Interference Exploitation Beamforming for Cognitive Radio Network with Energy Transfer

Instead of fully mitigating the interference, we design a data-aided beamforming scheme that exploits the multiuser interference to further improve the quality of the received signal at each secondary user, termed constructive interference (CI) scheme. A power minimization problem is formulated by jointly optimizing the beamforming design and the power splitting (PS) ratio, subject to the constraints of the received signal-to-interference-plus-noise ratio (SINR), the harvested energy requirement of the secondary users and interference power of primary users. To tackle this non-convex problem, we provide an approximate second-order cone programming (SOCP) relaxation method for the optimization problem, and derive the robust beamformer for CI exploitation in the presence of imperfect CSI. Simulation results show that the proposed CI scheme can reduce the transmission power significantly compared with the conventional interference suppression transmit beamforming designs.