BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Montreal
BEGIN:DAYLIGHT
DTSTART:20220313T030000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20221106T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20220908T211455Z
UID:83AA074E-B935-4E42-8F2F-860D7098A23B
DTSTART;TZID=America/Montreal:20220908T123000
DTEND;TZID=America/Montreal:20220908T150000
DESCRIPTION::  This talk discusses AI/ML-based techniques to model and miti
 gates nonlinear distortion and hardware impairments of Radios intended for
  5G and beyond wireless networks. The talk focuses on the neural network (
 NN) based digital pre-distortion techniques for linearizing multi-band\, a
 nd multiple-input multiple-output (MIMO) phased array transmitters. We wil
 l review the most popular NN-based behavioral models using Deep Neural Net
 works\, Augmented Neural Networks\, and Shallow Neural Networks. The talk 
 will concentrate on various aspects of DPD implementation using AI/ML tech
 niques and discuss the evolution of NN-based DPD modeling techniques for d
 ynamic nonlinear systems. We will also present State-of-the-art approaches
  such as Real-Valued Focused Time-Delay Neural Network. (RVFTDNN) and Conv
 olutional Neural Networks (CNN) and discuss their suitability for in-field
  applications. These models' performance will be assessed in terms of thei
 r capability to mitigate the transmitter's distortion and hardware impairm
 ents such as antenna's crosstalk\, PA's nonlinearity\, I/Q imbalance\, and
  dc offset for multi-band and MIMO applications for 5G and 6G applications
 .\n\nCo-sponsored by: Staracom\n\nSpeaker(s): Prof. Fadhel Ghannouchi\, \n
 \nRoom: M-2110\, Bldg: Pavillons Lassonde\, 2500 Chemin de Polytechnique\,
  Montreal\, Quebec\, Canada\, H3T1J4
LOCATION:Room: M-2110\, Bldg: Pavillons Lassonde\, 2500 Chemin de Polytechn
 ique\, Montreal\, Quebec\, Canada\, H3T1J4
ORGANIZER:mohammad.sharawi@polymtl.ca
SEQUENCE:4
SUMMARY:AI driven Digital Predistortion Techniques for Multi-Band/MIMO Wire
 less phased Array Transmitters
URL;VALUE=URI:https://events.vtools.ieee.org/m/322009
X-ALT-DESC:Description: <br /><p><strong>: </strong>&nbsp\;This talk discus
 ses AI/ML-based techniques to model and mitigates nonlinear distortion and
  hardware impairments of Radios intended for 5G and beyond wireless networ
 ks. The talk focuses on the neural network&nbsp\; (NN) based digital pre-d
 istortion techniques for linearizing multi-band\, and multiple-input multi
 ple-output&nbsp\; (MIMO) phased array transmitters. We will review the mos
 t popular NN-based behavioral models using Deep Neural Networks\, Augmente
 d Neural Networks\, and Shallow Neural Networks. The talk will concentrate
  on various aspects of DPD implementation using AI/ML techniques and discu
 ss the evolution of NN-based DPD modeling techniques for dynamic nonlinear
  systems. We will also present State-of-the-art approaches such as Real-Va
 lued Focused Time-Delay Neural Network. (RVFTDNN) and Convolutional Neural
  Networks&nbsp\; (CNN) and discuss their suitability for in-field applicat
 ions. These models' performance will be assessed in terms of their capabil
 ity to mitigate the transmitter's distortion and hardware impairments such
  as antenna's crosstalk\, PA's nonlinearity\, I/Q imbalance\, and dc offse
 t for multi-band and MIMO applications for 5G and 6G applications.</p>
END:VEVENT
END:VCALENDAR