• DocumentCode
    347881
  • Title

    A high throughput architecture for channel equalization based on a neural network using a wave pipeline method

  • Author

    Morin, Frédéric ; Vidal, Martin ; Massicotte, Daniel

  • Author_Institution
    Dept. of Electr. Eng., Quebec Univ., Trois-Rivieres, Que., Canada
  • Volume
    1
  • fYear
    1999
  • fDate
    9-12 May 1999
  • Firstpage
    560
  • Abstract
    The use of a wave pipelining method for the design of a systolic architecture dedicated to channel equalization is proposed. A description is given of the piecewise linear multilayer neural network (PL-MNN) algorithm and the architecture. To improve the throughput of the architecture we propose a wave-pipelined version of the multiplier-accumulator (MAC) the presents the bottleneck of the architecture. A 16/spl times/8-bit MAC is performed using a normal process complementary pass transistor (NPCPL) as a universal cell for the creation of conventional logic gates and is used to optimize the wave pipelined MAC. The throughput and the latency of the MAC have been evaluated at 650 MHz and 8 ns respectively. The performance has been evaluated in a 0.5 /spl mu/m CMOS technology in comparison with the systolic architecture with and without a conventional pipeline and the proposed wave pipeline structure.
  • Keywords
    CMOS logic circuits; VLSI; equalisers; logic gates; multilayer perceptrons; multiplying circuits; parallel algorithms; piecewise linear techniques; pipeline processing; systolic arrays; 0.5 mum; 16 bit; 650 MHz; 8 bit; 8 ns; CMOS technology; PL-MNN algorithm; VLSI; channel equalization; high throughput architecture; logic gates; multiplier-accumulator; neural network; normal process complementary pass transistor; performance evaluation; piecewise linear multilayer neural network; systolic architecture; universal cell; wave pipeline method; wave pipelined MAC; CMOS logic circuits; CMOS technology; Delay; Design methodology; Logic gates; Multi-layer neural network; Neural networks; Piecewise linear techniques; Pipeline processing; Throughput;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electrical and Computer Engineering, 1999 IEEE Canadian Conference on
  • Conference_Location
    Edmonton, Alberta, Canada
  • ISSN
    0840-7789
  • Print_ISBN
    0-7803-5579-2
  • Type

    conf

  • DOI
    10.1109/CCECE.1999.807260
  • Filename
    807260