认知无线电系统中的多中继分布式波束成形方法
摘要
In this paper, the cognitive scenario where a source-destination pair of secondary users coexists with a licensed transmitter-receiver pair with the help of a group of cognitive relay nodes is considered. To exploit the spatial spectrum holes, we study the distributed beamforming of multiple cognitive relay nodes each equipped with a single antenna for the maximization of signal-to-interference-plus-noise ratio (SINR) at the cognitive destination, while keeping the interference power at the licenced receiver below a certain threshold and satisfying the individual power constraint. Two beamforming schemes are proposed: 1) the optimal scheme maximizing SINR; and 2) a suboptimal zero-forcing scheme. The optimal beamforming problem can be reformulated as a quasi-convex optimization problem using semide nite relaxation (SDR), then the bisection and interior point methods can be used to solve this problem. It can also be proved that SDR yields the optimal solution to the original problem. The suboptimal zero-forcing scheme eliminates the interference at the licensed receiver as well as the signals from the licensed transmitter. It has a low complexity without iterations and simple constraint functions. Numerical results are presented to show the impacts of the number of relay nodes, the maximum transmit power of cognitive nodes, and the threshold of power of interferences at the licenced receiver to the average achievable rates of the two schemes. Furthermore, the performance will be improved if the impact of signals from licenced transmitter is taken into account, which is veri ed through simulation experiments.