BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260610T190501EDT-9928IN4phw@132.216.98.100
DTSTAMP:20260610T230501Z
DESCRIPTION:Abstract\n\nIn this thesis\, we present the concept of subband 
 decomposition CNN architecture that can learn the subband decomposition co
 efficients from a training dataset through the automated end-to-end traini
 ng framework of a neural network. We present a few variations of the archi
 tecture that provide trade-off between computational cost and performance.
  Further\, we analyze the properties of the subband decomposition architec
 tures and compare their performance with present state-of-the-art architec
 tures. We observe the subband decomposition CNN architecture to perform at
  part of the state-of-the-art CNN architectures at a fraction of the compu
 tation cost. We then apply the concept of subband decomposed CNN architect
 ure to the problem of image segmentation of lumen and media in IVUS images
 . We compare results with state-of-the-art CNN architectures and easily id
 entify that not only does the subband decomposition architecture outperfor
 m the competition\, but does that at a fraction of the computation cost.\n
 \nFurther\, we apply the concept of subband decomposed CNN architectures t
 o the problem of channel equalization and synchronization in OFDM radio re
 ceiver subsystems. To understand the practicability of the subband decompo
 sed CNN architecture-based receiver synchronization subsystem\, we impleme
 nt it on an FPGA and analyze the hardware resource utilization for real-li
 fe applicability.\n	 \n
DTSTART:20240408T160000Z
DTEND:20240408T180000Z
LOCATION:Room 603\, McConnell Engineering Building\, CA\, QC\, Montreal\, H
 3A 0E9\, 3480 rue University
SUMMARY:PhD defence of Pavel Sinha – Properties and Applications of a Struc
 turally Regularized CNN Architecture via Adaptive Subband Decomposition
URL:https://www.mcgill.ca/ece/channels/event/phd-defence-pavel-sinha-proper
 ties-and-applications-structurally-regularized-cnn-architecture-356575
END:VEVENT
END:VCALENDAR