• DocumentCode
    3032681
  • Title

    An implementation of MIMO detection in TD-LTE based on General Purpose Processor

  • Author

    Li Zhou ; He Zhiqiang ; Duan Ran ; He Lifeng

  • Author_Institution
    Key Lab. of Universal Wireless Commun., Beijing Univ. of Posts & Telecommun., Beijing, China
  • fYear
    2012
  • fDate
    8-10 Aug. 2012
  • Firstpage
    807
  • Lastpage
    811
  • Abstract
    This paper presents a method of implementing Multiple Input Multiple Output (MIMO) detection on General Purpose Processor (GPP) with Single Instruction Multiple Data (SIMD) instructions. In the receiver of Long Term Evolution (LTE) communication system MIMO detection plays a rather important role. How to reduce the processing latency of MIMO detection becomes increasingly significant. The advantages of GPP compared with programmable hardware provide us a considerable solution to cope with this problem. SIMD is another effective manner as it is able to increase the computing parallel degree. For each algorithm of MIMO detection, fixed-point operations which have higher parallel degree than floating-point operations may be a better choice to further reduce the process delay. Nevertheless, the performance attained by fixed-point is still a factor need to be considered while its latency is quite low.
  • Keywords
    Long Term Evolution; MIMO communication; parallel processing; telecommunication computing; LTE communication system; Long Term Evolution communication system; MIMO detection; SIMD; TD-LTE; fixed-point operation; general purpose processor; multiple input multiple output detection; single instruction multiple data instruction; Complexity theory; Graphics processing units; MIMO; Multiplexing; Registers; Vectors; Wireless communication; Floating-point and fixed-point; GPP; LTE; Latency; Performance; SIMD;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Communications and Networking in China (CHINACOM), 2012 7th International ICST Conference on
  • Conference_Location
    Kun Ming
  • Print_ISBN
    978-1-4673-2698-8
  • Electronic_ISBN
    978-1-4673-2697-1
  • Type

    conf

  • DOI
    10.1109/ChinaCom.2012.6417595
  • Filename
    6417595