Low-complexity hybrid precoding for sub-connected millimeter wave massive MIMO systems

• Published in: Signal processing Vol. 219; p. 109381
• Main Authors: Zhao, Lei, Li, Junjie, Huang, Songkang, Wu, Xiaoge, Jiang, Ming
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.06.2024
• Subjects: Energy efficiency; Hybrid precoding; Massive MIMO; Spectral efficiency; Hybrid precoding; Energy efficiency; Spectral efficiency; Massive MIMO
• ISSN: 0165-1684; 1872-7557
• DOI: 10.1016/j.sigpro.2023.109381

Hybrid analog-digital precoding architectures facilitate practical implementation of millimeter wave (mmW) massive multiple-input multiple-output (mmW-mMIMO) systems by reducing the number of radio frequency chains required. The spectral efficiency (SE) and energy efficiency (EE) optimization problems of these systems are typically non-convex and NP-hard, as the joint optimization that involves the analog and digital precoding functions along with the constant modulus constraints required by analog phase shifters, results in highly non-linear problems. To address such problems, we propose a decoupled two-stage design. More specifically, in the first stage, a closed-form approximation of the effective channel is proposed, and thus the SE maximization problem can be transformed to an effective channel gain maximization problem. Then, the analog precoding matrices can be generated to facilitate the second stage, where the digital precoding matrix is devised for maximizing the system’s EE. In addition, the circuit power consumption statistics of the system modules are exploited and a sub-connected mmW-mMIMO architecture is adopted during EE optimization. Finally, simulation results are provided to validate the effectiveness of our proposed mmW-mMIMO scheme with hybrid precoding. •An intractable hybrid precoding design problem of the sub-connected architecture is tackled.•A two-stage processing method for hybrid precoder in mmW-MU-mMIMO systems is proposed.•The complexity of the precoder is reduced by the transformed mmW-MU-mMIMO system.•Closed-form and optimal methods are used to maximize spectral and energy efficiency, respectively.•A dynamic sub-array assignment mechanism is proposed for optimum sub-array/user matching.