Multiuser MISO Beamforming Design for Balancing the Received Powers in Secure Cognitive Radio Networks

In this study, we consider multiuser multi-input single-output (MISO) beamforming design in underlay cognitive radio networks in the presence of a primary user (PU) and eavesdropper (Eve). We propose a multi-objective optimization problem (MOOP) for balancing two conflicting design objectives: maximizing the total intended power received at secondary receivers and minimizing the interference power at the PU, while satisfying the secrecy rate and PU interference threshold constraints. The MOOP was solved by a weighted sum method and semidefinite programming relaxation (SDR), and the rank-one optimal solutions were drawn with explanations. The upper and lower bounds were found by solving the single-objective and zero-forcing beamforming problems. The simulation results demonstrated a trade-off between the objectives in the Pareto optimal set, no improvement of performance with artificial noise, and better results with increasing number of transmission antennas.