基于粒子群算法的下行CR-NOMA网络资源分配

将非正交多址接入(Non-orthogonal Multiple Access,NOMA)技术应用于认知无线电(Cognitive Radio,CR)次网络,使次用户的信号在功率域叠加,可以进一步提高次网络的吞吐量。为此,将粒子群算法(Particle Swarm Optimization,PSO)应用于底层模式的CR-NOMA网络进行资源分配,并分为子信道分配和功率分配两个步骤。在子信道分配中,使用结合遗传算法思想的粒子群算法提高算法的全局搜索能力。在此基础上,使用基于罚函数的粒子群算法对子信道功率和信道内用户功率进行分配。仿真结果表明,提出的基于粒子群算法的CR-NOMA网络资源分配相比以往算法能获得更高的次网络吞吐量。

Particle Swarm Optimization-based Resource Allocation of Downlink CR-NOMA Network

The application of NOMA(Non-Orthogonal Multiple Access)technology to CR(Cognitive Radio)secondary network enables the signals of secondary users to be superimposed in the power domain,which can further improve the throughput of the secondary network.To this end,PSO(Particle Swarm Optimization)algorithm is applied to the underlay CR-NOMA network for resources allocation,and this resource allocation is divided into two steps involving subchannel allocation and power allocation.In the subchannel allocation,PSO combined with GA(genetic algorithm)is used to improve the global search ability of the algorithm,and with this,PSO based on penalty function is then used to allocate the subchannel power and user power in the channel.Simulation results indicate that the proposed CR-NOMA network resource allocation algorithm based on PSO can achieve more throughput of secondary network than previous algorithms.