李光远*,付辉*,张凯**,韩芳*,徐顺清***,孟溪****.基于毫米波大规模MIMO的共享型阵列混合波束成形算法[J].高技术通讯(中文),2025,35(2):175~182 |
基于毫米波大规模MIMO的共享型阵列混合波束成形算法 |
Hybrid precoding algorithms base on the fully-connected structure in millimeter wave massive MIMO systems |
|
DOI:10. 3772 / j. issn. 1002-0470. 2025. 02. 007 |
中文关键词: 大规模多输入输出; 毫米波; 波束成形; 频谱效率 |
英文关键词: massive multiple input multiple output(MIMO), millimeter wave, beamforming, spectrum efficiency |
基金项目: |
作者 | 单位 | 李光远* | (*黄河科技学院工学部郑州 450063)
(**北京邮电大学信息与通信工程学院北京 100876)
(***中国科学院计算技术研究所北京 100190)
(****中国电信股份有限公司研究院北京 100032) | 付辉* | | 张凯** | | 韩芳* | | 徐顺清*** | | 孟溪**** | |
|
摘要点击次数: 5 |
全文下载次数: 4 |
中文摘要: |
考虑到毫米波大规模多输入输出(multiple input multiple output,MIMO)系统中传统的全数字波束成形算法能耗较高、成本较大且模拟预编码算法性能较差,提出了一种共享型阵列的混合波束成形算法。首先通过信道矩阵的奇异值分解,提取它们的相位信息作为模拟预编码和模拟合成矩阵,然后将得到的模拟预编码和模拟合成矩阵与信道矩阵一起作为等效信道矩阵,再次进行奇异值分解,即可得到相应的数字预编码矩阵和数字合成矩阵。仿真结果表明,本文所提混合波束成形算法,当射频链路数与数据流数配置相等时,其频谱效率和能效都要明显优于传统的混合波束成形算法,且在工程上易于实现。 |
英文摘要: |
Considering the high cost and power consumption of the traditional fully-digital beamforming algorithm in millimeter wave massive multiple input multiple output(MIMO) system ,and the poor performance of analog precoding algorithms, a hybrid beamforming algorithm based on the full-connected structure is proposed. Firstly, the phase information is extracted as analog precoding and combining matrix by the singular value decomposition of channel matrix. Then the obtained analog precoding and combining matrix are used together with the original channel matrix as equivalent channel matrix to obtain the corresponding digital precoding and combining matrix by singular value decomposition again. Simulation results demonstrate the proposed hybrid beamforming algorithm can achieve better spectral and energy efficiency than the traditional hybrid beamforming algorithm, while the number of radio frequency chain and data streams are equal. Moreover, the whole singular value decomposition(W-SVD) algorithm is easy to implement in engineering. |
查看全文
查看/发表评论 下载PDF阅读器 |
关闭 |