Validation of New Biometrics for Human Identification

• Activity Number: 358
• Type: Contributed
• Date/Time: Wednesday, August 14, 2002 : 2:00 PM to 3:50 PM
• Sponsor: Section on Physical & Engineering Sciences*
• Author(s): John Irvine*+ and Steven Israel and Andrew Cheng and Mark Wiederhold and Brenda Wiederhold and Shannon McGehee
• Affiliation(s): SAIC and SAIC and SAIC and SAIC and Virtual Reality Medical Center and Virtual Reality Medical Center
• Keywords: Human identification ; signal processing ; Biometrics

Abstract

Numerous biometric techniques exist for verifying the identity of individuals. Traditional biometrics include fingerprint, face, and iris recognition. This paper examines recent developments of new biometric techniques based on observation of physiological functions related to circulation. These biometrics exploit information about dynamic physiological processes that can be interrogated using non-imaging sensors. In particular, we present biometric techniques based on electrocardiogram and pulse oximetry. The development and validation of these new biometric techniques pose some interesting challenges for design of the experiments and data analysis. Since a person's heart rate can vary with mental and emotional state, we developed a data collection protocol in which subjects perform a variety of tasks designed to elicit varying levels of stress or excitement. In developing and validating the new biometrics, it is necessary to identify features in the physiometric signals that are unique to individuals, but invariant to mental and emotional state. In this paper, we present the experiment procedures, summarize the data and processing, and present initial performance results.