Login | Register

Hybrid Beamforming Design for Millimeter Wave Massive MIMO Communications

Title:

Hybrid Beamforming Design for Millimeter Wave Massive MIMO Communications

Martignago Zilli, Guilherme ORCID: https://orcid.org/0000-0003-1228-4920 (2021) Hybrid Beamforming Design for Millimeter Wave Massive MIMO Communications. PhD thesis, Concordia University.

[thumbnail of Zilli_PhD_S2021.pdf]
Preview
Text (application/pdf)
Zilli_PhD_S2021.pdf - Accepted Version
Available under License Spectrum Terms of Access.
10MB

Abstract

Wireless connectivity is a key driver for the digital transformation that is changing how people communicate, do business, consume entertainment and search for information. As the world advances into the fifth generation (5G) and beyond-5G (B5G) of wireless mobile technology, new services and use cases are emerging every day, bringing the demand to expand the broadband capability of mobile networks and to provide ubiquitous access and specific capabilities for any device or application. To fulfill these demands, 5G and B5G systems will rely on innovative technologies, such as the ultradensification, the mmWave, and the massive MIMO. To bring together these technologies, 5G and B5G systems will employ hybrid analog-digital beamforming, which separates the signal processing into the baseband (digital) and the radio-frequency (analog) domains. Unlike conventional beamforming, where every antenna is connected to an RF chain, and the signal is entirely processed in the digital domain, hybrid beamforming uses fewer RF chains than the total number of antennas, resulting in a less expensive and less energy-consuming design. The analog beamforming is usually implemented using switching networks or phase-shifting networks, which impose severe hardware constraints making the hybrid beamforming design very challenging. This thesis addresses the hybrid analog-digital beamforming design and is organized into three parts.
In the first part, two adaptive algorithms for solving the switching-network-based hybrid beamforming design problem, also known as the joint antenna selection and beamforming (JASB) problem, are proposed. The adaptive algorithms are based on the minimum mean square error (MMSE) and minimum-variance distortionless response (MVDR) criteria and employ an alternating optimization strategy, in which the beamforming and the antenna selection are designed iteratively. The proposed algorithms can attain high levels of SINR while strictly complying with the hardware limitations. Moreover, the proposed algorithms have very low computational complexity and can track channel variations, making them suitable for non-stationary environments. Numerical simulations have validated the effectiveness of the algorithms in different operation scenarios.
The second part addresses the phase-shifting-network-based hybrid beamforming design for narrowband mmWave massive MIMO systems. A novel joint hybrid precoder and combiner design is proposed. The analog precoder and combiner design is formulated as constrained low-rank channel decomposition, which can simultaneously harvest the array gain provided by the massive MIMO system and suppress intra-user and inter-user interferences. The constrained low-rank channel decomposition is solved as a series of successive rank-1 channel decomposition, using the projected block coordinate descent method. The digital precoder and combiner are obtained from the optimal SVD-based solution for the single-user case and the regularized channel diagonalization method for the multi-user case. Simulation results have demonstrated that the proposed design can consistently attain near-optimal performance and provided important insights into the method's convergence and its performance under practical phase-shifter quantization constraints.
Finally, the phase-shifting-network-based hybrid beamforming design for frequency-selective mmWave massive MIMO-OFDM systems is considered in the third part. The hybrid beamforming design for MIMO-OFDM systems is significantly more challenging than for narrowband MIMO systems since, in these systems, the analog precoder and combiner are shared among all subcarriers and must be jointly optimized. Thus, by leveraging the OFDM systems' multidimensional structure, the analog precoder and combiner design is formulated as constrained low-rank Tucker2 tensor decomposition and solved by a successive rank-(1,1) tensor decomposition using the projected alternate least square (ALS) method. The digital precoder and combiner are obtained on a per-subcarrier basis using the techniques presented in the second part. Numerical simulations have confirmed the design effectiveness, demonstrating its ability to consistently attain near-optimal performance and outperform other existing design in nearly all scenarios. They also provided insights into the convergence of the proposed method and its performance under practical phase-shifter quantization constraints and highlighted the differences between this design and that for the narrowband massive MIMO systems.

Divisions:Concordia University > Gina Cody School of Engineering and Computer Science > Electrical and Computer Engineering
Item Type:Thesis (PhD)
Authors:Martignago Zilli, Guilherme
Institution:Concordia University
Degree Name:Ph. D.
Program:Electrical and Computer Engineering
Date:April 2021
Thesis Supervisor(s):Zhu, Wei-Ping
ID Code:988317
Deposited By: Guilherme Martignago Zilli
Deposited On:29 Jun 2021 23:06
Last Modified:29 Jun 2021 23:06
All items in Spectrum are protected by copyright, with all rights reserved. The use of items is governed by Spectrum's terms of access.

Repository Staff Only: item control page

Downloads per month over past year

Research related to the current document (at the CORE website)
- Research related to the current document (at the CORE website)
Back to top Back to top