Joint Tx/Rx Antenna Selection and Beamforming Optimization for Multi-User Full-Duplex mMIMO

Abstract

This paper proposes antenna selection-based beamforming optimization algorithms for full-duplex (FD) massive multiple-input multiple-output (mMIMO) large-scale arrays. The approach jointly selects the optimal transmit and receive antennas to optimize the propagation channels, thereby mitigating both self-interference (SI) and multi-user interference while preserving the desired signal strength. Illustrative results based on measured SI channels between crosspolarized 8 $\times$ 8 Tx and 8 $\times$ 8 Rx arrays demonstrate substantial improvements in both SI suppression and achievable total rate. On average, the proposed method achieves a 129.3% FD sumrate improvement over the Power-Greedy Selection scheme and a 65.5% improvement over the Random Selection scheme. Specifically, compared with the Power-Greedy and Random Selection schemes, the proposed method improves the uplink rate by average factors of 11.3 and 29.2, respectively, with maximum improvements of up to 13.9 and 35.3 times. An average 9.1 dB enhancement in SI suppression is observed. A comprehensive analysis of antenna selection effectiveness and downlink-uplink rate tradeoff in FD communications is provided.