is some maths, but this is a great explainer.

GPS – Bartosz Ciechanowski

Posted 4 months ago

I used to work in a department which designed some of the first GPS receivers, they were the size of a BBC Micro – before everything was integrated in a single chip.

Posted 4 months ago

Just you and me then…

Posted 4 months ago

Also old enough to remember them turning off selective availability – your position used to wander around by 100-200m whilst they had that switched on…

Posted 4 months ago

#metoo :o)

That’s a fascinating read – Thank You!

Posted 4 months ago

I recall the run up to the Gulf Wars – for about 3 months before them GPS devices seemed to give 1m accuracy – those using them were all guessing something was going to be happening. Before then it seems to be between 20m and 30m accuracy.

I’ll get a read of that tonight whilst waiting for Rugby training to end, thanks for the link.

Posted 4 months ago

Also old enough to remember them turning off selective availability – your position used to wander around by 100-200m whilst they had that switched on…

gulf war pt 1 caused that when they realised how many consumer grade GPS’s were in use by the military.

Posted 4 months ago

Stephen Fry explains GPS on QI

Posted 4 months ago

First, GPS. How does that work, Stephen?

“You send a signal from your GPS device,” he explained. “You’ve got to be at least three, usually four or five satellites – that receive your signal. And the difference in time it takes to get from one satellite to the other to the other, which is milliseconds, allows them to calculate your position to within 10 metres.”

That’s amazing.

close, but wrong.

Posted 4 months ago

I’ve not read the article but I always thought that it was the other way round – your gps is a ‘receiver’ and its the satellite that continuously sends out the signals – the maths is basically the same though.

(your phone, usb gps, etc. would be doing very well to transmit a signal to multiple satellites!)

Posted 4 months ago

i was about right.
What a great website.

Posted 4 months ago

They got more accurate around the mid-90’s because the US government allowed public use of their military satellites. Previous to that I think they were worried about nefarious use.

Posted 4 months ago

In the end the accuracy depends on the accuracy of timing, as it’s a comparison of the time stamp of the signal vs the time on your receiver that determines how long it took the signal to reach your receiver that tells you the distance. And given the signal travels very fast, the time differences are very small.

Fortunately we have clocks that are accurate to seconds in hundreds of millions of years, but increasing accuracy is vital….. knowing there’s a turn coming up to uncertainties of 10m or 1m is workable when on a bike ride, but when autonomous vehicles are a regular thing that 1m is not good if they’re overlapping by it.

British Atomic Clock Leads Race for Accuracy

https://www.imeche.org/news/news-article/shrinking-quantum-clocks-fill-growing-need-for-accurate-timekeeping

Posted 4 months ago

They got more accurate around the mid-90’s because the US government allowed public use of their military satellites. Previous to that I think they were worried about nefarious use.

Nope.

There are two GPS systems, commercial and military. The same satellites run both. The military system uses longer codes which give better spacial resolution as you can discern time more accurately. The codes necessary to use the military system change regularly, so you need to be updated with the new codes every few days (can’t remember how often they change them). The commerical system just uses a set of known codes published in the spec, so anyone can use them.

They then corrupted the commerical system with selective availability – made it wander around slowly by about 100-200m. As they didn’t want a Russian cruise missile hitting the White House guided in by a US DoD system.

The idea was companies would make commercial GPS receivers for the masses and military ones for the DoD. No one was really bothered about the latter and the big players jumped on the commerical receivers.

As mentioned above, come the 1st Gulf war they didn’t have enough military grade GPS receivers and had to issue the troops with commerical ones.

They then decided to switch off selective availability to improve accuracy of the commerical receivers (poss related to the Gulf war, can’t remember relative timing).

The advent of Glonass (Russia’s own system) made selective availability pretty pointless as you could just guide your cruise missile to 1600 Pennsylvania Avenue using that. So, selective availability never got switched back on.

The higher resolution military system still exists and I suspect that’s what US cruise missiles use amongst other things.

In theory they could switch back on Selective availability, but given how many systems rely on GPS, it would probably cause all sorts of safety critical problems!

Posted 4 months ago

The wide availability of DGPS and now RTK makes selective availability redundant. Plus RTK gives decent height info as well. Prices are crashing down too, I can get cm accuracy for ~£600

Posted 4 months ago

That’s a good website, OP, thanks for posting.

I’ve been interested in GPS from when it was new. I got my first handheld back in the 1990’s, a Garmin GPS 40. It was rubbish. I remember when I first got it, putting the 4 AA batteries in it and standing in the garden waiting for it to work. 20 minutes later it got a fix, shortly followed by the batteries running out.

Posted 4 months ago

It still amazes me how amazingly amazing GPS is. I’ve been at sea in various ships from before the time that we had our hands on GPS. First in the RN and later on into the commercial world.

BITD… We used Transit SatNav and learned our Astro Nav (sextants and stuff), giving us a 1 mile resolution on a good day, which is fine when you’re 100’s of mile out at sea. Transit gave us fixes at about 1 hour intervals too, whoop! Astro Nav, really only gives you a good fix at sunset and sunrise, but heck, all you need is an accurate clock (1 sec will do), a set of tables and a good sun spanner (and some cunning maths).

The first GPS we used were just commercial units, and we were only supposed to treat them as a back-up. But still, 200m on a bad day, and 24 hr coverage, fab!

Gradually we got better and better. cheaper and cheaper, GPS units.

Now I work on ships in the Oil and Gas Offshore world. With the differentially corrected systems, accuracy worse than 10cm is frowned upon… And it’s still getting better.

Posted 4 months ago

that’s a really interesting read, I’m looking forward to reading more of the posts on the site

Posted 4 months ago

Great article OP. I had a quick look, will read properly tomorrow.

Posted 4 months ago

@jam-bo can you tell me more about how one goes about getting cm accuracy for £600? Thanks!

Posted 3 months ago