why did old (like 90s/2000s) GPS take up to 15 minutes to get a lock, in giant bulky units, but my running watch gets it in 15 seconds?
In: 3751
The real ELI5 here is, GPS receivers and satellites have improved significantly in the past 20 years.
Nowadays your devices download the GPS almanac (using GPS-A functionality), which usually requires an Internet connection.
Rather than having to “remember” where you were (and hoping you haven’t moved since you last had GPS turned on), it has to find you all over again from scratch and that initial lock can take a while to find.
GPS does transmit its own almanac but it’s slow to do so in order to be compatible.
With an up-to-date GPS almanac, which includes all the latest course corrections and orbit calculations for all the GPS units, it’s quicker to get a first-lock.
So if you have Internet, and a modern device, you can get first-lock faster.
Also, a tiny GPS chip now interacts with American GPS, Chinese Beidou, Russian GLONASS, European Galileo etc. GPS constellations all over the planet, so it’s far quicker to find 3 sats (basic 2D fix) or 4 sats (3D fix) much quicker.
GPS from the 90s ? Wasn’t it opened to the public in 2000 ?
I thought this reddit stood for explain to me like I’m 5, yet reading most comments they’re waaaay beyond that.
I like simple answers, the rest I can Google if I’m interested in the technicals
One thing I don’t think anyone has mentioned is that modern chips in smartphones, watches etc will often have support for the other navigation systems too. This greatly speeds up getting a lock and also increases accuracy.
For example my Pixel 6 Pro supports GPS, GLONASS, BDS, GALILEO and QZSS.
If you install this app: https://play.google.com/store/apps/details?id=com.android.gpstest
Then you can see the different systems and satellites your phone is picking up.
Nobody’s actually posted the full explanation, at least not in the top ten comments I went through. This’ll be buried, but searchable.
1. The almanac (ephemerids, etc.) was (and is) broadcasted at a very low bitrate but is necessary for any GPS receiver to obtain a fix. The receiver needs to know exactly what orbit each satellite it wishes to use is following, since GPS works on light lag. Today it can be downloaded over the Internet, so the receiver has it very quickly.
2. Nearby WiFi networks are often catalogued and you can be located using them, but so are nearby cell towers. They all have their own unique ID, and this can be referenced to location using online services. If your phone can see WiFi BSSIDs x, y, z, and cell towers A, B, C, and D, an online locator service (Qualcomm operates one for its Snapdragon A-GPS systems) can get you to within around 50 metres in most cases.
3. (Which is often missed) Cell towers are sector-based. Your carrier will know which sector you’re on, which tower you’re on, and how strong the signal is, it then has a rough direction and range, which it can provide back to your device. If you’re doing multi-cell MIMO, you’ll be on more than one tower, enabling the carrier to triangulate you better.
What tends to happen is you first get a very rough location, from the cell tower data. This data is being piped to your smartphone constantly anyway (for things like hand-off to another tower, it’s basically “Tower 34 will be stronger than me soon, look for it and connect to it when you can”). It will then be sending out a request for an updated GPS almanac while it sends the tower ID and WiFi visibility to another service, and while it listens for GPS (or Glonass, or Galileo…) signals.
So you see the blue inaccuracy circle shrink on Google Maps as these requests complete and the Internet tells your device where it is with more accuracy. Finally, GPS will get a 2D lock (a WGS-84 surface position) which will shrink the circle to less than 10 metres, and then a 3D lock (includes height above or below WGS-84), which will get the circle down to less than one metre.
Maintaining a GPS lock is quite battery intensive, so the high-accuracy location isn’t maintained while the screen is off (unless you change this setting), and it’ll do periodic WiFi and cell tower scans to maintain background location.
A pure GPS unit will still take a minute or two to get a location.
Phones and other consumer devices use what’s called AGPS, or assisted GPS. They use less precise but quicker means of geolocation to get a quick fix while working to narrow it down with more precise methods.
Your phone knows what cell tower you’re connected to and can estimate distance based on signal strength, so with a database of cell tower locations there’s a quick estimate. There are databases of public Wi-Fi networks and their locations if you’re in a built-up area. Your phone knows your approximate altitude so could compare against a topographical map. It’s got an accelerometer so it knows approximately how far it’s moved since the last time it checked its location.
All those can be used to narrow it down while waiting for a definitive fix from GPS.
You know when you open the maps app on your phone and your location is a circle the size of a small town, then it narrows down to a block, and then it keeps shrinking until it’s an accurate dot? That’s AGPS at work.
The big reason? Computers have gotten faster and smaller.
In addition to this, most devices with GPS now can easily estimate where you might be through other means as well as make reasonable guesses as to information they’ll eventually receive from satellites, which makes solving the problem more efficient and means they don’t have to have a full cycle of the data that can be transmitted by the satellites
Besides what everyone said, you can try for yourself by removing your SIM card (or perhaps airplane mode). When there’s no coverage, your phone will take the good old time to get the GPS position.
Back then, gps also used to have an accuracy limit of 300 meters. It was done to prevent usage in weapons as GPS was primarily a military system. With a 300-meter inaccuracy, you would miss the White House
Latest Answers