Tightened security checks have propelled biometric technology into the limelight
Could biometrics finally be about to take off? There has been discussion for years over technology to identify individuals through fingerprints, iris or facial scans. But since 11 September, biometrics technology has been front-page news, especially in the USA. John Rockefeller, chairman of the US Senate aviation subcommitee, calls it "the future of aviation security" and recommended the technology be made an integral part of passenger and employee screening.
The US Air Transport Association also called on the government to establish a voluntary traveller identification programme using biometrics to speed frequent fliers through security checks. There has been a rash of new trials, such as a year-long pilot project announced in October by Amsterdam Schiphol to use iris scanning for border control.
Biometrics has gone from a technology that, in the words of IBM global airline marketing executive Lois McKeon, was "interesting but not essential" to top of the airline and regulatory agenda.
Biometrics was always central to IATA's Simplifying Passenger Travel (SPT) initiative, a programme which envisages passengers using biometrics matched against data pre-encoded on to a smart card for identification at automated check-in, security and immigration.
The events of 11 September have brought that vision closer, but only somewhat. Biometrics is not necessarily the answer to today's security questions, not particularly high on the list of priorities of the new US Transportation Security Administration (TSA) and something that airlines can ill afford to implement under current market conditions. "The interest is there, but the decision to go ahead is not," says McKeon of her recent contacts with carriers.
The technology still needs serious testing in an airline context, and it is not yet clear who is going to set the standards. Perhaps the best that can be said is that at least the process of getting such an agreement is now starting in earnest.
The public imagination sees biometrics as a system linked to the FBI's Most-Wanted list picking out potential hijackers from the check-in line. The technology to do this - facial recognition - is being trialled, but the challenge of implementing it is huge. Checking thousands of faces daily against large criminal databases would be an enormous computing challenge, even at a small airport.
Employee checks
Two more likely and more immediate applications of biometrics are to check airport employee identities when entering secure areas and to expedite the passage of frequent fliers through security controls. The former could be the most effective way biometrics can be used to ensure security. It involves testing a captive audience against a limited database, and has been trialled.
Tightening access to secure areas is invisible to travellers, however. They want to see something that shortens queues or makes them feel safer. Registering frequent fliers in advance for a smart card bearing details of their fingerprint or iris and then fast-tracking them through security might have this effect. It would allow screening personnel to concentrate their attention on the potentially higher-risk ad hoc passengers, shorten screening queues for all and help airports cope with the pressure on security procedures when passenger numbers get back to normal.
But airports must first decide on which biometric technology to use. No option is perfect. Iris scanning offers high accuracy and involves nothing more intrusive than staring at a camera for a few seconds, but people tend to be nervous about the idea of something shining in their eye.
Fingerprints have the advantage of a huge existing database to match against (at least when such a thing is technically possible), but have unfortunate criminal connotations. Some manual workers and some ethnic groups also have less pronounced fingerprints. And what if the finger is cut or in a bandage?
Facial scans - while less intrusive - need careful alignment of cameras and good lighting. They also have problems with facial features such as beards, and would be culturally unacceptable in Arab countries where women are veiled.
All these technologies need lengthy testing in airport settings. INSPASS and CANPASS are schemes that have been trialling hand scans for immigration fast- tracking for 10 years at eight airports in the USA and Canada with mixed success. Accuracy was poor and the process was slow, and many of those who registered found it easier to use normal channels.
There is also the question of who will set standards. Most in the industry agree that they are essential, but opinions differ about where they should come from. ICAO is one possible source: it has a working group studying the area which will release its report in the first quarter of 2002. SITA also says it is up to ICAO, along with IATA and the Airport Council International (ACI), to take the lead.
IATA assistant director of facilitation services Bob Davidson suggests, however, that it should be passport authorities who set the standard: "I don't believe that any state will accept airline-enrolled smart cards for crossing borders, so we must start with travel documents." He says passport authorities are ready to take such a step and the technology is ready, too. It is certainly interesting that several of the trials of biometrics at airports (for example, at Amsterdam and Tel Aviv) are for immigration control uses. Davidson also says the US Congress is looking at mandating the use of smart visas.
Rick Norton, executive director of the International Biometrics Industry Association, takes the opposite view. "It is a mistake to expect a government authority to standardise a biometric or biometrics system, and it is not necessary, either," he says. "The technology is available now, and we can start automating check-in or security straight away."
Norton points out the fallacy of even trying to agree to a standard before implementation. "What if we agree a standard, and then the world's best biometric is invented a year from now?" he asks. "Then we would have standardised around an obsolete technology. It is important to remember that it is not a single biometric, but biometrics in general, that will solve current problems. You need to have underlying systems capable of recognising multiple biometrics or changing to different ones."
The same point is made by SITA, and in November it took practical steps in that direction by announcing that it was opening its CUTE (common user terminal equipment) and CUSK (common user self-service kiosks) platforms to all biometrics vendors.
As its vice-president for industry relations, Catherine Mayer, says, systems may need to be able to recognise at least two or three different types of biometric measurement. "For example, what happens if a finger does not scan? That is a case for having a back-up biometric," she says. "The important thing is that the systems are able to exchange data in a format that is readable by everyone."
Government's role?
The new TSA could also offer guidance on biometrics standards, but Mayer, who spent considerable time in Washington lobbying on the Aviation and Transportation Security Act, thinks this is unlikely. "The TSA is not a standards making body - it is a policy body. In any case, it has a lot else on its plate," she says. "In my opinion, it will leave biometrics standards to industry bodies."
IBM's McKeon agrees: "If you wait for the government to do something, nothing happens for a long time. It's up to the airlines to take leadership, to go to government and to say 'this is what we want to do, and these are the technology providers we want to work with'." However, she also notes that airlines are holding back, worried about what regulations the TSA might hand down in the future. "The airlines are waiting for the government, and the government is waiting for the airlines," she says.
Most experts agree that the coming year will see an intensification of trials. Existing pilot projects, such as the ones at London Heathrow and Amsterdam, and forthcoming ones at Charlotte, North Carolina, in the USA are already under SPT auspices, aiming to test various aspects of the technology and to look at its scaleability and ease of fitting in with airport processes.
McKeon says the best hope is that in 6-12 months the industry will have come to more agreement on technologies and data standards, and real implementation can then follow. Given public and official pressure for visible change in the USA, airlines there are likely to be early adopters, she believes. In contrast, carriers in Europe and Asia - where confidence in airport security measures was less shaken by 11 September - will probably wait to see what technology emerges from the USA.
Davidson at IATA takes the long view. "The SPT vision looks down the road 10-15 years," he says. "I think we will see significant use of biometrics by then."
Source: Airline Business