下行非正交多址系统低复杂度功率分配算法设计

Low-complexity power allocation algorithm for downlink non-orthogonal multiple access

近年来,非正交多址接入(Non-Orthogonal Multiple Access,NOMA)作为5G关键技术而得到广泛关注。相比于已有的正交多址接入,NOMA可承载用户数不受资源数目的限制,能够满足5G应用场景中大规模连接需求,同时可以提高系统的频谱效率。由于NOMA通过功率域区分不同用户,功率分配方法直接影响着系统的性能,文中旨在设计有效的功率分配方案优化系统吞吐量。鉴于已有算法较高的复杂度,文中提出一种改进的功率分配算法,将功率分配问题拆分为两个子问题,即先进行带间功率分配,再进行带内功率分配,通过联合设计迭代注水功率分配和比例公平调度算法分别求解两个子优化问题。另外,考虑到信号检测分离过程存在残余干扰,还对差错传播模型进行了仿真分析。仿真结果表明,改进的算法虽然性能上略低于最优迭代注水功率分配算法,但是大幅度降低了计算复杂度,具有一定的实践意义。

In recent years,non-orthogonal multiple access（ NOMA） has received wide attention as the key technology of 5G. Compared with the existing orthogonal multiple access,NOMA can carry users without the limitation of the number of resources,which can meet the large-scale connection requirements in 5G application scenarios and improve the spectrum efficiency of the system. NOMA can distinguish different users through power domain,so power allocation directly affects the performance of the system. An effective power allocation is used to optimize system throughput. Due to the high complexity of existing algorithms,an improved power allocation algorithm is proposed. Power allocation is split into two sub-problems,called the power allocation inter-subbands and intra-subbands,which are solved by joint design of iterative water-filling algorithm and proportional fair scheduling algorithm. In addition,considering the existence of residual interference in the process of signal detection and separation,the error propagationmodel is simulated and analyzed. Simulation results show that the performance of the improved algorithm is slightly lower than that of the iterative water-filling algorithm,but it can reduce the computational complexity. Thus it has a practical significance.