• DocumentCode
    149934
  • Title

    Application design and performance evaluation for multiprocessor sensor nodes

  • Author

    Jingyao Zhang ; Zhenhe Pan ; Schaumont, Patrick ; Yaling Yang

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Virginia Polytech. Inst. & State Univ., Blacksburg, VA, USA
  • fYear
    2014
  • fDate
    6-9 April 2014
  • Firstpage
    3272
  • Lastpage
    3277
  • Abstract
    Currently, the main components of most sensor nodes are a microcontroller and a radio. Their real-time and peak performance would be a bottleneck when executing computation-intensive tasks. Also, energy consumption may be high due to the long task-execution time. Several works [2]-[6] demonstrate that adding a coprocessor would be a solution to the problems. So far, no work have been done to analyze the performance of a multiprocessor sensor nodes with a FPGA coprocessor. In this paper, a design flow for heterogeneous multiprocessor sensor nodes is provided. Then, the performance comparison between multiprocessor and single processor sensor node´s time and energy consumption is provided by executing applications on our in-house designed multiprocessor sensor node on a real testbed. The testbed results show that multiprocessor sensor node can either increase sensor node´s execution speed or reduce the sensor node´s energy consumption when the sensor node executes computation-intensive tasks.
  • Keywords
    coprocessors; energy consumption; field programmable gate arrays; microcontrollers; multiprocessing systems; wireless sensor networks; FPGA coprocessor; computation-intensive tasks; energy consumption; microcontroller; multiprocessor sensor nodes; Coprocessors; Energy consumption; Field programmable gate arrays; Hardware; Microcontrollers; Wireless communication; Wireless sensor networks;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Wireless Communications and Networking Conference (WCNC), 2014 IEEE
  • Conference_Location
    Istanbul
  • Type

    conf

  • DOI
    10.1109/WCNC.2014.6953072
  • Filename
    6953072