Dissertation/Thesis Abstract

A Smart Design Framework for a Novel Reconfigurable Multi-Processor Systems-on-Chip (ASREM) Architecture
by Dutta, Anandi, Ph.D., University of Louisiana at Lafayette, 2016, 127; 10163349
Abstract (Summary)

In this work, a smart reconfigurable Multi-Processor Systems-on-Chip (ASREM) has been introduced. The researcher examined a reconfigurable MPSoC architecture that incorporates one Processor-FPGA core to ensure flexibility and better design parameters. The objective of this work is to initiate a smart system that introduces this philosophy: the system would build a better system by itself. The system would learn about usage statistics by using an android application with the help of support vector machine (SVM) algorithm. Accomplishing this, the system would form a decision function from the SVM classifier to improve usability. According to the user’s preference, it would re-design the reconfigurable MPSoC to ensure customized and superior user experience. A Reconfigurable MPSoC architecture and task scheduling mechanism has been introduced to enable the design. An image processing task has been performed as a case-study on a FPGA-SoC platform and GPU as a proof of concept to ensure current standards. The novel and competent approach of designing system for hand-held devices adopted in this study enables customization of the device after manufacturing.

Indexing (document details)
Advisor: Bayoumi, Magdy
Commitee: Kumar, Ashok, Zhao, Danella
School: University of Louisiana at Lafayette
Department: Computer Engineering
School Location: United States -- Louisiana
Source: DAI-B 78/04(E), Dissertation Abstracts International
Subjects: Computer Engineering
Keywords: Android application, Association rule mining, Multi-processor systems-on-chip, Reconfigurable computing, Support vector machine
Publication Number: 10163349
ISBN: 978-1-369-18041-1
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy