BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20250309T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20241103T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20250116T231653Z UID:7964E4BA-7231-4AFD-8D8B-CAF755FC0C51 DTSTART;TZID=America/New_York:20250115T110000 DTEND;TZID=America/New_York:20250115T123000 DESCRIPTION:Speaker:\n\nDr. Bilal Taha\, Postdoctoral Fellow\, University o f Toronto\n\nAbstract:\n\nConventional biometric systems\, such as facial recognition and fingerprint authentication\, are widely adopted in modern security frameworks. However\, these technologies suffer from limitations\ , including susceptibility to spoofing and demographic biases. Photoplethy smography (PPG) signals offer a promising alternative for biometric user a uthentication\, addressing these challenges through their inherent resilie nce and unique physiological characteristics. This talk presents advanceme nts to establish the viability of PPG-based authentication. First\, a nove l approach combining signal processing techniques with neural network arch itectures is introduced\, achieving accurate and reliable user authenticat ion from PPG signals. Second\, generative adversarial networks (GANs) are utilized to synthesize realistic PPG signals\, enhancing the robustness an d generalizability of the system by augmenting training datasets. Finally\ , the feasibility of extracting PPG signals using mobile devices rather th an conventional pulse oximeters is demonstrated\, detailing both the signa l acquisition process and the development of a practical mobile-based auth entication model.\n\nSpeaker Biography:\n\nBilal Taha is currently a postd octoral fellow at the University of Toronto. He received his Ph.D. degree from the Department of Electrical and Computer Engineering\, University of Toronto\, in 2023. He was a Postgraduate Research Affiliate with the Vect or Institute as well as a Research Intern with the ICUBE Laboratory\, Univ ersity of Strasbourg\, and Amazon Alexa AI. His research interests include security systems\, biometrics\, and machine learning.\n\nSpeaker Website and Google Scholar Profile:\n\nhttps://www.comm.utoronto.ca/~btaha/\nhttps ://scholar.google.com/citations?hl=en&user=11cMSWsAAAAJ&view_op=list_works &sortby=pubdate\n\nCo-sponsored by: Royal Military College of Canada\, Dep t. of Elec. & Comp. Eng.\n\nSpeaker(s): Bilal\n\nVirtual: https://events.v tools.ieee.org/m/460732 LOCATION:Virtual: https://events.vtools.ieee.org/m/460732 ORGANIZER:chan-f@rmc.ca SEQUENCE:27 SUMMARY:IEEE Talk: Beyond Conventional Biometrics: Advancing User Authentic ation with PPG Signals URL;VALUE=URI:https://events.vtools.ieee.org/m/460732 X-ALT-DESC:Description: <br /><p><strong>Speaker: </strong></p>\n<p>Dr. Bil al Taha\, Postdoctoral Fellow\, University of Toronto</p>\n<p><strong>Abst ract</strong>:</p>\n<p>Conventional biometric systems\, such as facial rec ognition and fingerprint authentication\, are widely adopted in modern sec urity frameworks. However\, these technologies suffer from limitations\, i ncluding susceptibility to spoofing and demographic biases. Photoplethysmo graphy (PPG) signals offer a promising alternative for biometric user auth entication\, addressing these challenges through their inherent resilience and unique physiological characteristics. This talk presents advancements to establish the viability of PPG-based authentication. First\, a novel a pproach combining signal processing techniques with neural network archite ctures is introduced\, achieving accurate and reliable user authentication from PPG signals. Second\, generative adversarial networks (GANs) are uti lized to synthesize realistic PPG signals\, enhancing the robustness and g eneralizability of the system by augmenting training datasets. Finally\, t he feasibility of extracting PPG signals using mobile devices rather than conventional pulse oximeters is demonstrated\, detailing both the signal a cquisition process and the development of a practical mobile-based authent ication model.</p>\n<p><strong>Speaker Biography:</strong></p>\n<p>Bilal T aha is currently a postdoctoral fellow at the University of Toronto. He re ceived his Ph.D. degree from the Department of Electrical and Computer Eng ineering\, University of Toronto\, in 2023. He was a Postgraduate Research Affiliate with the Vector Institute as well as a Research Intern with the ICUBE Laboratory\, University of Strasbourg\, and Amazon Alexa AI. His re search interests include security systems\, biometrics\, and machine learn ing.</p>\n<p>Speaker Website and Google Scholar Profile:</p>\n<p><a href=" https://www.comm.utoronto.ca/~btaha/">https://www.comm.utoronto.ca/~btaha/ </a>&nbsp\;<br><a href="https://scholar.google.com/citations?hl=en&amp\;us er=11cMSWsAAAAJ&amp\;view_op=list_works&amp\;sortby=pubdate">https://schol ar.google.com/citations?hl=en&amp\;user=11cMSWsAAAAJ&amp\;view_op=list_wor ks&amp\;sortby=pubdate</a>&nbsp\;</p> END:VEVENT END:VCALENDAR