Hostname: page-component-76d6cb85b7-8p85h Total loading time: 0 Render date: 2026-07-13T03:35:04.735Z Has data issue: false hasContentIssue false

Design of ultra-wideband antenna based on one-dimensional convolutional neural network

Published online by Cambridge University Press:  21 March 2025

Pengcheng Gao
Affiliation:
College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu, China
Weigang Wang*
Affiliation:
College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu, China
Yuanjian Liu
Affiliation:
College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu, China
Qijian Liu
Affiliation:
College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu, China
Yonghao Chen
Affiliation:
College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu, China
Xiangming Yan
Affiliation:
Portland Institute, Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu, China
*
Corresponding author: Weigang Wang; Email: wangwg2020@163.com

Abstract

This paper presents a notched ultra-wideband antenna designed to suppress interference from narrowband communication systems. The antenna features a defected ground structure and a stepped microstrip feedline for improved impedance matching and enhanced bandwidth. A bent slot structure is incorporated into the radiating patch to achieve the band-notched characteristic. It has a wide tunable frequency range which allows for flexible adjustment of the notch frequency. Traditional optimization methods, such as numerical analysis, are computationally expensive and inefficient, while heuristic algorithms are less precise. To address these challenges, an improved one-dimensional convolutional neural network (1DCNN-IPS) model is proposed for optimizing the bent slot design more efficiently. The trained 1DCNN-IPS model can accurately predict the antenna’s electromagnetic parameters, reducing mean squared error and training times compared to traditional methods. This provides an efficient and precise solution for antenna structural optimization.

Information

Type
Research Paper
Copyright
© The Author(s), 2025. Published by Cambridge University Press in association with The European Microwave Association.

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable