Interference-Driven Antenna Selection for Massive Multiuser MIMO

Abstract

Low-complexity linear precoders are known to be close-to-optimal for massive multi-input multi-output (M-MIMO) systems. However, the large number of antennas at the transmitter imposes high computational burdens and high hardware overloads. In line with the above, in this paper we propose a low complexity antenna selection (AS) scheme which selects the antennas that maximize constructive interference between the users. Our analyses show that the proposed AS algorithm, in combination with a simple matched filter (MF) precoder at the transmitter, is able to achieve better performances than systems equipped with a more complex channel inversion (CI) precoder and computationally expensive AS techniques. First, we give an analytical definition of constructive and destructive interference, based on the phase of the received signals from phase-shifted-keying (PSK) modulated transmissions. Then, we introduce the proposed antenna selection algorithm, which identifies the antenna subset with the highest constructive interference, maximizing the power received by the user. In our studies, we derive the computational burden of the proposed technique with a rigorous and thorough analysis and we identify a closed form expression of the upper bound received power at the user side. In addition, we evaluate in detail the power benefits of the proposed transmission scheme by defining an efficiency metric based on the achieved throughput. The results presented in this paper prove that antenna selection and green radio concepts can be jointly used for power efficient M-MIMO, as they lead to significant power savings and complexity reductions.

MIMO antenna selection using biogeography‐based optimization with nonlinear migration models

“…Moreover, for all the algorithms, we set the population size NP, the maximum number of generations G, and the elitism parameter p same as in other studies. 18,19 These values are NP = 80, G = {12, 25, 40} for (L T , L R ) = {(2, 4), (3,5), (4, 6)}, respectively, and p = 4. For the last scenario, (L T , L R ) = (8,8), we select NP = 80, G = 50 and p = 4.…”

“…The antenna selection (AS) method becomes even more important within the emerging massive MIMO technology. [5][6][7][8] Thus, it is desirable to have a larger antenna number than the RF chains number in each link end and to choose the antenna subset that will be in use. However, the reduction of the antenna number will certainly cause performance deterioration.…”

Extending the user capacity of MU-MIMO systems with low detection complexity and receive diversity

“…The complexity of this algorithm is of ( ) due to the requirement of vector norm calculations of all rows. Thus, low selection complexity is achieved compared with the most popular methods in [18][19][20].…”

“…On the other hand, average symbol error probability ( ) of conventional MU-MIMO over constant channel ∈ × and with same total average power constraint as in the proposed scheme can be calculated for average SNR Note that for realistic GL-MU-MIMO system of instantaneous channels � , the results of equations (11), (13), and (15) should be averaged over large number of channel realizations [19,20] …”

“…WiMAX and 3GPP-LTE standards [21] and becomes an essential part of the promising massive MIMO systems [20].…”