I think there have been attempts to do this with physical beacons that can orient phones nearby, but I would imagine it can be hard to maintain in the harsh operating environment of the subway.
What I'd be more curious about is being able to trust the beacons.
Certainly there is a mechanism where if a phone can get a GNSS/GPS fix, it can then listen for other signals like Wifi and cell towers, and correlate the two: so if GPS isn't available other stuff can heard and used for location.
But if there's no GPS, but there are other signals, how can you map the signals to an actual location?
You'd have to have someone (Apple? Google?) go in with survey equipment and map the non-GNSS (beacon) signals to locations 'manually' and put them in a database.
Probably not economically feasible but I don't think technically impossible given analytics from devices.
And concidering cell signal is likely covered down there and the locations of cell towers are almost certainly mapped they could use some sort of triangulation to generate a pretty reasonable map even without having ever been there.
As I said, I doubt they would put the money into it... Would be a really fun project to build out if you had access to the data though... Maybe don't want to do it because it would make it more common knowledge about how easy it is to track device location even without GPS...
I don't think you would need physical beacons, cell towers or fixed wifi APs[1] will likely be good enough and using some statistics would make removing cases where an AP moved/get renamed / goes down temporarily.
Generally speaking, agencies already have to put things like signs, tactile indicators, etc. next to landmarks like stairs and exits in public spaces; so making the accessible thing also a beacon wouldn’t be that much more complicated.
I believe it’s possible to just have the landowner/transit agency install them and publish their locations according to some standards. I believe shopping malls and airports are already mapped this way sometimes.
