• DocumentCode
    2896854
  • Title

    Parallel high throughput soft-output sphere decoder

  • Author

    Qi, Q. ; Chakrabarti, C.

  • Author_Institution
    Sch. of Electr., Comput. & Energy Eng., Arizona State Univ., Tempe, AZ, USA
  • fYear
    2010
  • fDate
    6-8 Oct. 2010
  • Firstpage
    174
  • Lastpage
    179
  • Abstract
    Multiple-Input-Multiple-Output communication systems demand fast sphere decoding with high performance. We propose a high throughput soft-output fixed complexity sphere decoder (PFSD) that is parallel and has comparable performance to list fixed complexity sphere decoder (LFSD) and K-best sphere decoder. In addition, we propose a parallel QR decomposition algorithm to lower the preprocessing overhead, and a low complexity LLR algorithm to allow parallel update of LLR values. We demonstrate the BER and computation complexity advantages of the PFSD algorithm in a 4×4 16-QAM system. The PFSD algorithm has been mapped onto Xilinx XC4VLX160 FPGA. The resulting PFSD decoder can produce 8 candidate vectors per clock cycle, and achieve upto 75Mbps throughput for 4×4 64-QAM configuration at 100MHz with low control overhead.
  • Keywords
    MIMO communication; computational complexity; decoding; error statistics; field programmable gate arrays; parallel architectures; quadrature amplitude modulation; 16-QAM system; 64-QAM; BER; MIMO communication; PFSD algorithm; Xilinx XC4VLX160 FPGA; bit-error-rate; computational complexity; k-best sphere decoder; list fixed complexity sphere decoder; low complexity LLR algorithm; multiple-input multiple-output communication system; parallel QR decomposition algorithm; parallel high throughput; soft-output fixed complexity sphere decoder; sphere decoding; Bit error rate; Complexity theory; Decoding; Generators; MIMO; Measurement; Sorting; FPGA; Parallel Implementation; Sphere Decoding;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signal Processing Systems (SIPS), 2010 IEEE Workshop on
  • Conference_Location
    San Francisco, CA
  • ISSN
    1520-6130
  • Print_ISBN
    978-1-4244-8932-9
  • Electronic_ISBN
    1520-6130
  • Type

    conf

  • DOI
    10.1109/SIPS.2010.5624783
  • Filename
    5624783