基于强化学习的大规模天线阵天线选择算法

Antenna Selection Algorithm of Large-scale Antenna Array Based on Reinforcement Learning

传统天线选择算法过于依赖信道状态信息(CSI),然而以用户为中心的大规模多输入多输出(UC-MMIMO)系统难以获得足够CSI。针对以上矛盾,将强化学习方法引入到天线选择的问题中,提出了一种基于强化学习的天线选择算法。通过仿真说明所提算法相对于传统的天线选择方法对CSI依赖程度大大降低,并且有着更低的算法复杂度。

Traditional antenna selection algorithms rely too much on channel state information(CSI),but user-centric massive multiple-input multiple-output(UC-MMIMO)systems are diffi-cult to obtain sufficient CSI.In view of the above contradictions,the reinforcement learning method is introduced into the problem of antenna selection,and an antenna selection algorithm based on re-inforcement learning is proposed.Simulation results show that the proposed algorithm greatly re-duces the dependence on CSI compared with the traditional antenna selection methods,and has a lower algorithm complexity.