Precoded Massive MU-MIMO Uplink Transmission under Transceiver I/Q Imbalance

In massive multiple-input multiple-output (MIMO) systems, combined with digital array processing, the amount of the associated radio frequency (RF) front-ends is inevitably high. This paper addresses how imperfections in these RF front-ends affect the overall system performance in precoded massive multi-user MIMO (MU-MIMO) uplink transmission. In particular, we focus on transceiver in-phase/quadrature (I/Q) imbalances and their mitigation with RF-aware spatial processing. We first derive the essential distortion and interference models for OFDMA-based massive MU-MIMO uplink system under I/Q imbalances, and then propose augmented spatial post-processing to be carried out in the uplink receiver (RX) for mitigating the harmful effects efficiently. Numerical examples show that the augmented spatial RX processing clearly outperforms the conventional linear processing, and thus provides significant performance improvements with practical low-cost RF front-ends.