Analysis of Electromagnetic Testing and Defect Detection Tools for 5G Generation Radio Antennas
Year: 2025
Type: article
Abstract: The purpose of this article is to review the key elements of fifth-generation (5G) wireless systems and the methods used for electromagnetic testing and defect detection of active antenna systems. Due to the introduction of massive MIMO technologies and millimeter-wave (mmWave) communication in 5G networks, traditional testing approaches are no longer effective, prompting the need for new non-invasive methods. The article aims to analyze current approaches to post-production non-invasive testing of active antenna systems and explore the potential of machine learning elements to enhance the efficiency of defect detection. The study begins with an overview of the technological features of 5G systems, including the use of millimeter-wave frequency bands, massive MIMO antenna arrays, and beamforming techniques, which have increased the complexity of telecommunication equipment. A classification of electromagnetic testing methods is provided, comparing conductive testing methods with non-invasive Over-The-Air (OTA) techniques such as Direct Far-Field (DFF), Compact Antenna Test Range (CATR), and Near-Field to Far-Field (NFTF) transformations. The paper also discusses the use of electromagnetic scanning, including Relative Element Value (REV) methods, for diagnosing antenna defects without physical connections. Given the increasing volume of testing data, the work highlights the integration of machine learning techniques, including deep neural networks and convolutional networks, for defect classification, big data analysis for pattern detection, and predictive analytics for forecasting antenna failures. Special attention is given to how artificial intelligence methods enhance the effectiveness and accuracy of defect diagnosis in modern 5G systems. In conclusion, the study shows that the growing complexity of active antenna systems in 5G networks necessitates the evolution of testing methodologies. Traditional conductive approaches are no longer viable for large-scale production, and non-invasive OTA testing has become the standard. Electromagnetic scanning, combined with machine learning techniques, significantly improves defect detection rates and diagnostic accuracy. Integrating intelligent diagnostic methods into the 5G antenna testing process enables better quality control, reduced production losses, and enhanced reliability of telecommunication equipment.
Authors: Viktor Bondarchuk, Anatolii Matsui, Viktor Kalich
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4411385236