多连接结构的大规模MIMO混合预编码

Hybrid Precoding with Multiple-connected Structure for Massive MIMO Systems

为在大规模多输入多输出(Multiple-Input Multiple-Output,MIMO)系统中更灵活利用空间资源以提升频谱效率,提出了一种基于多连接结构的低复杂度混合预编码方案。该方案允许每个射频链路连接的天线存在交集。进一步,通过直接提取空口信道的共轭相位设计模拟预编码矩阵,并提出串行最大化动态子阵列增益算法,以调节射频链路与发射天线的连接关系,从而获得更高的空间分集增益;再对等效信道实施迫零数字预编码。仿真结果表明,所提方案能以更低硬件成本取得超过传统全连接结构方案的频谱效率,以相当的复杂度较传统部分连接结构方案提升频谱效率20%~390%,且对相位量化误差不敏感。

In order to improve the spectral efficiency of massive multiple-input multiple-output(MIMO)systems by more flexibly utilizing spatial resources,a low complexity hybrid precoding scheme based on the multiple-connected structure is proposed.The scheme allows the antennas connected to each radio frequency(RF)chain overlapping.Furthermore,the analog precoding matrix is firstly designed by directly extracting the conjugate phase of the air interface channel,and an algorithm successively maximizing the dynamic subarray gain is proposed to adjust the connection between the RF chains and the transmitting antennas,thereby obtaining higher spatial diversity gain.Then a zero-forcing digital precoding is implemented for the equivalent channel.The simulation results show that the proposed scheme can achieve higher spectral efficiency than traditional fully-connected structure schemes at lower hardware cost,improve spectral efficiency significantly by 20% to 390% compared with traditional partially-connected structure schemes at comparable complexity,and is also insensitive to the phase quantized error.