Beamforming design for weighted rate balance in cognitive network

Abstract

When a cognitive radio network shares the spectrum with a primary network, the harmful interference to the primary users is one of the most important concerns. In this paper, we propose a new cognitive scheme which has two cognitive sub-networks sharing the same relay station, considering the interference to the primary user. We study the Weighted Rate Balance problem with the constraint of the transmit power of the cognitive network. Joint beamforming at the cognitive base station and the relay station is used to maximize the Weighted Rate Balance of SUs. To solve the non-convex problem, we adopt the classic altering optimization method, namely the Blahut-Arimoto algorithm. We show that we only need to solve several semidefinite programming(SDP) sub-problems, which have linear matrix inequalities(LMIs), in each iterative step. Simulation results show that, compared with the widely studied diagonally structured beamforming design, the proposed approach performs better.