Robust Beamforming in Cognitive Radio

In cognitive radio, it is crucial to control the interference from secondary users (SUs) to primary users (PUs). This paper studies the use of transmit beamforming in the cognitive secondary network for enhancing the performance of a SU while controlling the interference to the PUs. In particular, we propose to maximize the service probability of the SU with a number of probability constraints on the interference level at the PUs with the aid of imperfect channel state information (CSI). Modeling the CSI uncertainty as an additive Gaussian noise, it is shown that the optimum can be realized by second-order cone-programming (SOCP) in tandem with a one-dimensional search. Results reveal that the proposed approach provides a technique to tradeoff the performance between the PUs and the SU, making an analytical connection between non-robust and worst-case systems.