• DocumentCode
    3087460
  • Title

    Research on Efficiency of Signal Processing on Embedded Multicore System

  • Author

    Liu, Tao ; Ji, Zhenzhou ; Wang, Qing ; Zhu, Suxia

  • Author_Institution
    Sch. of Comput. Sci. & Technol., Harbin Inst. of Technol., Harbin, China
  • fYear
    2010
  • fDate
    17-19 Sept. 2010
  • Firstpage
    907
  • Lastpage
    911
  • Abstract
    A prototype signal processing system of FPGA embedded multicore platform is presented to evaluate the performance and efficiency of signal processing on embedded multicore platform. The multicore processing algorithms of GSM and ADPCM are implemented. For the requirements in embedded applications, we point out the performance difference that come from the various processor core number and bus width, the impact from bus bandwidth also has been analyzed. It has been clearly shown that shared storage multicore structure fits embedded multicore system very well by optimized configuration for processor cores, bus structure and storage resources. The prototype system shows high performance in signal data processing, and the OpenMP multithread model is convenient for development of embedded multicore applications with good scalability.
  • Keywords
    application program interfaces; field programmable gate arrays; microprocessor chips; multi-threading; multiprocessing systems; performance evaluation; signal processing; ADPCM; FPGA embedded multicore platform; GSM; OpenMP multithread model; bus structure; performance evaluation; processor cores; shared storage multicore structure; signal processing system; storage resources; Algorithm design and analysis; Biological system modeling; Computational modeling; GSM; Multicore processing; Signal processing; Signal processing algorithms; ADPCM; FPGA; GSM; OpenMP; embedded multicore;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Pervasive Computing Signal Processing and Applications (PCSPA), 2010 First International Conference on
  • Conference_Location
    Harbin
  • Print_ISBN
    978-1-4244-8043-2
  • Electronic_ISBN
    978-0-7695-4180-8
  • Type

    conf

  • DOI
    10.1109/PCSPA.2010.224
  • Filename
    5635845