Beamforming Strategies In Cognitive Radio Networks

Cognitive radio is a promising solution to alleviating spectrum scarcity in wireless communications. The basic idea of CR is the hierarchical access model of spectrum sharing. Unlicensed secondary users opportunistically access to a spectrum which is originally allocated to licensed primary users, provided that the access do not harmfully affect the PUs. Relay assisted wireless communication has attracted great attentions to improve performance of cognitive radio networks. Different relay schemes have been presented such as amplify-and-forward and decode-and-forward, in which the relay nodes play different roles on processing the received signal before retransmitting. The physical layer security problem, first introduced by Wyner, has received a growing concern in past few years. Although the security discussion is meaningful and critical for the CRN, research in this area is still insufficient. In this paper, we study beamforming strategies in two-way relay cognitive radio networks and MISO cognitive secure networks.We first consider a two-step two-way relay cognitive network with one primary user, two secondary users and multiple relay nodes, each with single antenna. Our aim is to design the optimal beamforming to maximize the sum rate of the secondary network with power and interference constraints. Secondly, we focuses on a multiple-input single-output cognitive radio network, where both the primary and the secondary users communicate with confidential messages. A multi-antenna eavesdropper whose channel state information has been known is considered in our CRN. While we do not assume its wiretap target, which means each user may have been tapped, and define it as a target-unknown eavesdropper. A specific application of spectrum leasing is proposed to ensure the secrecy performance of the whole network, in which SU accesses the spectrum in reward of helping improve the secrecy performance of the entire CRN. The transmit covariance of the ST is designed to maximize the worst-user’s secrecy rate.