christo16
christo16

Reputation: 4843

Wifi Triangulation- Distance from AP to iPhone

I'm trying to calculate a rough distance from an AP (Access Point) to an iPhone. A couple of constants:

1) I've already solved the problem of finding and determining signal strength of nearby APs (had to use private APIs, don't worry this isn't going in the App Store)

2) All the APs are the same brand & model

Now given a APs RSSI, how can I get a rough distance?

Example: RSSI is -60, what other info would I need to get to a distance?

Thanks!

Upvotes: 1

Views: 3766

Answers (3)

danyowdee
danyowdee

Reputation: 4698

Adding to andynormancx:

The thing that is totally going to kill you isn't attenuation as much as interference:

I know more than enough rooms on our campus where there are a mere 20cm between full signal and none at all.

All you can do is look whether you see a certain AP. If so, that's a good indication that you're approximately less then a few ten meters away from it.

If you don't see a certain AP you only know, that you don't see a certain AP. What you don't know is if it's...

  1. because you're out of range or if you
  2. simply have to move a few centimeters to get out of the destructive interference you are standing in.

Edit:

This, by the way, is what core location does, when there is no GPS signal... At least, it can tell me where this laptop is with an accuracy of +/- 70 meters.

That said, when you have the right number of APs, operating in considerably spaced-out channels and know where they are, you might be able to get better than core location by one order of magnitude — if the building you are in is built out of reinforced concrete for the floors and main structure with comparably thin, "removable" walls, you should be able to locate in which of these "cells" the user is.

If you are in a large conference hall, well then hopefully your exit signs on site are good enough anyway ;-)

Upvotes: 3

andynormancx
andynormancx

Reputation: 13762

You can't, not unless all these access points are running in a completely open space with no obstructions. The received signal strength will vary depending on several other factors other than distance.

Walls, trees, windows, doors etc will all attenuate the signal significantly. And of course even different types of walls for example will have very varying levels of attenuation (almost total for some types/thicknesses of walls). I expect reflections off of those same objects will also have a similar effect.

Upvotes: 3

Mikeb
Mikeb

Reputation: 6361

You can apply a model of simple signal propagation loss, and assuming you know the source power (how much energy the access point is putting out) you could approximate the distance. It would be tricky to avoid complications - things like walls will attenuate the power and so the calculation will think the AP is farther away than it actually is.

Upvotes: 2

Related Questions