Robust utility scheme with admission control over MIMO cognitive radio network

Multiple-input multiple-output (MIMO) cognitive radio network is considered in this paper, in which single primary user (PU) coexists with multiple secondary users (SUs). Standard cognitive radio techniques require perfect channel state information (CSI) and are sensitive to the errors in channel estimation. Under the assumption of the norm-bounded channel imperfections, we develop a robust energy-aware design which admits the maximum number of the served SUs and enhances the transmit rate of the secondary links by using the least transmit power while maintaining the quality-of-service (QoS) constraint of the PU and the SUs. Instead of using deterministic target, the signal-to-interference-and-noise ratio (SINR) threshold becomes one of on-going optimization variables, and acts as the admission control implicitly. Due to its non-convexity and NP-hardness, the underlying problem is reformulated in semidefinite programming (SDP) form and solved via iterative bisection search approach. Simulation results are provided to validate the robustness and efficiency of the proposed scheme.