Uplink Performance Analysis for Massive MIMO Linear Processing

This paper considers an uplink massive MIMO system, in which the base station (BS) is equipped with a very large antenna array to serve multiple users simultaneously in the presence of out-of-cell interferers. Multi-cell minimum mean-square-error (M-MMSE) detection is used to mitigate co-channel int...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2019, Vol.7, p.180749-180760
Main Authors: Abid, Wahiba, Roy, Sebastien, Ammari, Mohamed Lassaad
Format: Article
Language:English
Subjects:
Antenna arrays
Antennas
Cochannel interference
Complexity
Complexity theory
Detectors
Error detection
Fading
Fading channels
Intercell interference
Linear receivers
Massive MIMO
Massive MIMO system
MIMO (control systems)
multi-cell MMSE
Receivers
Receivers & amplifiers
SINR
two-layer receiver processing
Uplink
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper considers an uplink massive MIMO system, in which the base station (BS) is equipped with a very large antenna array to serve multiple users simultaneously in the presence of out-of-cell interferers. Multi-cell minimum mean-square-error (M-MMSE) detection is used to mitigate co-channel interference (CCI) and multipath fading effects. However, this receiver processing brings high computational complexity to bear in a real-time massive MIMO scenario. To overcome this higher complexity, novel two-layer linear receiver processing schemes are proposed in this work, which can achieve a good trade-off between performance and complexity. The proposed architecture consists of two layers: 1) splitting the antenna array into a number of subsets, and performing M-MMSE processing at the subset level; 2) combining the resulting subset outputs using either MRC or M-MMSE detectors. Taking into account both small- and large-scale fading, we investigate the system performance and the computational complexity of the proposed receivers. To further characterize the advantages of the proposed schemes, they are compared with conventional detectors. Numerical simulations show that the proposed schemes approach the performance of conventional M-MMSE processing, albeit with significantly reduced complexity. We also derived tight expressions for intra-cell and inter-cell residual interference powers at the output of the first processing layer. An important observation is that the inter-cell interference dominates the total interference, especially when shadowing is strong or the subset size is comparable to the total number of users. However, performing M-MMSE at the second processing layer provides significant gains in this context.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2019.2959580