Max-Min Power Control in Downlink Massive MIMO With Distributed Antenna Arrays

• Published in: IEEE transactions on communications Vol. 69; no. 2; pp. 740 - 751
• Main Authors: Akbar, Noman, Bjornson, Emil, Yang, Nan, Larsson, Erik G.
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Antenna arrays; Antennas; Channel estimation; Control algorithms; Control theory; Covariance; Distributed wireless networks; Downlink; Downlinking; Fading; Fading channels; Massive MIMO; max-min power control; MIMO (control systems); optimization; Power control; Power management; Quality of service architectures; Fading channels; Distributed wireless networks; Power control; optimization; Channel estimation; Massive MIMO; Downlink; max-min power control; Antenna arrays
• ISSN: 0090-6778; 1558-0857; 1558-0857
• DOI: 10.1109/TCOMM.2020.3033018

In this paper, we investigate optimal downlink power allocation in massive multiple-input multiple-output (MIMO) networks with distributed antenna arrays (DAAs) under correlated and uncorrelated channel fading. In DAA massive MIMO, a base station (BS) consists of multiple antenna sub-arrays. Notably, the antenna sub-arrays are deployed in arbitrary locations within a DAA massive MIMO cell. Consequently, the distance-dependent large-scale propagation coefficients are different from a user to these different antenna sub-arrays, which makes power control a challenging problem. We assume that the network operates in time-division duplex mode, where each BS obtains the channel estimates via uplink pilots. Based on the channel estimates, the BSs perform maximum-ratio transmission in the downlink. We then derive a closed-form signal-to-interference-plus-noise ratio (SINR) expression, where the channels are subject to correlated fading. Based on the SINR expression, we propose a network-wide max-min power control algorithm to ensure that each user in the network receives a uniform quality of service. Numerical results demonstrate the performance advantages offered by DAA massive MIMO. For some specific scenarios, DAA massive MIMO can improve the average per-user throughput up to 55%. Furthermore, we demonstrate that channel fading covariance is an important factor in determining the performance of DAA massive MIMO.