Optimizing sum-capacity through power allocation for SLNR-precoding-based cognitive networks

Cognitive radio (CR) has great potential to improve the spectral efficiency of future wireless networks. This paper focuses on maximizing the sum-capacity of cognitive wireless networks based on signal to leakage noise ratio (SLNR) precoding and hybrid opportunistic spectrum access scheme. We propose a globally optimal power allocation scheme based on a combination of the Branch and Bound framework (B&B) and convex relaxation technique to maximize the sum capacity of all secondary users (SUs). Simulation results indicate that, with the proposed power allocation scheme, the sum capacity of the secondary network can be improved compared to conventional SLNR-precoding-based power allocation schemes.