Joint beamforming methods for full-duplex multiuser networks

Abstract

In this dissertation, we introduce the joint beamforming algorithms to solve the non-convex problem of sum rate maximization in full-duplex (FD) multiuser multiple-input multiple-output (MU-MIMO) systems. In the FD transmission of the multiuser network, there are a self-interference (SI) signal in which the downlink signal of the base station (BS) is received by itself and a co-channel interference (CCI) signal in which the transmitted signal of the uplink user is received by the downlink user. Since SI and CCI are coupled in FD transmission, beamforming vectors for uplink and downlink multiuser are required to be jointly designed. However, objective of the joint optimization problem is non-convex and hard to solve due to the coupled effect. To properly address the coupled design issue, we reformulate the problem into an equivalent dual uplink channel problem, using uplink and downlink duality known as MAC-BC duality. Although the equivalent optimization problem is still non-convex, we propose two algorithms based on iterative method to solve the simplified problem. Minorization-maximization (MM) algorithm allows us to apply the approach which makes an affine approximation for non-convex objective function using the first-order approximation. To avoid performance degradation resulting from the approximation, we also devise an alternating algorithm based on iterative water-filling (IWF) without any approximation to the objective function. We prove that alternating IWF algorithm provides better performance of sum rate and lower computational complexity than MM algorithm by numerical results. Furthermore, we develop the precoder design algorithm which can provide the optimal beamforming vector and the SI pre-suppression vector simultaneously. Finally, we show that the FD transceiver structure is simplified by using the proposed precoder and the maximum sum rate of FD MU-MIMO network is improved.