Low complexity asymptotically unitary algorithm for hybrid beamforming in mmWave communication systems  

作　　者：Li Xiaohui Meng Meimei Lin Yingchao Hei Yongqiang 

机构地区：[1]State Key Laboratory of Integrated Service Networks, Xidian University, Xi'an 710071, China [2]Collaborative Innovation Center of Information Sensing and Understanding, Xidian University, Xi'an 710071, China

出　　处：《The Journal of China Universities of Posts and Telecommunications》2017年第1期18-25,共8页中国邮电高校学报（英文版）

基　　金：supported by the National Natural Science Foundation of China(61201134);State Key Science and Research Project(MJ-2014-S-37);the 111 Project(B08038)

摘　　要：In millimeter wave （mmWave） massive multiple-input multiple-output （MIMO） systems, because of the high hardware cost and high power consumption, the traditional fully digital beamforming （DBF） cannot be implemented easily. Meanwhile, analog beamforming which is implemented with phase shifters has high availability but suffers poor performance. Considering the advantages of above two, a potential solution is to design an appropriate hybrid analog and digital beamforming structure, where the available iterative optimization algorithm can get performance close to fully digital processing, but solving this sparse optimization problem faces with a high computational complexity. The key challenge of seeking out hybrid beamforming （HBF） matrices lies in leveraging the trade-off between the spectral efficiency performance and the computational complexity. In this paper, we propose an asymptotically unitary hybrid precoding （AUHP） algorithm based on antenna array response （AAR） properties to solve the HBF optimization problem. Firstly, we get the optimal orthogonal analog and digital beamforming matrices relying on the channel＇s path gain in absolute value by taking into account that the AAR matrices are asymptotically unitary. Then, an improved simultaneously orthogonal matching pursuit （SOMP） algorithm based on recursion is adopted to refine the hybrid combining. Numerical results demonstrate that our proposed AUHP algorithm enables a lower computational complexity with negligible spectral efficiency performance degradation.In millimeter wave （mmWave） massive multiple-input multiple-output （MIMO） systems, because of the high hardware cost and high power consumption, the traditional fully digital beamforming （DBF） cannot be implemented easily. Meanwhile, analog beamforming which is implemented with phase shifters has high availability but suffers poor performance. Considering the advantages of above two, a potential solution is to design an appropriate hybrid analog and digital beamforming structure, where the available iterative optimization algorithm can get performance close to fully digital processing, but solving this sparse optimization problem faces with a high computational complexity. The key challenge of seeking out hybrid beamforming （HBF） matrices lies in leveraging the trade-off between the spectral efficiency performance and the computational complexity. In this paper, we propose an asymptotically unitary hybrid precoding （AUHP） algorithm based on antenna array response （AAR） properties to solve the HBF optimization problem. Firstly, we get the optimal orthogonal analog and digital beamforming matrices relying on the channel＇s path gain in absolute value by taking into account that the AAR matrices are asymptotically unitary. Then, an improved simultaneously orthogonal matching pursuit （SOMP） algorithm based on recursion is adopted to refine the hybrid combining. Numerical results demonstrate that our proposed AUHP algorithm enables a lower computational complexity with negligible spectral efficiency performance degradation.

关 键 词：massive MIMO mmWave communication hybrid beamforming asymptotically unitary low complexity 

分 类 号：TN928[电子电信—通信与信息系统] TN958[电子电信—信息与通信工程]