Interference alignment in cognitive radio networks

Abstract

A fast algorithm to perform interference alignment (IA) in a cognitive radio network is proposed. Here, the authors consider a network consisting of multiple primary users (PUs) and secondary users (SUs). Although the PUs are cooperating through IA and their rates are maximised by water-filling algorithm, in the secondary network, the authors try to choose the precoding matrices of the SUs such that the interferences at the primary receivers are aligned away from the desired subspaces of the primary receivers. Besides, the interference cancellation of the PUs on the secondary links is considered through designing the post-processing matrices for the secondary links. In this way, their mutual harmful effects are suppressed and SUs could transmit their information simultaneously on the same frequency or time without causing/receiving any interference to/from the primary links. Meanwhile, the feasibility condition for the secondary network regarding the mentioned constraints is investigated. In addition, the rates of the secondary links are maximised by water-filling algorithm. The proposed iterative algorithm is an improved version of minimum weighted leakage interference algorithm with a much faster convergence rate. The analytical results are confirmed through simulations.