An Efficient Zero-Forcing Precoding Design for Cognitive MIMO Broadcast Channels

We consider linear precoding design for an underlay cognitive radio multiple-input multiple-output broadcast channel in the presence of multiple primary users (PUs). Under the assumption of imperfect channel state information (CSI) of the PUs, the objective of this letter is to maximize the sum rate of the secondary system, subject to the power budget at the secondary base station and the interference power constraints at the PUs. The design problem is non-convex, and thus is difficult to solve in general. Herein, we first convert a non-convex constraint related to the imperfect CSI of the PUs into a convex constraint, and then invoke a rank relaxation method to transform the considered problem into a convex-concave problem based on a downlink-uplink duality result. Simulation results are provided to demonstrate the effectiveness and robustness of the proposed design against CSI imperfection.