Hybrid Precoding for Beamspace MIMO Systems With Sub-Connected Switches: A Machine Learning Approach

Abstract

By employing lens antenna arrays, the number of radio frequency (RF) chains in millimeterwave (mmWave) communications can be significantly reduced. However, most existing studies consider the phase shifters (PSs) as the main components of the analog beamformer, which may result in a significant loss of energy efficiency (EE). In this paper, we propose a switch selecting network to solve this issue, where the analog part of the beamspace MIMO system is realized by a sub-connected switch selecting network rather than the PS network. Based on the proposed architecture and inspired by the cross-entropy (CE) optimization developed in machine learning, an optimal hybrid cross-entropy (HCE)-based hybrid precoding scheme is designed to maximize the achievable sum rate, where the probability distribution of the hybrid precoder is updated by minimizing CE with unadjusted probabilities and smoothing constant. Simulation results show that the proposed HCE-based hybrid precoding can not only effectively achieve the satisfied sum-rate, but also outperform the PSs schemes concerning energy efficiency.

SINR Maximization Beam Selection for mmWave Beamspace MIMO Systems

“…We consider a scenario where the total power consumed by the system include the total transmit power and the total power consumed by the selected RF chains. In this paper, the energy efficiency η is modelled as follows [33], [34]…”

Spectrum and Energy Efficiency Optimization for Hybrid Precoding-Based SWIPT-Enabled mmWave mMIMO-NOMA Systems

“…Hence, considered a promising technology for the fifth-generation (5G) and beyond wireless communication systems [2], [5], [6]. In conventional multiantenna systems (i.e., for cellular frequency band 2 − 3 GHz), multi-stream beamforming, i.e., precoding, is entirely realized in the digital domain at baseband to cancel interference among several data streams [5], [7]. For such conventional full-digital precoding, in which each antenna connects to a dedicated energy-intensive radio frequency (RF) chain [2], [5], [7]- [9], may not be a practical solution for mmWave mMIMO systems due to the excessive power consumption and prohibitive hardware complexity.…”

“…In conventional multiantenna systems (i.e., for cellular frequency band 2 − 3 GHz), multi-stream beamforming, i.e., precoding, is entirely realized in the digital domain at baseband to cancel interference among several data streams [5], [7]. For such conventional full-digital precoding, in which each antenna connects to a dedicated energy-intensive radio frequency (RF) chain [2], [5], [7]- [9], may not be a practical solution for mmWave mMIMO systems due to the excessive power consumption and prohibitive hardware complexity. Meanwhile, the large antenna array in mMIMO systems, which can be pack in small-formfactor at mmWave frequencies enables precoding multiple data streams and provides the beamforming gain needed to overcome path loss [2], [5].…”

“…Meanwhile, the large antenna array in mMIMO systems, which can be pack in small-formfactor at mmWave frequencies enables precoding multiple data streams and provides the beamforming gain needed to overcome path loss [2], [5]. Hence, mmWave mMIMO systems preclude such full digital precoding transceiver architecture at present, and will almost certainly use hybrid precoding to realize beamforming with fewer RF chains to reduce energy consumption [5], [7], [10]. Specifically, the main idea of hybrid precoding is employing a linear network of variable phase shifters in the RF domain along with the baseband digital precoding [5], [7], [10]- [16], which makes analog processing in the RF domain more attractive.…”

“…Hence, mmWave mMIMO systems preclude such full digital precoding transceiver architecture at present, and will almost certainly use hybrid precoding to realize beamforming with fewer RF chains to reduce energy consumption [5], [7], [10]. Specifically, the main idea of hybrid precoding is employing a linear network of variable phase shifters in the RF domain along with the baseband digital precoding [5], [7], [10]- [16], which makes analog processing in the RF domain more attractive. Based on the RF-antenna mapping strategy for mmWave mMIMO systems, which defines the number of required phase shifters [2], [5], [11], the hybrid precoding transceiver architectures can be classified as either partially-connected or fully-connected structures.…”