Energy and Spectral Efficiency Maximization of Full-Duplex Communications Under Imperfect CSI

In a full-duplex (FD) multiuser multiple-input-multiple-output (MU-MIMO) base station system with imperfect channel state information (ICSI), there are challenges, such as co-channel interference, residual self-interference (SI), channel estimation errors, multiuser interference, and noise. To maximize the spectral efficiency (SE) of a system affected by interference, we proposed an efficient signal-to-leakage-plus-noise-ratio precoding scheme (SLNR-PS). The optimal precoding vectors that maximize the sum rates were obtained by solving the signal-to-leakage-noise ratio (SLNR) optimization problem using the generalized eigenvalue decomposition method. The SE is then calculated by combining the sum rates of the downlink and uplink transmissions. By developing an efficient power consumption model and precoding design schemes, we investigate the maximum energy efficiency (EE) of these systems. The results of the simulation indicate that SLNR-PS outperforms zero-forcing and block-diagonalization methods at all SNR levels. Additionally, it has been verified that the SE enhancement of the FD system compared to the half-duplex system relies on the suggested SLNR-PS technique, considering various factors, such as the number of antennas, Rician factors, and channel state information (CSI) inaccuracies. Through the utilization of these proposed methodologies, we have further demonstrated that the achievable EE of the FD system exceeds that of the half-duplex system, taking into account transmit power, circuit power consumption, power amplifier efficiency, and CSI inaccuracies.