BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20260308T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20261101T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260328T022255Z UID:16AF5362-D628-4261-8373-AA2A6F2D1087 DTSTART;TZID=America/New_York:20260327T170000 DTEND;TZID=America/New_York:20260327T190000 DESCRIPTION:The rapid expansion of the Internet of Things (IoT) is transfor ming sensing infrastructures into large-scale distributed cyber-physical s ystems. However\, conventional IoT architectures remain largely cloud-cent ric\, relying on continuous data transmission and centralized processing. Such designs introduce fundamental challenges in energy consumption\, comm unication bandwidth\, latency\, and privacy exposure\, particularly as the number of connected devices continues to grow. This talk introduces the e merging paradigm of the Artificial Intelligence of Things (AIoT)\, where i ntelligence is embedded directly within sensing devices through edge compu ting and TinyML techniques. By enabling machine learning inference on reso urce-constrained microcontrollers\, AIoT systems can transform traditional sensors into autonomous\, context-aware intelligent nodes capable of perf orming real-time decision making. Beyond presenting representative AIoT sy stems and applications\, this talk also examines broader research challeng es shaping the future of AIoT systems\, including platform fragmentation\, intelligent sensor deployment\, secure federated learning\, and the need for autonomous and scalable IoT architectures. In particular\, we discuss how edge-intelligent AIoT systems can support emerging digital twin framew orks\, enabling event-driven monitoring and scalable healthcare applicatio ns while reducing continuous data streaming. The presentation concludes wi th an overview of modern development pipelines for deploying machine learn ing models on edge devices using frameworks such as TinyML toolchains\, hi ghlighting future research opportunities in distributed intelligent sensin g systems.\n\nSpeaker(s): Minhee\, \n\nBldg: Scullen Conference Room (Pang born Hall 110)\, The Catholic University of America\, 620 Michigan Avenue N.E\, 20064\, District of Columbia\, United States\, 20064\, Virtual: http s://events.vtools.ieee.org/m/549200 LOCATION:Bldg: Scullen Conference Room (Pangborn Hall 110)\, The Catholic U niversity of America\, 620 Michigan Avenue N.E\, 20064\, District of Colum bia\, United States\, 20064\, Virtual: https://events.vtools.ieee.org/m/54 9200 ORGANIZER:krishna.kandi@ieee.org SEQUENCE:48 SUMMARY:Edge-Intelligent AIoT Systems: Enabling Scalable and Privacy-Aware Intelligent Applications URL;VALUE=URI:https://events.vtools.ieee.org/m/549200 X-ALT-DESC:Description: <br /><p dir="ltr">The rapid expansion of the Inter net of Things (IoT) is transforming sensing infrastructures into large-sca le distributed cyber-physical systems. However\, conventional IoT architec tures remain largely cloud-centric\, relying on continuous data transmissi on and centralized processing. Such designs introduce fundamental challeng es in energy consumption\, communication bandwidth\, latency\, and privacy exposure\, particularly as the number of connected devices continues to g row. This talk introduces the emerging paradigm of the Artificial Intellig ence of Things (AIoT)\, where intelligence is embedded directly within sen sing devices through edge computing and TinyML techniques. By enabling mac hine learning inference on resource-constrained microcontrollers\, AIoT sy stems can transform traditional sensors into autonomous\, context-aware in telligent nodes capable of performing real-time decision making. Beyond pr esenting representative AIoT systems and applications\, this talk also exa mines broader research challenges shaping the future of AIoT systems\, inc luding platform fragmentation\, intelligent sensor deployment\, secure fed erated learning\, and the need for autonomous and scalable IoT architectur es. In particular\, we discuss how edge-intelligent AIoT systems can suppo rt emerging digital twin frameworks\, enabling event-driven monitoring and scalable healthcare applications while reducing continuous data streaming . The presentation concludes with an overview of modern development pipeli nes for deploying machine learning models on edge devices using frameworks such as TinyML toolchains\, highlighting future research opportunities in distributed intelligent sensing systems.</p>\n<p>&nbsp\;</p> END:VEVENT END:VCALENDAR