A comprehensive study on channel estimation techniques using machine learning for massive MIMO systems

Estimating the Wireless channel parameters using the received signal to noise ratio feedback is an effective method in realistic wireless multiple-input multiple-output (MIMO) systems. In the future, being 5G-and-beyond wireless networks, the significance of machine learning algorithms in improving...

Full description

Saved in:
Bibliographic Details
Main Authors: Madurai, Kaaviya, Sivaraman, Deepa, Balasubramaiyan, Bhuvaneswari, Jebakumar, Jeneetha Jebanazer
Format: Conference Proceeding
Language:English
Subjects:
Algorithms
Computer simulation
Deep learning
Estimation
Machine learning
MIMO communication
Network reliability
Signal to noise ratio
Support vector machines
System effectiveness
Wireless networks
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Estimating the Wireless channel parameters using the received signal to noise ratio feedback is an effective method in realistic wireless multiple-input multiple-output (MIMO) systems. In the future, being 5G-and-beyond wireless networks, the significance of machine learning algorithms in improving the reliability of wireless networks is reviewed in this paper. Channel estimation using Least squares (LS) and the minimum mean square error (MMSE) being the most popular and commonly used techniques suffer due to the relatively high estimation error. Various machine learning and deep learning-based channel estimation techniques are studied in this paper and the reported works suggest using Support Vector Machines (SVM) to address the practical issue of a limited amount of actual channel samples required for training. An SVM-based model with pilot-assisted channel estimation delivers the best performance in both computer simulation and transmission testing of massive MIMO systems, making it a viable choice for future wireless networks and massive MIMO systems.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0152775