• DocumentCode
    586261
  • Title

    Ergodic Capacity of Multi-User MIMO Systems Using Pilot-Based Channel Estimation, Quantized Feedback and Outdated Feedback as Well as User Selection

  • Author

    Jin, Fan ; Hanzo, Lajos

  • Author_Institution
    Sch. of ECS, Univ. of Southampton, Southampton, UK
  • fYear
    2012
  • fDate
    3-6 Sept. 2012
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    We derive the ergodic channel capacity of a closed-loop MIMO broadcast system under both random user selection (RUS) and semi-orthogonal user selection (SUS) principles, when considering the effects of channel estimation errors, channel state information (CSI) quantization errors and CSI feedback-delay. We intend to answer the question of "How many simultaneous users should be active in a time slot in order to achieve the maximum sum-rate in a MIMO broadcast system?". It is shown that we should refrain from supporting full-rank transmissions due to the excess inter-user interference caused by imperfect CSI at both the receiver and transmitter under RUS. By contrast, full-rank transmissions may be feasible under the SUS principle. An approximate ergodic capacity equation is also derived for RUS, and then an adaptive-rank transmission strategy is derived, which is capable of maximizing the achievable ergodic capacity.
  • Keywords
    MIMO communication; broadcast channels; channel capacity; channel estimation; closed loop systems; feedback; interference (signal); multiuser channels; radio receivers; radio transmitters; random processes; CSI feedback-delay; CSI quantization errors; RUS principles; SUS principles; adaptive-rank transmission strategy; channel estimation errors; channel state information quantization errors; closed-loop MIMO broadcast system; ergodic capacity equation; ergodic channel capacity; full-rank transmissions; interuser interference; multiuser MIMO systems; outdated feedback; pilot-based channel estimation; quantized feedback; radio receiver; radio transmitter; random user selection; semiorthogonal user selection priciple; Approximation methods; Channel estimation; Delay; MIMO; Quantization; Signal to noise ratio; Vectors;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Vehicular Technology Conference (VTC Fall), 2012 IEEE
  • Conference_Location
    Quebec City, QC
  • ISSN
    1090-3038
  • Print_ISBN
    978-1-4673-1880-8
  • Electronic_ISBN
    1090-3038
  • Type

    conf

  • DOI
    10.1109/VTCFall.2012.6399232
  • Filename
    6399232