毫米波大规模MIMO-NOMA系统功率分配和混合预编码设计

Design of Power Allocation and Hybrid Precoding for Millimeter-wave Massive MIMO-NOMA Systems

针对基于非正交多址接入(Non-Orthogonal Multiple Access,NOMA)的毫米波大规模多输入多输出(Multiple Input Multiple Output,MIMO)系统中边缘用户可达速率较低的问题,提出了一种联合功率分配和混合预编码的优化方案来保证边缘用户在内的所有用户都具有较好的速率性能。基于最大最小公平性准则,设定了一个含有复杂目标函数和高维非凸约束的优化问题。采用固定变量法将该非凸问题转换为功率分配和混合预编码2个子优化问题处理。对于功率分配的设计,通过卡罗需-库恩-塔克(Karush-Kuhn-Tucker,KKT)条件得到了封闭形式下的最优功率分配系数;对于混合预编码的设计,提出了基于迫零(Zero Forcing,ZF)和压缩边界粒子群优化(Boundary Compressed-Particle Swarm Optimization,BC-PSO)算法相结合的求解方法,得到混合预编码的次优解。提出了一种交替优化算法来交替优化功率分配和混合预编码,直到满足设定的精度要求。仿真结果表明,该方案的最大最小速率优于传统的NOMA方案。

To solve the problem of low achievable rate of edge users in millimeter wave massive Multiple Input Multiple Output(MIMO)systems based on Non-Orthogonal Multiple Access(NOMA),an optimization scheme of joint power allocation and hybrid precoding is proposed to ensure that all users including edge users have better rate performance.Firstly,based on the principle of maximum-minimum fairness,an optimization problem with complex objective functions and high-dimensional non-convex constraints is formulated.Then,the fixed variable method is used to transform the non-convex problem into two sub-optimization problems of power allocation and hybrid precoding.For the design of power allocation,the optimal power allocation coefficient under closed form is obtained by using Karush-Kuhn-Tucker(KKT)condition.For the design of hybrid precoding,a solution method based on the combination of Zero Forcing(ZF)and Boundary Compressed-Particle Swarm Optimization(BC-PSO)algorithm is proposed to obtain the sub-optimal solution of hybrid precoding.Finally,an alternative optimization algorithm is proposed to alternately optimize power allocation and hybrid precoding until the setting accuracy requirements are met.The simulation results show that the maximum and minimum rate of this scheme is better than that of the traditional NOMA scheme.