• DocumentCode
    1343999
  • Title

    Vector Perturbation Precoding Revisited

  • Author

    Maurer, Johannes ; Jaldén, Joakim ; Seethaler, Dominik ; Matz, Gerald

  • Author_Institution
    Inst. of Commun. & Radio-Freq. Eng., Vienna Univ. of Technol., Vienna, Austria
  • Volume
    59
  • Issue
    1
  • fYear
    2011
  • Firstpage
    315
  • Lastpage
    328
  • Abstract
    We consider the downlink of a multiuser system with multiple antennas at the base station. Vector perturbation (VP) precoding is a promising variant of transmit-side channel inversion allowing the users to detect their data in a simple, noncooperative manner. VP precoding has so far been developed and analyzed under the assumptions that the transmitter has perfect channel state information (CSI) and that the receivers know perfectly a channel-dependent transmit power normalization factor and have infinite dynamic range. We demonstrate that the violation of any of these idealizing assumptions degrades the performance of VP significantly and almost always results in an error floor. Motivated by this observation, we propose a novel scheme which we term transmit outage precoding (TOP). With TOP, the transmitter uses a prearranged power scaling known by the receivers and refrains from transmitting when channel conditions are poor. We further show how to augment TOP and conventional VP to deal with a finite dynamic range at the receiver. The performance of the proposed schemes under various levels of transmit CSI is studied in terms of a theoretical diversity analysis and illustrated by numerical results.
  • Keywords
    antenna arrays; channel coding; precoding; radio receivers; radio transmitters; wireless channels; CSI; TOP; VP precoding; channel state information; channel-dependent transmit power normalization factor; multiple antenna; multiuser system; power scaling; radio receiver; radio transmitter; transmit outage precoding; transmit-side channel inversion; vector perturbation precoding; Base stations; Channel estimation; Decoding; Dynamic range; Receivers; Signal to noise ratio; Diversity; imperfect channel state information; precoding;
  • fLanguage
    English
  • Journal_Title
    Signal Processing, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    1053-587X
  • Type

    jour

  • DOI
    10.1109/TSP.2010.2084083
  • Filename
    5595019