针对毫米波大规模MIMO系统采用全数字预编码时,所需射频链路数量过多而导致能量消耗高的问题,提出了一种基于透镜的波束选择方案。该方案首先通过分析用户受干扰的可能性,将所有的用户分为干扰用户组和非干扰用户组,然后对于非干扰用户,直接利用最大功率准则进行波束选取,而对于干扰用户,则通过低复杂度增量算法选择合适的波束使系统和速率最大化。仿真结果表明,在有效减少系统所需射频链路数量和降低计算复杂度的基础上,该方案的系统和速率能够达到接近全数字预编码方案的水平,并且能够获得更高的能量效率。
To solve the high energy consumption problem of millimeter wave massive MIMO systems using traditional full digital precoding caused by too many required radio frequency chains, a beam selection scheme based on lens was proposed. The proposed beam selection scheme divides all the users into the two groups of interference-users as well as noninterference-users, and then, for noninterference-users, uses the maximum power criterion to conduct beam selection, while for interference users, uses the low complexity incremental algorithm to select appropriate beams to maximize the system sum-rate. The simulation results show that the proposed beam selection scheme can nearly reach the full digital precoding scheme' s system sum-rate on the basis of effectively reducing the number of re- quired RF chains and the computation complexity.