• DocumentCode
    2086346
  • Title

    Power estimation methodology for VLIW Digital Signal Processors

  • Author

    Ibrahim, Mostafa E A ; Rupp, Markus ; Fahmy, Hossam A H

  • Author_Institution
    Inst. of Commun. & RF Eng., Vienna Univ. of Technol., Vienna
  • fYear
    2008
  • fDate
    26-29 Oct. 2008
  • Firstpage
    1840
  • Lastpage
    1844
  • Abstract
    In this contribution the modeling of power consumption for the VLIW processor TMS320C6416T is presented taking into account typical software algorithms in signal and image processing. The modeling is performed at the functional level making this approach distinctly different from other modeling approaches in low level technique. This means that the power consumption can be identified at an early stage in the design process, enabling the designer to explore different hardware architectures and software algorithms. Some typical signal and image processing algorithms are used for the purpose of validating the proposed model. The estimated power consumption is compared to the physically measured power consumption, achieving a very low resulting average estimation error of 1.05% and a maximum estimation error of only 3.3%.
  • Keywords
    digital signal processing chips; multiprocessing systems; VLIW digital signal processors; VLIW processor TMS320C6416T; average estimation error; image processing; maximum estimation error; power consumption; power estimation methodology; signal processing; software algorithms; very long instruction word; Algorithm design and analysis; Digital signal processors; Energy consumption; Estimation error; Image processing; Process design; Signal processing; Signal processing algorithms; Software algorithms; VLIW;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signals, Systems and Computers, 2008 42nd Asilomar Conference on
  • Conference_Location
    Pacific Grove, CA
  • ISSN
    1058-6393
  • Print_ISBN
    978-1-4244-2940-0
  • Electronic_ISBN
    1058-6393
  • Type

    conf

  • DOI
    10.1109/ACSSC.2008.5074746
  • Filename
    5074746