CNN Based Hybrid Precoding for MmWave MIMO Systems With Adaptive Switching Module and Phase Modulation Array

Hybrid precoding is considered as a promising candidate to balance between the sum rate and the power and cost in the mmWave MIMO systems. However, it still suffers from huge power consumption due to considerable radio frequency links and phase shifters. In this study, an energy-efficient adaptive s...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on wireless communications Vol. 21; no. 12; pp. 10489 - 10501
Main Authors: Hei, Yongqiang, Liu, Chao, Li, Wentao, Ma, Longyuan, Lan, Maomao
Format: Journal Article
Language:English
Published: New York IEEE 01.12.2022
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
adaptive switching network
Adaptive systems
Algorithms
Arrays
Artificial neural networks
convolutional neural network
Convolutional neural networks
Costs
Energy efficiency
Hybrid power systems
Hybrid precoding
matching theory
Millimeter waves
MIMO communication
Modules
Optimization
Phase modulation
Phase shifters
Power consumption
Power management
Precoding
Radio frequency
Switches
Switching
ISSN:1536-1276, 1558-2248
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Hybrid precoding is considered as a promising candidate to balance between the sum rate and the power and cost in the mmWave MIMO systems. However, it still suffers from huge power consumption due to considerable radio frequency links and phase shifters. In this study, an energy-efficient adaptive switching module (ASM) and phase modulation array (PMA) assisted hybrid precoding scheme is proposed, in which ASM is introduced to enhance the energy efficiency while PMA is employed to implement analog precoding to reduce power consumption. The ASM-PMA hybrid precoder design problem related to three matrix variables is formulated. Besides, the joint ASM-PMA hybrid precoder/combiner design at both transmitter and receiver is also investigated. Owing to the tough task of joint optimization of matrix variables, we can divide it into two sub-problems. Correspondingly, alternating optimization hybrid precoding (AOHP) algorithm is suggested to solve the PMA. Then, matching theory and convolutional neural network (CNN) based algorithms are respectively proposed to seek the optimal ASM. For the CNN based scheme, relaxed adaptive switching CNN (RAS-CNN) is designed to solve the binary integer problem in the offline train stage, while in the online deployment stage, ideal adaptive switching CNN (IAS-CNN) is adopted. Furthermore, CNN accepts estimated channel as input to yield robust precoding performance and lower computational complexity. Simulation results verify that, the proposed hybrid precoding can outperform other schemes in energy efficiency with satisfactory spectral efficiency.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1536-1276
1558-2248
DOI:10.1109/TWC.2022.3184326