License: Creative Commons Attribution 4.0 International license (CC BY 4.0)
When quoting this document, please refer to the following
DOI: 10.4230/DagSemProc.07041.8
URN: urn:nbn:de:0030-drops-11120
URL: http://dagstuhl.sunsite.rwth-aachen.de/volltexte/2007/1112/
Go to the corresponding Portal


Devos, Harald ; Eeckhaut, Hendrik ; Christiaens, Mark ; Stroobandt, Dirk

Energy Scalability and the RESUME Scalable Video Codec

pdf-format:
07041.DevosHarald.Paper.1112.pdf (0.4 MB)


Abstract

In the context of the RESUME-project a scalable wavelet-based video decoder was built to demonstrate the benefits of reconfigurable hardware for scalable applications. emph{Scalable} video means that the quality of service (QoS), i.e., the frame rate, resolution, color depth, ldots of the decoded video can easily be changed by only decoding those parts of the video stream that contribute to the desired QoS. With the emergence of high-performance FPGAs (Field Programmable Gate Array), both the required performance for real-time decoding and flexibility, by allowing reconfiguration, are offered.

Since the amount of calculations scales with the QoS, energy dissipation is expected to scale similarly. To investigate the relation between QoS and energy dissipation we actually measured the energy dissipation of a scalable video decoder implementation on a FPGA. The measurements show how dissipation effectively scales with the QoS, but also depends on the decoded data and the used design method. This is illustrated by comparing two different implementations of the inverse discrete wavelet transform (IDWT).



BibTeX - Entry

@InProceedings{devos_et_al:DagSemProc.07041.8,
  author =	{Devos, Harald and Eeckhaut, Hendrik and Christiaens, Mark and Stroobandt, Dirk},
  title =	{{Energy Scalability and the RESUME Scalable Video Codec}},
  booktitle =	{Power-aware Computing Systems},
  pages =	{1--12},
  series =	{Dagstuhl Seminar Proceedings (DagSemProc)},
  ISSN =	{1862-4405},
  year =	{2007},
  volume =	{7041},
  editor =	{Luca Benini and Naehyuck Chang and Ulrich Kremer and Christian W. Probst},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/opus/volltexte/2007/1112},
  URN =		{urn:nbn:de:0030-drops-11120},
  doi =		{10.4230/DagSemProc.07041.8},
  annote =	{Keywords: Wavelet-based Scalable Video, Energy Measurement, Hardware Generation, FPGA}
}

Keywords: Wavelet-based Scalable Video, Energy Measurement, Hardware Generation, FPGA
Collection: 07041 - Power-aware Computing Systems
Issue Date: 2007
Date of publication: 30.07.2007


DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI