• DocumentCode
    2549640
  • Title

    An adaptive Viterbi decoder based on FPGA dynamic reconfiguration technology

  • Author

    Xiang-Ju, Qin ; Ming-Cheng, Zhu ; Zhong-Yi, Wei ; Du, Chao

  • Author_Institution
    Coll. of Inf. Eng., Shenzhen Univ., China
  • fYear
    2004
  • fDate
    6-8 Dec. 2004
  • Firstpage
    315
  • Lastpage
    318
  • Abstract
    The Viterbi algorithm is the most popular decoding algorithm in communication systems, but its hardware complexity increases in the order of 2k with the constraint length K. This work proposed an improved Viterbi algorithm, called adaptive Viterbi algorithm (AVA), only keeping some of the best or most-likely states, so as to reduce the amount of computations and storage memories needed. Using run-time dynamic reconfiguration technology, the parameters of the decoder can be dynamically reconfigured in response to channel noise. Simulation results and careful analysis of the decoder stucture demonstrate that total dynamic reconfiguration technology can be used to improve performance of adaptive Viterbi decoder. Comparison with one Viterbi decoder that has implemented in Xilinx FPGA by us, it is also shown that adaptive Viterbi decoder can save hardware resources by 20%.
  • Keywords
    Viterbi decoding; field programmable gate arrays; reconfigurable architectures; FPGA dynamic reconfiguration technology; Xilinx FPGA; adaptive Viterbi algorithm; adaptive Viterbi decoder; channel noise; communication systems; decoder stucture; hardware complexity; hardware resources; run-time dynamic reconfiguration technology; storage memories; Analytical models; Chaotic communication; Computational modeling; Costs; Decoding; Educational institutions; Field programmable gate arrays; Hardware; Runtime; Viterbi algorithm;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Field-Programmable Technology, 2004. Proceedings. 2004 IEEE International Conference on
  • Print_ISBN
    0-7803-8651-5
  • Type

    conf

  • DOI
    10.1109/FPT.2004.1393288
  • Filename
    1393288