Hacker News Reader

Comment by jofer | Hacker News Reader

I've noticed the same thing, but I'm always surprised by it. Particularly with smartphones.

A GPS is basically just a very, very precise clock and a radio receiver.

Why don't smartphones use the GPS clock?

I'm sure there's a good reason (and some probably do), but it sill surprises me. (Presumably it's battery related -- the GPS clock can't be used unless GPS is fully enabled or something similar.)

View on HN

The precise clock is located on the satellites, not receivers. The receivers figure out the correct time by listening for the signals from multiple satellites and matching them against each other and it’s own internal tables. This is the expensive part.

Indeed, but the GPS fix you're getting is 4-dimensional: if you have position, you also have precise time. Therefore - listening for the GPS time is not necessary all the time, just update when a precise-enough fix is achieved.

Probably cheaper in terms of battery life to just use NTP than to periodically get a GPS fix.

Of course. I was thinking "when another app is requesting a GPS fix anyway, might as well use the temporal part, as you're getting it for free."

It's worth noting that to get time with .1s accuracy you don't actually need a GPS fix, just to pick up the 50-bit-per-second navigation signal for a couple seconds, from one satellite.

How would that work unless you know the distance to the satellite?

If you're on Earth, you can assume that the satellite is closer than 30000 km or so. Light takes 0.1s to cross that distance.

The distance to an overhead GPS satellite is somewhere in the range of .07-.09 light-seconds. The planet's width is basically irrelevant when your goal is .1 second accuracy.

Based off of some quick and dirty calculations for a GPS satellite at approximately 20,000km above sea level, the difference in distance between a satellite directly overhead and one straight out over the horizon is around 5500km.

That's actually only about 18ms difference in delay. Unless there's something I'm failing to account for here, you should be able to get more than 0.1s precision based off of a single satellite's signal.

Subtract a flat ~75ms to account for the minimum delay (~66ms @ 20,000km) and half of the variable delay, and you shouldn't really be more than about 9ms out.

Reply on news.ycombinator.com