5

A general question: is it possible to retrieve information about how far away e.g. a computer is from a wifi-router. For instance I want to get data on my computer if I'm 10 meters away from my home-wifispot or 2 meters.

Any idea if that is even possible?

Edit: How about bluetooth? Is it possible to get information about how far away bluetooth-connected devices are one from another?

Mario Kutlev
  • 4,897
  • 7
  • 44
  • 62
matt
  • 42,713
  • 103
  • 264
  • 397

3 Answers3

4

I would recommend a measuring line or just good-old-fashioned guesstimating.

There is no "simple" way to do it (complex ways may involve building "accurate" signal maps ahead of time or trying to fit a better equation which is still subject to anumber of the limitations with the naive rule) and the rule of thumb "1/r^2" is just that -- a general rule of thumb. On the other hand, perhaps there is some existing software that will show you your RSS strength and make the task feel accomplished :-)

You will find useful links if you google for "RSS signal distance". This kind of task seems quite a common topic in academia w.r.t. small wireless devices ("motes") as well and there have been some interesting approaches to this problem such as the fitting of secondary low-frequency acoustic sensors.

  • a geusstimation is perfect. i wonder how about bluetooth? is it possible to track distances with bluetooth? – matt Nov 01 '10 at 20:31
  • The same concept holds -- the greater the distance (for a fixed gain and the same power output and assuming ideal transmission) the lower the signal received. When I walk downstairs and my bluetooth starts getting fuzzy/break-up I can therefor say "> 20feet" with good confidence. If it cuts (loses connection entirely) out I'm somewhere "> 30feet" (and I likely won't be a complete liar). Of course all of this depends upon your devices, orientation, and environment. –  Nov 02 '10 at 00:18
1

You can query the signal strength which is some kind of indication of distance and obstructions and a few other factors all rolled into one measure. With just plain wifi though this isn't possible directly.

Flexo
  • 87,323
  • 22
  • 191
  • 272
  • 1
    If your home environment is fairly static, you could measure signal strength at a few distances, and then interpolate between them, to refine your distance estimate based on signal strength. Of course, it will be highly direction-dependent because of obstacles. – Irish Buffer Nov 01 '10 at 17:55
  • And worse yet if your neigbhour has a microwave or the atmosphere where you live does interesting things at night then your "home environment" may be a lot less static than your realise. – Flexo Nov 01 '10 at 17:57
  • 1
    With a good, detailed map/plan of the area and pre-existing knowledge of the locations of access point(s) within that and a fair bit of processing something crude might just about be feasible, but certainly not recommended without constraining the problem a lot more. – Flexo Nov 01 '10 at 17:59
0

Try measuring the response time of the router to pings, with the data rate set to constant to avoid that effecting the response time. Take lots of samples and remove outliers to reduce errors, but you will still have a substantial quantization error. Subtract the latency of the router and computer, divide by 6 then multiply by the speed of light and hopefully you will have the distance to a resolution of a few metres.

Timothy Baldwin
  • 3,551
  • 1
  • 14
  • 23