1: \begin{abstract}

2: Full-duplex (FD) systems have emerged as an essential enabling technology  to further increase the data rate of wireless

3: communication systems. The key idea of FD is to serve multiple users over the same bandwidth with a base station (BS) that can

4: simultaneously transmit and receive the signals. The most challenging issue in designing an FD system is  to address both the harmful effects

5: of residual self-interference  caused by the transmit-to-receive antennas at the BS as well as the co-channel interference 

6: from an uplink user (ULU) to a downlink user (DLU). An efficient solution to these problems  is to assign

7: the ULUs/DLUs in different groups/slots, with each user served in multiple groups. Hence, this paper studies the joint design of transmit beamformers,  ULUs/DLUs group assignment, and time allocation for each group. The specific aim is to maximize the

8: sum rate under the ULU/DLU minimum throughput constraints. The utility function of interest is a difficult nonconcave problem, and the involved

9: constraints are also nonconvex, and so this is a computationally troublesome problem. To solve this optimization problem, we

10: propose a new path-following algorithm for computational solutions to  arrive at  least the local optima. Each iteration

11: involves only a simple convex quadratic program. We prove that the proposed algorithm iteratively improves the objective

12: while guaranteeing convergence. Simulation results confirm the fast convergence of the proposed algorithm with substantial

13: performance improvements over existing approaches.

14: \end{abstract}

15: