Fulltext available Open Access
License: 
Title: Implementation and Evaluation of Optimization Strategies for Audio Signal Convolution
Language: English
Authors: Heidtmann, Stefan 
Issue Date: 6-Sep-2019
Abstract: 
The focus of the thesis is the implmentation of the discrete convolution and the uniformlypartitioned convolution on the CPU with C++ and on the GPU with OpenCL. When comparing the performance of algorithms on different processor types, it is uttermost importance to exhaust their processing power. To obtain the maximum performanceoptimization of the code is required. There are several types of strategies available for optimizing the code. Their effect on the performance is evaluate to show which are absolute necessary and which have only a negligible effect.

Der Fokus dieser Thesis ist die Evaluierung von C++ und OpenCL Implementierung der diskreten und der partitionierten schnellen Faltung. Für eine aussagekräftigen Evaluierung der Rechenleistung der Algorithmen auf der CPU und der GPU muss die Leistung der Prozessoren ausgelasten werden. Um das zu erreichen wurden verschiedene Optimierungsstrategien implementiert um deren Einfluss auf Leistung getestet.
URI: http://hdl.handle.net/20.500.12738/8857
Institute: Department Informatik 
Type: Thesis
Thesis type: Master Thesis
Advisor: Fohl, Wolfgang 
Referee: Meisel, Andreas 
Appears in Collections:Theses

Files in This Item:
File Description SizeFormat
Thesis_Stefan_Heidtmann.pdf2.06 MBAdobe PDFView/Open
Show full item record

Page view(s)

206
checked on Dec 25, 2024

Download(s)

411
checked on Dec 25, 2024

Google ScholarTM

Check

HAW Katalog

Check

Note about this record


Items in REPOSIT are protected by copyright, with all rights reserved, unless otherwise indicated.