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:20260414T162848Z
UID:AEBD8375-3A62-4997-9AA8-9C7B3783DF56
DTSTART;TZID=America/New_York:20260519T180000
DTEND;TZID=America/New_York:20260519T193000
DESCRIPTION:Abstract:\n\nThis talk presents recent advances in WiFi sensing
  for human activity recognition (HAR)\, demonstrating how existing wireles
 s communication infrastructure can be leveraged as a powerful sensing moda
 lity. By exploiting Channel State Information (CSI) readily available in c
 ommodity WiFi devices\, it is possible to infer both large-scale and fine-
 grained human activities without requiring wearable sensors or dedicated h
 ardware. The presentation begins with an overview of the emerging paradigm
  of Integrated Sensing and Communication (ISAC)\, including developments s
 uch as IEEE 802.11bf and future 6G systems\, where communication signals a
 re repurposed for environmental sensing. It then introduces signal process
 ing techniques for extracting meaningful features from CSI\, including tim
 e-frequency analysis and Doppler-based representations that capture motion
  dynamics. A key contribution discussed in the talk is the use of lightwei
 ght and scalable machine learning approaches\, such as random convolutiona
 l kernels and deep neural networks\, for efficient end-to-end activity rec
 ognition.\n\nSpeaker(s): Professor Shahrokh Valaee\, \n\nRoom: EV003-309\,
  Bldg: Electrical & Computer Engineering Department EV\, Concordia Univers
 ity\, 1515 Ste. Catherine West\, MONTREAL\, Quebec\, Canada\, H3G 1M8
LOCATION:Room: EV003-309\, Bldg: Electrical & Computer Engineering Departme
 nt EV\, Concordia University\, 1515 Ste. Catherine West\, MONTREAL\, Quebe
 c\, Canada\, H3G 1M8
ORGANIZER:anader.benyamin@ieee.org
SEQUENCE:48
SUMMARY:WiFi Sensing for Human Activity Recognition
URL;VALUE=URI:https://events.vtools.ieee.org/m/554912
X-ALT-DESC:Description: <br /><p><strong>Abstract:</strong></p>\n<p style="
 text-align: justify\;">This talk presents recent advances in <strong><span
  style="mso-fareast-font-family: 'Times New Roman'\; mso-fareast-theme-fon
 t: major-fareast\;">WiFi sensing for human activity recognition (HAR)</spa
 n></strong><strong>\,</strong> demonstrating how existing wireless communi
 cation infrastructure can be leveraged as a powerful sensing modality. By 
 exploiting <strong><span style="mso-fareast-font-family: 'Times New Roman'
 \; mso-fareast-theme-font: major-fareast\;">Channel State Information (CSI
 )</span></strong> readily available in commodity WiFi devices\, it is poss
 ible to infer both large-scale and fine-grained human activities without r
 equiring wearable sensors or dedicated hardware. The presentation begins w
 ith an overview of the emerging paradigm of <strong><span style="mso-farea
 st-font-family: 'Times New Roman'\; mso-fareast-theme-font: major-fareast\
 ;">Integrated Sensing and Communication (ISAC)</span></strong><strong>\,</
 strong> including developments such as IEEE 802.11bf and future 6G systems
 \, where communication signals are repurposed for environmental sensing. I
 t then introduces signal processing techniques for extracting meaningful f
 eatures from CSI\, including time-frequency analysis and Doppler-based rep
 resentations that capture motion dynamics. A key contribution discussed in
  the talk is the use of<strong> </strong><span style="mso-fareast-font-fam
 ily: 'Times New Roman'\; mso-fareast-theme-font: major-fareast\;">l</span>
 <span style="mso-fareast-font-family: 'Times New Roman'\; mso-fareast-them
 e-font: major-fareast\;">ightweight and scalable machine learning approach
 es</span>\, such as random convolutional kernels and deep neural networks\
 , for efficient end-to-end activity recognition.</p>
END:VEVENT
END:VCALENDAR