Beamforming for MISO Cognitive Radio Networks Based on Successive Convex Approximation

Abstract

This paper presents a novel beamforming optimization method for downlink underlying multiple-input single-output (MISO) cognitive radio (CR) networks. We formulate a beamforming optimization problem to maximize the sum rate of a CR network with one primary user (PU) pair and multiple secondary user (SU) pairs, subject to the quality of service requirement of the PU given the transmit power budgets at the base stations (BSs). To find the solution of the nonconvex problem, an iterative solving algorithm is proposed based on successive convex approximation (SCA). In developing the algorithm, we first reformulate the original nonconvex objective function as the difference of two concave functions. A concave substitute function is then derived using the one-order Taylor expansion. Based on this concave substitute, a convex semidefinite programming (SDP) is derived and solved. The new solution is then utilized to construct a new substitute. This process is repeated until a smooth point is reached. Simulation results show the effectiveness of the proposed SCA-based beamforming algorithm to achieve spectrum sharing.