Joint overlay and underlay resource allocation with weighted fairness in OFDM-based cognitive radio systems

In this paper, the problem of hybrid overlay and underlay spectrum access is investigated for OFDM-based cognitive radio CR systems. Both the metrics for system e.g. capacity and users e.g. fairness are integrated into the unified framework of weighted-capacity maximization with the interference constraint in CR systems. For easing the procedure of resource allocation, two criteria, respectively, for subcarrier assignment and power allocation are theoretically derived based on the Karush-Kuhn-Tucker conditions. Under the guidance of the two criteria, max-capacity subcarrier assignment and optimal power allocation are proposed to flexibly distribute spectrum resources to secondary users. In order to reduce the computational complexity further, a suboptimal power allocation algorithm, referred to as cap-limited waterfilling, is also presented by decomposing the interference constraint. Simulation results show that the capacity and fairness performance of the proposed algorithms is considerably better than the conventional ones in references. The proposed suboptimal algorithm with substantially lower complexity approaches to optimal power allocation in the vicinity of only 1% performance gap. Meanwhile, joint access model is greatly beneficial to spectrum efficiency enhancement in CR systems. Copyright © 2014 John Wiley & Sons, Ltd.