Spectral Efficiency of Distributed Large-Scale MIMO Systems With ZF Receivers

Distributed multiple-input multiple-output (D-MIMO) is a promising technique for next-generation wireless networks, which offers a remarkable spectral efficiency gain over the conventional colocated MIMO (C-MIMO). In contrast to C-MIMO, which can be regarded as a special case of D-MIMO, performance analysis of D-MIMO is a challenging problem. This is because radio channels between a user and the distributed radio ports (RPs) are characterized by nonidentical path-loss and shadowing effects that render the classical analytical methods nontractable. In this paper, new accurate expressions for the uplink spectral efficiency of D-MIMO and C-MIMO systems are presented and compared for given large-scale coefficients. The authors further consider the uplink spectral efficiency for a single-cell distributed large-scale MIMO (D-LMIMO) system with linear zero-forcing (ZF) receivers that account for path loss along with shadow fading and multipath fading effects. Exact expressions for the average spectral efficiency over shadow fading in the asymptotically very large number of RPs antenna regime is explicitly derived, and a tight closed-form lower bound on the asymptotic spectral efficiency is presented. The authors demonstrate that, the transmit power of each user in D-LMIMO can be scaled down proportionally to the inverse of the number of RP antennas with no performance reduction. Moreover, the authors study the spectral efficiency of a D-MIMO in a multicell environment taking into account accurate cochannel interference (CCI) models. These expressions provide meaningful insights into the impact of signal-to-noise ratio (SNR), RPs and user positions, number of RPs antennas, shadow fading, and out-of-cell interference on the spectral efficiency of D-MIMO over practical scenarios. Finally, numerical results are validated by simulation to confirm the authors' analysis.


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01645117
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jun 22 2017 4:12PM