• DocumentCode
    163046
  • Title

    A Novel Joint Precoder and Receiver Design with Imperfect CSI for Multi-User MIMO Downlink

  • Author

    Ugurlu, U. ; Wichman, R. ; Ribeiro, C. ; Wijting, C.

  • Author_Institution
    Sch. of Electr. Eng., Aalto Univ., Aalto, Finland
  • fYear
    2014
  • fDate
    14-17 Sept. 2014
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    This paper studies the design of a linear precoder and receiver for the downlink of multi-user multiple-input multiple-output (MIMO) systems with imperfect channel state information (CSI). First, we modify the regularized block diagonalization scheme for the case of imperfect CSI, where we assume the variance of the channel estimation error is known. Next, we propose a novel joint design for the optimized precoder and receiver pair based on the minimum mean square error (MMSE) criterion, where we introduce two transmit channel matrices designed for high signal-to-noise ratio conditions. We compare the performance of these schemes with the generalized MMSE channel inversion filtering via numerical analysis. The results demonstrate the superiority of our joint precoder and receiver design.
  • Keywords
    MIMO communication; channel coding; channel estimation; least mean squares methods; linear codes; matrix algebra; precoding; radio receivers; wireless channels; MMSE; block diagonalization scheme; channel estimation; channel matric transmission designed; imperfect CSI; imperfect channel state information; joint precoder and receiver design; linear precoder; linear receiver; minimum mean square error criterion; multiuser MIMO downlink; multiuser multiple input multiple output systems downlink; numerical analysis; signal-to-noise ratio; Base stations; Channel estimation; Downlink; Joints; MIMO; Receivers; Signal to noise ratio;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Vehicular Technology Conference (VTC Fall), 2014 IEEE 80th
  • Conference_Location
    Vancouver, BC
  • Type

    conf

  • DOI
    10.1109/VTCFall.2014.6965827
  • Filename
    6965827