A slightly related question, if anyone knows - has phone GPS gotten worse in recent generations? More reliance on local wifi networks or something like that?
I ask because I do a lot of backcountry hiking, camping, and foraging and rely on true GPS-only navigation. My most recent two phones (iphone and pixel) have noticeably worse GPS performance than previous phones, and I even changed OS ecosystems mostly hoping for better GPS, but it didn't help. Maybe I've had bad luck, but two noticeably bad phones in a row seems like it may be a pattern.
And is there any way to find phones with very good GPS performance?
Also, RTK is an interesting way to correct the signal to get sub-centimeter accuracy. Using the timing differences between satellites with a stationary unit and then sending the that to the rover is a cool workaround and can be used without expensive equipment now.
VRS RTK can even get 1cm RMS without needing a stationary unit. Just need atmospheric correction data for your approx location. Which has been amazing for outdoor mobile robotic applications.
True. But the GPS receiver doesn't know whether you are on the surface or not (and at what elevation), so it must always assume 3d space, hence a sphere.
I love these incredibly simple and elegant classic technologies. GPS is one of the best. It seems like it would be incredibly complicated and mysterious, but it's actually quite straightforward.
I'm working on a presentation now to explain how GPS works to second graders. If they understand it, I'll take some photos and do a write-up.
even though the concepts are straightforward, the implementation requires great care in order to maintain and extract the required precision. Throw that tech into space takes everything to an even higher level requiring radiation hardening, weight management, and long term reliability. You can't send repair crews to fix them if they break. As an engineer, I am in awe of those who design and build these things.
The fact that they deliberately manufacture the satellite clocks to tick at the wrong frequency on the ground (10.22999999543 MHz instead of 10.23 MHz) so that relativity makes them tick correctly in orbit is one of my favorite engineering details in any system.
Decreased vertical precision is an artefact of measurement geometry more than e.g. number of frequencies.
Horizontal position has the benefit of having satellites at almost all azimuths. But the vertical position estimate only gets satellites from at most half of possible elevations (above the horizon).
Reminder of Bartosz Ciechanowski "GPS" article:
* https://ciechanow.ski/gps/
* 2022: https://news.ycombinator.com/item?id=29981188
* 2023: https://news.ycombinator.com/item?id=36180316
* Others: https://news.ycombinator.com/from?site=ciechanow.ski
Standford's "An Introduction to Satellite Navigation" course is also instructive (recorded 2014):
* https://www.youtube.com/playlist?list=PLGvhNIiu1ubyEOJga50LJ...
A slightly related question, if anyone knows - has phone GPS gotten worse in recent generations? More reliance on local wifi networks or something like that?
I ask because I do a lot of backcountry hiking, camping, and foraging and rely on true GPS-only navigation. My most recent two phones (iphone and pixel) have noticeably worse GPS performance than previous phones, and I even changed OS ecosystems mostly hoping for better GPS, but it didn't help. Maybe I've had bad luck, but two noticeably bad phones in a row seems like it may be a pattern.
And is there any way to find phones with very good GPS performance?
I don't know, but I have noticed that the GPS in my watch (Garmin) seems to be better than the one in my phone.
Also, RTK is an interesting way to correct the signal to get sub-centimeter accuracy. Using the timing differences between satellites with a stationary unit and then sending the that to the rover is a cool workaround and can be used without expensive equipment now.
VRS RTK can even get 1cm RMS without needing a stationary unit. Just need atmospheric correction data for your approx location. Which has been amazing for outdoor mobile robotic applications.
The explanation about the spheres is slightly inaccurate. With one satellite, you won't get a circle on the earths surface, but a sphere.
GPS is not constrained to earths surface (or the oblate spheroid approximating it), luckily.
> With one satellite, you won't get a circle on the earths surface, but a sphere.
With one satellite you get a sphere in 3D space, but if you are on a surface (like that of the Earth), that gets translated into circle.
If you are in a plane in the sky (3D space), then you get a spherical 'location fix'.
True. But the GPS receiver doesn't know whether you are on the surface or not (and at what elevation), so it must always assume 3d space, hence a sphere.
I love these incredibly simple and elegant classic technologies. GPS is one of the best. It seems like it would be incredibly complicated and mysterious, but it's actually quite straightforward.
I'm working on a presentation now to explain how GPS works to second graders. If they understand it, I'll take some photos and do a write-up.
even though the concepts are straightforward, the implementation requires great care in order to maintain and extract the required precision. Throw that tech into space takes everything to an even higher level requiring radiation hardening, weight management, and long term reliability. You can't send repair crews to fix them if they break. As an engineer, I am in awe of those who design and build these things.
The fact that they deliberately manufacture the satellite clocks to tick at the wrong frequency on the ground (10.22999999543 MHz instead of 10.23 MHz) so that relativity makes them tick correctly in orbit is one of my favorite engineering details in any system.
Wish they could solve the GPS altitude weakness
Watches that use GPS for altitude are terribly inaccurate
It is interesting to run the opensource GPSTEST app on a smartphone and watch the MSL "settle" over time but each sat seems to disagree
* https://github.com/barbeau/gpstest
btw watches are now getting THREE multi-band L1+L5 GPS chipsets, should help things
quad-band GNSS coming soon too!
* https://the5krunner.com/2026/03/06/tri-band-gps-garmin/
Decreased vertical precision is an artefact of measurement geometry more than e.g. number of frequencies.
Horizontal position has the benefit of having satellites at almost all azimuths. But the vertical position estimate only gets satellites from at most half of possible elevations (above the horizon).
See "Vertical Dilution of Precision":
https://en.wikipedia.org/wiki/Dilution_of_precision