Biometry: Security Technology In Spotlight After Sept. 11

Posted Mar 29, 2002 by Daniela Wiegmann
MUNICH (dpa) - For the workers at Infineon, a computer chip factory, the face of workplace security very much resembles their own.

Each day, before entering the offices, employees glance into a camera. In a matter of seconds, the shape of each worker's face is stored electronically and checked against a database of those permitted to enter the high security area.

The technology used to identify and compare the physical characteristics of employees is widely known as biometry.

Infineon, based in Munich, Germany, is one of the few large firms that are already taking advantage of the capabilities of this new technology, which is garnering attention after the terrorist attacks on United States last year.

"Interest has increased sharply since September 11," says Astrid Albrecht, a European security expert.

Numerous security projects based upon biometry are already being planned by German firms, among them the Frankfurt Airport, one of the world's largest air hubs. Frankfurt Airport intends to verify the identity of all its personnel using biometry when they arrive at work.

It's an exciting time for the producers of this technology, as they watch to see whether the long-awaited breakthrough of biometry has finally arrived. Still, their largest problem is not a new one: the large number of mistakes that the technology makes.

With national identity cards common throughout much of the world, outside of the United States, biometry has taken on new importance.

With his suggestion last fall that a fingerprint be included on the German national identification card, for example, German Interior Minister Otto Schily brought biometry into the limelight.

Willi Berchtold, head of the Munich chip manufacturer Giesecke and Devrient, suggested taking things a step further. He called for the introduction of a standardized identity card in all of Europe, including biometry measurements and a digital signature.

Yet Infineon Technology's Soenke Mehrgardt sees this as but one of a variety of potential uses. The technology already exists to secure computers, cell phones, guard posts, rooms, and automobiles through fingerprint sensors or other biometric processes.

The biometric recognition process is based on the assumption that certain physical features of humans - such as the iris, fingerprints, the voice, lip movements, and face shapes - do not change substantially over the course of one's life.

These physical characteristics, therefore, can be electronically evaluated and identified.

In practice, a person's physical characteristics are stored on a chip as a reference model and are then compared with the original.

This means that one's own body serves as the key - and with it a potential end to the countless PIN numbers and passwords that are currently an annoying part of everyday life around the world.

While manufacturers of the technology have been singing its praises for years, the impetus for a genuine breakthrough for the technology has never really occurred. The major cause of concern surrounding the technology is its consistently high failure rate.

"Regardless of which process is used, none has achieved a 100 per cent identification rate," reports Henning Arendt, director of BioTrusT, a Europe-based biometry project.

All humans regenerate their lost cells every few weeks, including cells of the supposedly unchanging characteristics.

For systems with too low of a tolerance threshold, even a small cut on the finger will bring an erroneous misidentification. For some of the identification processes based upon biometry, the mistake rate reaches double digits.

"It's unimaginable that 10 per cent of the residents of one country would have trouble crossing the border," Arendt speculates.

That's exactly why biometry has been best suited for use with smaller groups up to this point. A firm with a limited population can feasibly, for example, control access to its computer network using a biometric fingerprint sensor on the keyboard.

But before the technology is ripe enough to be put into widespread use, more expansive testing must be completed, Arendt reports. He raises another untested concern as well: the possibility that the systems could be fooled through photographs, tape recordings, or video.