• DocumentCode
    3476083
  • Title

    Low power DCT using highly scalable multipliers

  • Author

    Castellanos, Ricardo ; Kalva, Hari ; Shankar, Ravi

  • Author_Institution
    Dept. of Comput. Sci. & Eng., Florida Atlantic Univ., Boca Raton, FL, USA
  • fYear
    2009
  • fDate
    7-10 Nov. 2009
  • Firstpage
    1925
  • Lastpage
    1928
  • Abstract
    Low power consumption in computing systems is a key requirement for devices such as cell phones and cameras. In this paper we present a low power DCT implementation using a highly scalable multiplier. This paper focuses on IDCT with playback applications such as digital photo displays. The proposed solution exploits the fact that the size of the multiplications varies per stage in a multistage IDCT implementation and configuring multipliers to match the needs of each stage saves power. Results are compared with Wallace and array multipliers. We show that using a scalable multiplier and dynamically reconfiguring the width of the multiplier leads to significant power savings (over 72%) with negligible degradation in decoded image quality.
  • Keywords
    decoding; discrete cosine transforms; image coding; Wallace multipliers; array multipliers; camera; cell phones; digital photo displays; discrete cosine transforms; highly scalable multipliers; image quality decoding; low power DCT; low power consumption; Decoding; Discrete cosine transforms; Discrete transforms; Energy consumption; Image storage; Matrix decomposition; Quantization; Signal processing algorithms; Sparse matrices; Video compression;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Image Processing (ICIP), 2009 16th IEEE International Conference on
  • Conference_Location
    Cairo
  • ISSN
    1522-4880
  • Print_ISBN
    978-1-4244-5653-6
  • Electronic_ISBN
    1522-4880
  • Type

    conf

  • DOI
    10.1109/ICIP.2009.5413517
  • Filename
    5413517