Agilent Claims Location Technology Advance

CANNES, France – 3GSM World Congress – Pinning down the location of mobile phone users has become a big issue this year – partly because it’s become a requirement in the U.S. (to aid emergency services) and partly because it’s seen as a big potential source of new money-spinning applications.

The big challenge for operators, however, is that there’s a bunch of ways to locate users and each one demands an awkward compromise between accuracy and cost.

Agilent Technologies Inc. (NYSE: A) claims to have solved the problem with its acceSS7 subscriber location solution, announced here at the 3GSM World Congress.

The acceSS7 delivers an unparalleled combination of high accuracy and low cost, according to Andy Belcher, vice president and general manager of Agilent’s telecommunications systems division. “This is probably the breakthrough the market needed,” he says.

On accuracy, Agilent says that its acceSS7 can, on average, locate users within 100 meters – slightly more in rural settings and slightly less in urban.

On cost, Agilent says the acceSS7 can be deployed for between 50 cents and $1.00 per subscriber. This is for a new, standalone system. As the acceSS7 is an add-on to Agilent’s existing system for managing the health of GSM-based networks, the incremental cost could be a lot less for some operators.

Competitors that specialize in this area, such as Cambridge Positioning Systems and TruePosition Inc., cast doubt on Agilent’s claims, saying they’re too good to believe.

To dig deeper into the matter, a bit of background is needed.

First off, the cost of location technologies is often tied to where hardware and software has to be added in networks. A typical operator might have hundreds of thousands of users, tens of thousands of base stations, and a few hundred base-station controllers.

Current location technologies fall into the following categories:

• Cell ID makes use of existing info in the network and is thus very cheap. However, accuracy ranges from 500 meters to 30 kilometers in some cases, according to Agilent.
• Enhanced Observed Time Difference (EOTD), a technology patented by Cambridge Positioning Systems, deduces the position of users by measuring how long it takes signals to reach base stations. This delivers good accuracy (50 meters in most cases) but is expensive, according to Agilent, because large numbers of handsets may have to be modified. It also relies on sending “Where are you?” signals to the handset, which consumes bandwidth.
• Global Positioning System (GPS), the satellite-based technology, delivers pinpoint accuracy – down to a meter or so – but is very expensive, because hardware has to be added to handsets. It also has trouble locating users inside buildings and may not work at all when the user is making a phone call. In addition, it can take 20 minutes for GPS to lock on to users, according to Agilent. So-called “Assisted GPS” aims to solve the time-lag problem.
• Time Difference On Arrival (TDOA), a technology patented by True Position, affords another way of deducing location, by timing signals between users and base stations. This is accurate and doesn’t require handsets to be modified. However, in a typical network, equipment has to be added to tens of thousands of base stations – so the cost works out at $3 to $5 per subscriber, according to a company official, who asked not to be identified.
  
  

Agilent says it uses a whole bunch of technologies to achieve its goals, but all of them are based on adding probes to the few hundred base-station controllers in a typical network. No modifications are necessary in handsets of base stations, which keeps the cost down.

The probes aren’t new. They’re already used to collect data for Agilent systems used by many operators to monitor the health of their networks. “We own this space of link monitoring,” says David Craig, manager of Agilent’s mobile location services business unit. The acceSS7 merely processes this data in a different way to deduce the position of users. The probes sample signaling traffic twice a second. “You’ve got a wealth of information to get a good result from,” Craig notes.

The result is based on the following data:

• Monitoring power levels of the serving and neighboring cells. This is key, according to Craig.
• ”Software enhancement.” For instance, if a user is changing position rapidly, the software deduces that he/she is probably in a car on a road. It confirms this by comparison with other users in cars following the same line.
• Timing advance correcting factor: Another way of comparing the timing of signals from different users to deduce their relative position.
• Cell ID
• Comparing readings with the operator’s planning information covering radio frequency characteristics.
  
  

One of the drawbacks of Agilent’s acceSS7 is that it’s only applicable to GSM-based networks, which includes GPRS (general packet radio service) networks. It can’t be used in CDMA networks.

In real life, it’s likely that operators will deploy a mixture of location technologies to address different requirements. In fact, big equipment vendors like Ericsson AB (Nasdaq: ERICY) sell platforms supporting multiple technologies, with this in mind.



Right now, Agilent’s acceSS7 is in trials with some major operators, according to Belcher, who contends that “it will be available on a big scale in three or four months.” — Peter Heywood, Founding Editor, Light Reading
http://www.lightreading.com