QUICK! You desperately need a cab, but can鈥檛 see one anywhere. Or you鈥檝e
broken down on a country road late at night, but don鈥檛 know where on earth you
are. Or perhaps you鈥檙e new in town and have a craving for pizza. But where is
the nearest Italian restaurant?
In a few years鈥 time, you won鈥檛 think twice about these problems. A quick
call from your mobile to a wapsite will solve everything. In the US and Europe,
cellphones will soon be able to pinpoint their location. This will open the door
to a vast range of new services. Working out your position instantly, remote
taxi-hailing and finding a restaurant are just the start. You need never be lost
again.
The drive to improve location fixing started with the US Federal
Communications Commission. It wants to ensure that emergency services can
pinpoint people who call in from mobiles but don鈥檛 know where they are. The FCC
ordained that by this time next year all operators of cellular networks in the
US must have decided how they intend to locate mobile phones. They must then put
those plans into action. 鈥淚 think that the world is watching what鈥檚 happening
with the FCC as a benchmark,鈥 says Melanie Cox of the British-based company
Cambridge Positioning Systems. Though the European Commission wants similar
plans in place by 2002, the emphasis in Europe is less on complying with
regulation as seizing opportunities. 鈥淐ompanies see the potential revenues from
services,鈥 says Cox.
Advertisement
Analysts agree that a multibillion-dollar market is about to open up.
What they can鈥檛 agree on is which method phones will use to divine their
position. Competition between technologies is fierce, and it鈥檚 too early yet to
say which will win.
At present the best you can do to locate a cellphone is find out which cell
it鈥檚 in. In central London a cell might be as little as 200 metres across, but
out in the sticks it may cover several square kilometres. Either way, the
resolution is useless for all but the least demanding needs.
You might think that the obvious way to locate a phone is to use the Global
Positioning System. GPS receivers rely on a network of satellites orbiting the
Earth. By measuring the time taken for a radio signal to travel from one of
these satellites, the receiver can calculate how far away it is. Do this with
three satellites鈥攚hose positions are known鈥攁nd the receiver can work
out its longitude and latitude. Adding a GPS chip and antenna to your phone will
allow it to do the same. In an ideal world, the fix would be accurate down to a
few metres, so a breakdown crew could find out, for example, which side of a
motorway your car was on.
But conventional GPS has its drawbacks. It doesn鈥檛 work if walls, car roofs
and so on prevent direct communication with the satellites. Also, adding
components that consume power would be the kiss of death in a market where the
watchwords are small, light and powerful.
With this in mind, SnapTrack, a company based in San Jose, California, has
invented something called server-aided GPS, which it boasts will allow
position-fixing from just about anywhere a mobile call can be made. Its main
innovation is an ultra-sensitive receiver that can pick up satellite signals
even after they鈥檝e been weakened by walls and windows. The company has also
tackled the problems of increased weight and power consumption by shifting as
much processing as possible to a remote server.
Emergency measure
When a user requests a fix, the server finds the phone鈥檚 rough position from
the cellular network and informs the phone which GPS satellites are in view. The
phone simply works out its distance from the satellites and sends this
information to the server, which then calculates the caller鈥檚 position. In an
emergency, those details would go to the emergency services, but they could just
as easily be sent to other service providers or back to the phone itself. To
keep the weight down, SnapTrack has developed a combination chip that will
handle all the signal processing within the phone, says company spokesman John
Cunningham.
If adding a new chip and antenna to a mobile is not acceptable, alternatives
to GPS do exist. These Earthbound solutions find phones by using the antenna
base stations, the phones themselves and the signals that pass between them.
Cambridge Positioning Systems would also need a network of listening stations
and some added software in phones. Computers could then calculate the position
of a phone relative to three base stations and a single listening station, using
a technique called enhanced observed time difference or E-OTD
(see 鈥淲hat a difference a microsecond makes鈥).
In another option, called time difference of arrival (TDOA), three base
stations measure the times that signals arrive from a cellphone. From the
differences between those times, a computer plots curves on a map that intersect
at the phone鈥檚 location.
Both E-OTD and TDOA hit trouble if a mobile phone is not in touch with three
base stations. But in San Ramon, California, US Wireless is developing a method
that works even if you are only in contact with one base station. Its secret
weapon, installed on base stations, is called the RadioCamera. To set it up,
cars travel the roads within a cell and send out phone signals from precisely
known locations. The RadioCamera chops up the waveband into 5-hertz channels.
For each channel, it then measures the distortions to the incoming signal caused
by reflection and interference. This gives a 鈥減rofile鈥 of the noise across the
waveband at every location. So when somebody wants to know where they are, the
RadioCamera compares the noise profile of their call with the stored profiles.
If there鈥檚 no exact match, it estimates the location that鈥檚 most likely to give
the observed profile.
The main drawback with Earthbound solutions is that operators have to install
costly equipment. In some cases, though, this may be the only option. For
example, since nobody is going to make new analogue phones, it makes sense to
add any new technology to the network, not the phones. For these systems, then,
TDOA and RadioCamera have the edge.
There is also competition to make techniques as precise as possible.
Estimates of the resolution of E-OTD, TDOA and RadioCamera vary depending on who
you talk to, but they all fall between 30 and 100 metres. As yet, none comes
close to GPS, though this will change when 3G phones arrive. The higher data
rates of these phones will allow more precise measurement of time, which should
improve resolution of E-OTD and TDOA by up to a factor of ten.
Today, Cunningham claims, the 5-metre resolution of GPS is in a league of its
own. If people in towns want directions for how to get to, say, the nearest
pizzeria, he thinks his system will lead the way. 鈥淥ne hundred metres doesn鈥檛
get me a darn thing in an urban environment,鈥 he says.
Big Brother
While that may be true if you want directions, there are plenty of services
for which you don鈥檛 need such high resolution, says Cox. If you鈥檙e looking for
nearby restaurants, knowing where you are to within 50 metres is fine. And if
all you want is the local weather, such distances are neither here nor
there.
Among the services that will emerge as location fixing takes off is
鈥渁sset-tracking鈥 for firms such as taxi operators and delivery companies. To
make sure you know where your taxis or motorbikes are, simply give the drivers
mobile phones.
But this opportunity for companies鈥攁nd government agencies鈥攖o
play Big Brother is bothering some people. It鈥檚 fine for people to locate you in
an emergency, but who wants their every move tracked? 鈥淭his technology is
rushing ahead without any regard to privacy issues and questions of what the
legal standards should be,鈥 says David Sobel of the Electronic Privacy
Information Center in Washington DC. He thinks the law urgently needs to decide
when tracking can and can鈥檛 be used. 鈥淭echnologies that have invasive potential
are frequently rolled out in an application that most people find useful and
non-threatening,鈥 he says. 鈥淚t鈥檚 only later that some of the more troublesome
uses of the technology become apparent.鈥
Here again, some techniques score over others. With GPS and E-OTD, people
would press a button or punch in a number to find their position. But
RadioCamera and TDOA can work on any call, so users would have to trust the
network operators not to track them. The only way they could avoid detection
would be to switch off their phones.
What with cost, expediency, precision and the legal status of tracking,
there鈥檚 plenty to be decided before any single service begins to dominate.
鈥淭here are all these trade-offs that wireless operators have to go through,鈥
says Casey Hill of Motorola鈥檚 Location Solution Center in Schaumburg, Illinois.
鈥淚 haven鈥檛 seen any one thing that solves all the problems, so you鈥檒l probably
see a combination of all these things.鈥
To locate a phone, Cambridge Positioning Systems employs some simple physics.
A person makes a call to a coordinating centre. This collects the times at
which a signal from a base station arrives at both the mobile phone
(t1) and a listening station (t2). The distance from base station to
listening station (d2) is known.
From these values it鈥檚 relatively easy to calculate how far the phone is from
the base station (d1), because d2鈥 d1 = &ngr;(t2
鈥 t1), where &ngr; is the speed of the radio waves. Knowing the distance to
the phone gives a circle of radius d1 around the base station.
Calculate similar radii from two other base stations and the point where the
circles intersect gives the caller鈥檚 location.
