3D fingerprint scanner beats Apple's

July 06, 2015 // By R. Colin Johnson
Password security is already becoming a thing of the past with biometrics taking their place. For instance, fingerprint identification is being built-in to every new Apple mobile device, thus forcing the rest of the herd to follow.

Even retinal scanners are becoming common in corporate settings, when in the past their expense was only justified by high-security areas like the White House, the Central Intelligence Agency (CIA) and the National Security Agency (NSA).

Now, however, academic researchers sick-and-tired of memorizing long passwords and bolstered by the fact that even Captcha is no longer secure, have invented a 3-D fingerprint scanner that is not only immune to false-negatives due to oil or moisture on the skin, but even looks beneath the surface of the skin, to create an ultra-secure identification system that could make passwords a thing of the past.

"In terms of robustness, the ultrasound sensor is less prone to errors due to dry/wet/oily fingers since it can image the dermis (beneath the surface) rather than just the epidermis," Professor David Horsley (University of California Davis) told EE Times. He is also co-director of the Berkeley Sensor and Actuator Center, along with co-director professor Bernhard Boser at the University of California at Berkeley. "Secondly, conducting fingerprint recognition from 3D features makes these images harder to spoof, since you need to create a 3D model of the finger to reproduce them."

 

An ultrasonic fingerprint sensor on a board here measures a three-dimensional (3D) volumetric image of the finger's surface and the tissues beneath--making it near impossible to defeat.(Source: University of California)
An ultrasonic fingerprint sensor on a board here measures a three-dimensional (3D) volumetric image of the finger’s surface and the tissues beneath--making it near impossible to defeat.  (Source: University of California)

Today a determined hacker can lift your fingerprints from any glass you touch, using the same methods that the police do to identify criminals, making it relatively easy to reproduce that image of a 2D fingerprint and spoof a device protected that way. Not so with a 3D fingerprint that looks beneath the skin with ultrasonic microelectromechnical system (MEMS) sensor.

"With a 3D fingerprint, the subsurface features are private," Horsley told us.

To prove the concept, Horsley's research team collaborated with Invensense Inc. (San Jose, California), using Invensense to fabricate the device using its Invensense shuttle service which gives MEMS developers access to its patented MEMS-on-CMOS Nasiri fabrication toolkit.

 

Block diagram show the various layers, vias and other structures that make the 3D microelectromechanical system (MEMS) fingerprint detector to its underlying complementary metal oxide semiconductor (CMOS) application specific integrated circuit (ASIC).(Source: University of California)
Block diagram show the various layers, vias and other structures that make the 3D microelectromechanical system (MEMS) fingerprint detector to its underlying complementary metal oxide semiconductor (CMOS) application specific integrated circuit (ASIC). (Source: University of California)
 

"Invensense provided the fabrication service. We used a modified version of their NF Shuttle which is a multi-project wafer (MPW) service," Horsley told us. "They also provided funding to our students through a collaboration membership at our research center, the Berkeley Sensor and Actuator Center (BSAC)."