Computer Engineering Department,
San Jose State University

"ECG Biometric Systems: Evolution, Challenges and Future Directions"

Dec 3, 2021 Schedule:

2:30 to 3:30 PM Eastern Time (US and Canada)

Meeting # 94535057477


As the Covid-19 pandemic grips the world, people are practicing social distancing to prevent its spread. As a result, physical contact is discouraged. Hence, biometric authentication can pave the way for the future, ensuring safety and function. Therefore, the increased usage of biometric technologies raises legitimate concerns about security, privacy, and effectiveness. Despite all the benefits, the biometric traits are permanent and difficult to revoke if compromised, unlike keys or passwords. In addition, unprotected storage of biometric reference data poses severe privacy threats such as identity theft and cross-matching. This talk presents a resilient solution for biometric systems using ECG signals. First, I will present a robust, and accurate ECG-based biometric authentication using a ring Residual Neural architecture. Second, the spoofing attack on ECG biometric will be presented using the GAN and synthetic ECG for both cross-subject attacks and cross-device attacks. Third, we explore countermeasures that utilize ECG signal characteristics such as heart rate variability and PPG signal to detect and reject fake samples. Finally, I will address the vulnerability of biometric systems to physical side-channel attack resulting in template theft/privacy, illegal access based on combination of biometrics with two recent advances in hardware security -physically unclonable functions (PUFs) and hardware obfuscation.