Low-power Visual Pattern Classification in Analog VLSI

منو

" Low-power Visual Pattern Classification in Analog VLSI "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	150186
Doc. No	:	ET21978
Main Entry	:	Seth Bridges
Title Proper	:	Low-power Visual Pattern Classification in Analog VLSI
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	Today's mobile devices - cell phones, hand-held computers, sensor networks, etc. - re-quire large amounts of computational power to realize applications such as adaptive filtering,noise cancellation, activity recognition, biometric identification, and robotic control. Theaddition of cameras to many of these devices further increases this computational load byadding image processing tasks. Unfortunately, the general-purpose digital microcontrollerson which these devices compute consume tens or hundreds of milliwatts - power levels thatlimit battery operation to several days. This thesis seeks to reduce the power and area con-sumption of these algorithms by employing analog and mixed-signal circuits that computecontinuously-valued quantities.,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL45174.pdf
Title and statement of responsibility and	:	Low-power Visual Pattern Classification in Analog VLSI [Thesis]