A technique for energy versus quality of service trade-off for MPEG-2 decoder

Krishnan Srinivasan, Vijay Ramamurthi, Karam S. Chatha

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Scopus citations

Abstract

Many implementations of MPEG-2 decoder permit a certain degradation in PSNR due to human audio/video perceptory imperfections. This paper presents a technique for exploiting the acceptable degradation in PSNR towards energy minimization by application of dynamic voltage management techniques. The technique is based on a two phase approach that includes static characterization and dynamic adaptation. We present a novel model that relates the type of the frame, size of the frame, and number of macroblocks processed to the PSNR of the stream. The model is utilized for characterizing the benchmark streams. The results of static characterizations are then applied towards general streams by dynamic adaptations. The experimental results demonstrate that for an acceptable PSNR of 30 to 32 dB the technique results in energy savings from 24% to over 85%.

Original languageEnglish (US)
Title of host publicationProceedings - IEEE Computer Society Annual Symposium on VLSI
Subtitle of host publicationEmerging Trends in VLSI Systems Design
EditorsA. Smailagic, M. Bayoumi
Pages313-316
Number of pages4
DOIs
StatePublished - Sep 24 2004
EventProceedings - IEEE Computer Society Annual Symposium on VLSI: Emerging Trends in VLSI Systems Design - Lafayette, LA, United States
Duration: Feb 19 2004Feb 20 2004

Publication series

NameProceedings - IEEE Computer Society Annual Symposium on VLSI: Emerging Trends in VLSI Systems Design

Other

OtherProceedings - IEEE Computer Society Annual Symposium on VLSI: Emerging Trends in VLSI Systems Design
Country/TerritoryUnited States
CityLafayette, LA
Period2/19/042/20/04

ASJC Scopus subject areas

  • General Engineering

Fingerprint

Dive into the research topics of 'A technique for energy versus quality of service trade-off for MPEG-2 decoder'. Together they form a unique fingerprint.

Cite this