Matched-Filter Detector With Quadrature Interference Cancellation for Coded Uplink Multiuser MIMO Systems With Massive IoT Devices

We propose a new interference cancellation (IC) approach designed for coded uplink multiple-input multiple-output (MIMO) systems that support a massive number of Internet-of-Things (IoT) devices. The performance of MIMO spatial multiplexing depends on the symbol detection schemes, which in general offer a trade-off between the error rate performance and computational complexity. Previous studies have shown that a simple matched-filter (MF)-based detector with powerful channel coding can achieve near-optimal performance as the number of base station (BS) antennas increases even without IC. However, if the number of BS antennas is the same order as that of the IoT devices supported by the BS, its achievable rate will be significantly degraded. Furthermore, channel estimation errors due to the lack of orthogonal pilot sequences should also become a dominant factor. This paper overcomes this problem by introducing a MF detector with quadrature interference cancellation (MF-QIC), which can improve error performance and spectral efficiency. Taking polar codes as our channel coding example, we propose a code design based on mutual information tailored for MF-QIC. Our simulation results demonstrate that the proposed low-complexity polar-coded MF-QIC outperforms other conventional approaches even in the presence of imperfect channel state information.