• DocumentCode
    1341606
  • Title

    On the Rate Gap Between Multi- and Single-Cell Processing Under Opportunistic Scheduling

  • Author

    Bang, Hans Jørgen ; Gesbert, David ; Orten, Pål

  • Author_Institution
    Elliptic Labs., Oslo, Norway
  • Volume
    60
  • Issue
    1
  • fYear
    2012
  • Firstpage
    415
  • Lastpage
    425
  • Abstract
    Base station (BS) coordination is a key technique to handle intercell interference (ICI) in cellular networks. Nevertheless, recent work on scheduling indicates that the value of coordination is less prominent when the number of users grows large. More specifically, the loss in sum rate due to ICI in uncoordinated networks can be made arbitrarily small as the number of users goes to infinity. However, the gap in performance for a finite number of users has remained unknown so far. From this perspective we study the gains of multicell zero-forcing beamforming (ZFBF) on the downlink of a Wyner-type network. We first identify the beamforming weights and the optimal scheduling policy under a per-base power constraint. To compare ZFBF with single-cell processing (SCP) we focus on the extra number of users that is needed per cell to compensate for ICI. Specifically, we find the number of users n1 with ZFBF and n2 with SCP that gives the same mean postscheduling signal-to-interference-plus-noise ratio (SINR) as an interference free network with n users. The results show that the ratio [(n2)/(n1)] grows logarithmically with n. Finally, we demonstrate that the difference in sum-rate between SCP and multicell ZFBF goes to zero as O([(lnlnn)/(lnn)]). As a consequence of the slow convergence there is a significant gain with multicell ZFBF for all practical numbers of users.
  • Keywords
    adjacent channel interference; array signal processing; cellular radio; interference suppression; scheduling; ICI; SCP; Wyner-type network; ZFBF; base station; cellular networks; intercell interference; multicell processing; opportunistic scheduling; optimal scheduling policy; signal-to-interference-plus-noise ratio; single-cell processing; zero-forcing beamforming; Array signal processing; Convergence; Electronic mail; Interference; Optimal scheduling; Signal to noise ratio; Upper bound; Base station (BS) coordination; multiuser scheduling; zero-forcing beamforming (ZFBF);
  • fLanguage
    English
  • Journal_Title
    Signal Processing, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    1053-587X
  • Type

    jour

  • DOI
    10.1109/TSP.2011.2170681
  • Filename
    6035798