Fresh impetus has been given to biometric checking of travellers due to the threat of terrorism and also ICAO's new standard for passports. But widespread implementation is still some distance away

Biometrics technology - the identification of passengers by facial, iris, hand or fingerprint scanning - has supposedly been just over the horizon for years. IATA's Simplifying Passenger Travel (SPT) initiative set out the airline vision in the late 1990s: check-in, passport control and boarding all done automatically, using biometric identification instead of manual checks and paper documents.

Originally, it was all about improving service to travellers. Then came 11 September. Suddenly, biometrics had a new purpose. Attention turned to the ability to weed out terrorists and allow regular travellers to pass through security more quickly. In fact, many thought the attacks would provide a spur to the wider implementation of biometrics for passenger travel. But has it?

On one level, not a great deal of progress seems to have been made. There are still various trials going on, but as Robert Bailey, vice-president for government at SITA points out, these have tended to focus on specific areas rather than the end-to-end process enshrined in the SPT. "Most trials have been in the area of expedited border checks or frequent flyer checking at the border," he says. "They have also focused so far on proof of technology and, to a less extent, on process change."

The focus on border controls is borne out by a list of SPT trials provided by Jeffrey Durante, programme director for SPT at IATA. Of seven tests on the list, six are purely concerned with border controls, including INSPASS in the USA. The exception was a three-month trial at Tokyo Narita, earlier this year, of facial and iris recognition for check-in, security and boarding, which involved Japan Airlines and the Narita Airport Authority, as well as government and technology companies.

The emphasis on border-control trials is not entirely surprising. As Durante says: "Governments will pay for the trials of processes under their control, but they are not going to pay for trials of airline processes." He admits that, in the current climate, it is hard to make a return-on-investment case to carriers for forays into biometrics, even though he thinks such a case exists. "The problem is that it costs money to save money, and, at the moment, there is no money with carriers," he says.


Terrorism has, of course, prompted more interest in border security. Durante says it has become the primary driver within SPT, although he says the focus is still ultimately on customer convenience.

Martin Huddart, vice-president for business development at Ingersoll-Rand Security and Safety, and chairman of the International Biometric Industry Association, agrees. He says there has been "a big shift from convenience and speed to checking against the watch-list".

That said, there are hopeful signs on the horizons for fans of the larger SPT project. One that has been universally hailed as a big step forward is the decision by ICAO in May to adopt facial recognition as the globally interoperable biometric for machine-readable passports and travel documents. Mary McMunn, chief of the facilitation section at ICAO, says the face was chosen because it was unobtrusive, more acceptable to travellers, and because "most databanks in the world contain pictures, as opposed to other types of biometric".

Although border-control needs were the first consideration, she says facial recognition technology, which involves comparing a traveller's face to key measurements of it in an encoded digital version on a chip, does have advantages for SPT. Trials of iris and fingerprint technology, McMunn adds, may work with experienced travellers who know their way around airports, but could be problematic with less regular air travellers.

The agreement on a passport standard at ICAO opens up the enticing possibility that large numbers of travellers will soon be equipped with a biometric-enabled document, which might then be used for other applications in the SPT vision. In theory, this could lead to much wider adoption of biometrics than merely the enrolment of frequent travellers in special schemes.

Passports also possess big security advantages over other types of biometric identity documents, such as smartcards. One big danger for any biometric card is from terrorists managing to fake them, or registering under false identities. Passport authorities, in theory at least, are fully experienced in countering this problem, something that cannot be said for an airline's frequent-flyer scheme. The ICAO standard will also incorporate rigorous public-private key encryption - a standard that is still a work in progress. This will ensure that data, once written on the chip by the relevant authority, cannot be altered without detection.

It is tempting to see the new biometric passports as the key to realising the rest of the SPT vision. Durante says that, for SPT, they are "a foundation that provides the mass-market adoption necessary to justify the development of add-on airline and airport processes". In theory, he adds, a person could be identified by their passport at check-in, and then pass all subsequent airport processes merely by showing their face. But, he says, although the ICAO standard "is available, whether it will be used is another matter".

One obvious drawback of basing all progress on biometric passports is that it could be a long time before they become common among travellers. A new US law, has given the process a major boost by requiring the 28 countries in its visa-waiver scheme to issue biometric-enabled passports to the ICAO standard by 26 October 2004. Several nations have already indicated they have started on this process.

But, as Bailey says, due to passport expiry dates, it will take 11 years for all UK passport-holders to acquire a biometric passport; the other 27 countries are in a similar position. "The new standard is certainly a leap forward, but there is a long way to go," he adds. "It will funnel efforts that were a bit anarchic before, with each technology provider vying to implement its own solution, but it won't change everything overnight."

Other measures

Huddart also warns against regarding the ICAO standard as the last word on the subject. He makes the distinction between two types of biometrics checking. Firstly, "reference biometrics" used to check an identity against a database of pre-registered people, such as a frequent flyer or watch-list. And secondly, "operational biometrics" employed to ensure that a person is the same individual as encoded on the document that they present. That could include checking the identity of someone who has just stepped off an aircraft.

So far, Huddart says, most of the focus has been on the reference side, with less work on operational applications. He thinks that one biometric test, based just on the face, is not enough for many reference uses. "For reference, I think you need a second biometric, such as fingerprints. Otherwise, if you are searching a large database, the chance that two people are alike is quite high. You will end up saying a person looks like one of 10 people, and needing a manual check to be sure." Trials on the operational side might also throw up other areas where a facial biometric is not ideal. "Hand geometry is very common for this application," he adds.

As it happens, the ICAO standard has foreseen this, allowing iris scanning and fingerprinting as secondary biometrics. But neither will be mandatory for passports. This seems to open the door to combinations of biometrics being used by different applications.

Technology companies, which have spent the last two years developing various solutions, clearly hope so. "At one level, the facial standard is a barrier, since people think it is the only thing they can do - but that's not true," says Huddart. "The ICAO standard is a positive move, but not by any means the end-game."

Neither does the passport solution serve the purpose for domestic travellers - in particular those within the USA or the borderless Shengen area within the European Union (EU). On the European side, IATA, SITA and the leading European airlines are involved in a project called S-Travel, which aims to use biometrics to check in and board passengers, using smartcards within the Shengen area.

A trial is due to start later this year on Alitalia flights between Athens and Milan. Bailey says the biometrics standards chosen will aim to conform to SPT and ICAO standards. A more pressing issue, perhaps, is how to ensure the trial conforms with EU data privacy laws.

In the USA, the picture is less clear. The Transportation Security Administration (TSA) is setting up a Transport Workers Identity Card scheme using biometrics, and another for Airport Access Control. Both are testing technologies that might be used in a voluntary Trusted Traveller programme for frequent flyers within the USA.

ICAO standard

As yet, the TSA has not announced which biometric it will use in this programme. McMunn says there is no obligation on it to follow the ICAO standard, although she says that, as airlines handle travel documents in check-in processes as often as passport authorities do in border controls, there is a strong logic to it doing so. But John McKeon, head of biometrics and smartcard technology at the travel and transport division of IBM Global Services, believes that the border control standard will ultimately underpin the TSA scheme.

There are other areas of airport management and security for which biometrics could also be used, and which might create competing technologies to the ICAO standard. An example is access control for employees, which, as Bailey points out, is easier to implement than the SPT solution, since it works on a captive group of users, and does not require major changes to the layout of airports.

Qantas's SmartGate programme, allowing flightcrew to pass through immigration in Sydney, is only one example.

There are also plans to extend this to other airports, and, perhaps, to Australian passport-holders. SmartGate also employs facial recognition, but other airports are using hand geometry, fingerprints, or iris scanning for employee access.

Huddart thinks such applications are initially more likely to appeal to cash-strapped airlines as a cost-cutter in preference to passenger-centred solutions. He also points to frequent flyer lounge access as possibly suiting biometric applications. Ingersoll-Rand is in discussion with several airlines about this, he says. "The ICAO standard is being viewed as a one-size-fits-all cure-all, but I think there are lots of other areas where biometric technology could be applied," Huddart explains, adding that he expects a variety of such applications to flourish, with standards following after some years, once technologies are proved in live situations.

Others expect steady progress towards the SPT model, but perhaps no dramatic breakthroughs in the near future. Durante says the aim will be to set up end-to-end trials, from check-in to border control. He also takes heart from the rapid growth of self-service kiosks, in particular the first signs that SITA's common-user, self-service check-in (CUSS) is being adopted at some airports.

These, he says, could be a key to the future integration of check-in and border controls, allowing visa and immigration-related questions to be answered by the passenger on a screen at check-in - ideally then connecting to the relevant country's immigration systems to get the permission to proceed.

A version of this is already in operation for visitors to Australia, while Canada, the UK and the USA have all expressed an interest. "More work is needed in the area of information exchange between airlines and government," Durante says. "It is critical for SPT that we get a board/no-board message back from the arrival government at the time of check-in."

This, he adds, will be much easier to achieve with CUSS than with a range of individual airline terminals. Durante says an SPT trial with an unnamed airline, airport and government authority using CUSS kiosks will start soon.

CUSS can be operated without biometrics - the usual form of identification is a credit card - but the system would obviously be much more secure when linked to biometric-enabled passports. For its part, SITA says that the kiosks are ready for biometrics from a technology standpoint, and that they are simply waiting for the first customer.

Passenger applications

Lois McKeon, business unit executive, safety and security in the travel and transport division at IBM Global Services, says there might be more time for passenger applications in biometrics in the USA, now the rush to federalise baggage screening and install explosives-detection systems is slackening off.

A TSA lead on its Trusted Traveller scheme would also help, she adds. "Airlines would love to have these type of passenger processes, but they can't do anything that impacts on security without the TSA, and security is one of the big areas where there is scope to speed processes," McKeon says.

Bailey just expects more of the same, with technology companies, airlines, airports and the government slowly ironing out the remaining barriers to biometrics and SPT. "Progress may seem slow overall, but underneath fast progress is being made," he says.

And what of those long security queues at airports? "I don't think there will ever be no security for trusted flyers, but there could be smarter screening of passengers," says Huddart. "It is all about risk management, and focusing on those of higher risk."

Lois McKeon agrees: "The trusted traveller will perhaps get security back to pre-11 September levels," she says. "They maybe won't have to take their shoes off. It will be a bit like the car-pool lane on the freeway."


Source: Airline Business