It's easy to take GPS for granted. The global positioning system now powers not only satellite navigation systems but also a plethora of innovative apps on mobile devices.
Unless, that is, you happen to go inside a building or down a street where the signals don't penetrate.
GPS can be very patchy, especially in cities, and this has prompted the development of alternative positioning systems that can enhance or replace GPS when necessary.
Any positioning system needs fixed points of reference and because wi-fi access points don't move around much they form a good foundation for what is called wi-fi localisation.
Every wi-fi signal contains a unique number called a MAC address that lets you work out which access point it came from and just as with any other radiation the signal emitted by an access point weakens the further it goes.
In theory this means if you measure the power of a received wi-fi signal and know the location and transmission power of the access point then you can work out your distance from that access point.
If you can do this with three or more signals then you should be able to work out your position quite accurately.
In practice there are many difficulties with this method.
Not all access points emit signals of the same strength and just as GPS signals are weakened by obstacles so wi-fi signals are weakened to different degrees by different obstacles.
That makes it harder to accurately determine your range to an access point from the received signal strength - you can't easily tell the difference between being far away across open air or very close but with a heavy obstacle in the way.
To compensate for at least some of these factors most wi-fi localisation systems rely on a technique called fingerprinting. Rather than counting on theory to determine how distance affects wi-fi signals you go out and actually measure them.
To enable wi-fi localisation for a city you survey many points throughout the area.
At each place you record your exact position (determined from GPS or other sources) along with the received signal strength of all wi-fi access points within range.
The particular combination of access point signal strengths provides a kind of fingerprint of the location.
Then when a mobile device wants to know its location it just has to examine the fingerprint database and find the one that most closely matches the received signal strengths of detected access points.
Such a detailed survey isn't cheap, mind you, but it is worthwhile.
Google conducted one alongside its photographic Street View survey, though it notoriously collected a lot of private information like passwords and emails along with the MAC addresses and signal strengths needed for localisation.
With mobile broadband prices as high as they are I'm sure many people wish wi-fi could reach more than a few metres.
But the parochial nature of wi-fi does give it the unexpected ability to plug gaps where the global positioning system doesn't quite live up to its name.
- © Fairfax NZ News