Experimental fingerprint authentication avoids storing biometric templates

A new research paper out of Pohang University of Science & Technology (POSTECH) showcases a biometric authentication technology that can “identify users without storing their fingerprint information as digital data.”
Published in the international journal Nature Sensors, the paper addresses the problem of irreplaceability, in that biometrics are unique, and storing them in any form is risky. “Even when the stored information is encrypted, the authentication process still requires access to biometric data, leaving potential security vulnerabilities,” says a news release.
The system uses ultrasound and a specially designed surface to perform fingerprint matching physically, rather than capturing a fingerprint image and comparing it with a stored digital template. Depending on whether the fingerprint matches an enrolled user, the ultrasound energy is directed toward one of two detection regions.
The technology uses a metasurface-driven loop-diffractive neural network, or loop-DNN. The metasurface controls the direction, intensity and phase of ultrasound, allowing the system to perform the matching process without conventional digital fingerprint processing.
How acoustic fingerprint matching works
When a registered fingerprint is presented, acoustic energy concentrates in a designated detection region. An unregistered fingerprint directs the energy toward another region. The system compares the signal intensities to determine whether the fingerprint matches, without reconstructing a fingerprint image or extracting digital features.
Researchers explore alternative to stored fingerprint templates
In experiments involving fingerprints from four individuals, the system achieved 97.92 percent accuracy across 192 validation samples. The small experimental dataset means its performance in real-world authentication remains unproven.
Further technical development is needed before practical deployment. The study demonstrates the potential for replacing conventional digital fingerprint matching with physical computation.
According to professor Junsuk Rho, who led the study, “rather than developing a faster algorithm or stronger encryption technology, this study changes the starting point of biometric authentication itself.”
“By eliminating the need to separately store biometric information, we expect this approach to develop into a new authentication method that can reduce the risk of fingerprint information being exposed through hacking.”
Article Topics
biometric authentication | biometrics | biometrics research | fingerprint biometrics | metasurface | ultrasonic fingerprints







Comments