• DocumentCode
    414888
  • Title

    Multi-target channel inversion power control for block-fading channels with multiple antennas

  • Author

    Rui Zhang ; Ying-Chang Liang ; Cioffi, J.M.

  • Author_Institution
    Stanford University
  • Volume
    1
  • fYear
    2004
  • fDate
    20-24 June 2004
  • Firstpage
    593
  • Lastpage
    597
  • Abstract
    This paper studies the capacities of block-fading (BF) additive white Gaussian noise (AWGN) channels with multiple transmit and receive antennas, under the assumption that perfect channel state information is available at both the transmitter (CSIT) and the receiver (CSIR). We study both ergodic capacity under no transmission delay constraint and delay-limited capacity for delay-constrained case, and consider the tradeoff between capacity and complexity. Optimal power allocation schemes to achieve ergodic and delay-limited capacities are known as variants of standard water-filling algorithms, which usually requires large computational complexity at the transmitter side. In this paper, we propose a new sub-optimal power allocation scheme, so-called multi-target channel inversion (MT-CI), which can be easily implemented at the transmitter and still asymptotically achieve the performance of optimal schemes. Furthermore, MT-CI transforms the original fading channel into parallel AWGN channels and hence greatly reduces the receiver complexity. In addition, the proposed scheme is shown to be also suitable for simultaneous transmission of multimedia services which may require different levels of quality of service (QOS).
  • Keywords
    AWGN; Additive white noise; Channel state information; Delay; Fading; Mutual information; Power control; Quality of service; Receiving antennas; Transmitters;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Communications, 2004 IEEE International Conference on
  • Conference_Location
    Paris, France
  • Print_ISBN
    0-7803-8533-0
  • Type

    conf

  • DOI
    10.1109/ICC.2004.1312558
  • Filename
    1312558