Robust Power Allocation for OFDM Based Underlay Cognitive Radio Networks with Channel Uncertainties

Abstract

In consideration of all possible channel uncertainties without the assumption of perfect channel state information (CSI), a robust power allocation algorithm for underlay orthogonal frequency division multiple cognitive radio networks is presented. This algorithm can assign transmission power of each secondary user (SU) on each sub-carrier based on total transmission power minimization of SUs under the constraints corresponding to signal-to-interference-noise ratio of SUs and the interference power constraint to guarantee the quality of service of primary users (PUs). In addition, the CSI errors are assumed to be bounded with ellipsoid and interval sets. Through the worst case approach, the original optimization problem is converted into a convex one solved by Lagrange dual decomposition method. The proposed robust algorithm provides a trade-off between robustness and system performance. Simulation results prove that the suboptimal solution can achieve a satisfactory performance for both SUs and PUs at the same time.