Full-Duplex Bi-directional MIMO under Imperfect CSI and Hardware Impairments

Abstract

In this paper, we consider a bi-directional in-band Full-Duplex (IBFD) multiple-input multiple-output (MIMO) system, where reduced hardware (HW) complexity analog self-interference cancellation (SIC) units are deployed at both IBFD MIMO nodes. The system is assumed to operate with imperfect channel state information (CSI) knowledge, and both transmitter (TX)s possess non-ideal radio frequency (RF) chains. Focusing on the TX power minimization problem with per node signal to interference-plus-noise ratio (SINR) constraints, we design a low-complexity iterative algorithm for the TX and receiver (RX) beamformers. According to the proposed algorithm, TX power is adjusted via the classical Perron-Frobenius (PF) theorem, whereas TX and RX beamforming (BF) vectors are designed in order to minimize the interference-distortion plus noise (IDN) components at RX, while maximizing the intended signal power. Our representative simulation results with realistic parameters for the analog canceller HW demonstrate that the proposed algorithm outperforms the relevant State-of-the-Art (SotA) designs in terms of outage probability of the provided available TX power.