Blind-based demodulation scheme for virtual massive MIMO systems

摘要

In beyond-fifth-generation (B5G) networks, massive multiple-input-multiple-output (MIMO) systems using a large number of base station antennas have attracted significant attention. As the number of antennas increases, the hardware scale of massive MIMO systems is complicated. In recent years, a virtual massive (VM)-MIMO system which performs MIMO transmission by using virtual antenna arrays creating a single antenna element has been investigated. In this paper, we propose a blind-based demodulation scheme for a VM-MIMO system to cancel the interference with oversampling. In our proposed scheme, as a signal whose level is less than a certain power threshold cannot be used in the high-order modulation scheme with oversampling, we apply filtering wherein signals with the noise power level are removed. Through this procedure, multi-level modulation signals can be utilized in VM-MIMO. Moreover, our simulation evaluations show that the bit error rate improves considerably by using our proposed scheme on the uplink commutation environment including up to ten users.