Discrete broadcast beamforming with antenna and user selection

Antenna selection is an effective approach to reduce hardware cost as well as increasing channel diversity. In this paper, broadcast beamforming design with antenna selection is considered for cognitive radio networks. In the presence of primary users, it is usually difficult to satisfy quality of service (QoS) constraints of the all secondary users while keeping the interference towards the primary users below a certain threshold. User selection is a good method in order to make the problem feasible. In this paper, discrete phase-only beamforming is considered due to its simplicity and certain advantages. The joint problem is converted into linear form and solved optimally by using mixed integer linear programming. Simulation results show that significant power saving is possible compared to fixed antenna systems.