Hsiao-Chun Wu’s wearable biometric authentication via PPG wins NSF grant

A release from Louisiana State University (LSU)’s College of Engineering says electrical and computer engineering professor Hsiao-Chun Wu is the recipient of a $245,631 grant from the U.S. National Science Foundation (NSF). Published through the IEEE, Wu’s research examines “novel robust continuous time-efficient biometric profiling and authentication using multispectral photoplethysmography.”

That’s a highly technical way of saying that Wu is exploring biometric authentication using the kind of simple wearable device often used to measure heart rate.

Photoplethysmogram (PPG) technology measures volumetric blood-flow changes in the peripheral circulation. Apple has previously explored using PPG in its AirPods.

Per the abstract, “because PPG signals are very easy to acquire, compared to other biometrics (fingerprints, iris/retina, etc.), especially by low-cost wearable electronic devices, they are widely adopted to measure heart rates.” Wu proposes to develop “a novel robust PPG based authentication system, which is capable of continuously authenticating the user instead of one-time authentication as carried out by the conventional techniques.”

Imagine a patch, ring, thimble or other wearable device capable of “continuously sensing a wide range of physiological signals for use in biometric authentication and health monitoring.” In Wu’s model, authentication is not an isolated event but a consistent state.

“As society sees increasing use of digitally interconnected devices and systems, the need for secure and user-friendly authentication methods is more critical than ever,” Wu says. “Existing methods, such as passwords, fingerprints, and facial recognition are incompatible with wearable technologies.”

By using physiological signals to enable continuous time-efficient authentication, Wu hopes to not so much erase the seams as to shift the focus to how the whole outfit looks when it’s on.

Wu’s team is set to create “advanced interdisciplinary learning opportunities” at LSU and Michigan Technological University to engage undergraduate and graduate students in “cutting-edge research across electrical engineering, biomedical engineering, signal processing, machine learning, intelligent systems and data science.”

Meanwhile, he says “we have already built the hardware prototype successfully for real-time training data collection and real-time authentication.” Per the abstract, the equal-error-rate at the authentication stage reaches 5.5 percent and the identification accuracy reaches 98 percent at the identification stage.

Previous research into photoplethysmography as a tool for authentication against deepfakes has found that video deepfakes with digitally altered faces retain the effect of heartbeat. In other words, generative AI is now good enough to fool heart rate and blood flow detectors on video.

In 2022, in a profile of a project to detect deepfakes with heart rate estimation, Biometric Update editor in chief Chris Burt wrote that, “if deepfake production techniques start to take into account physiological data related to heart rate or blood flow, the technique will become less effective.”

For video, that moment seems to have arrived. But if Wu’s technology takes off, we may end up wearing human-authenticating devices on our person constantly, like jewelry.

The professor appears to have slid in under the wire: the Trump administration has cut nearly 1400 grants from the NSF worth nearly $1 billion, and promises in its new budget to cut its funding by 55 percent.

This week marks the 75th anniversary of its founding on May 10, 1950.

Article Topics

biometric authentication  |  biometrics  |  continuous authentication  |  heartbeat biometrics  |  LSU  |  photoplethysmogram (PPG)  |  research and development  |  wearables