| 注册
首页|期刊导航|无线电工程|基于分数低阶极坐标图和视觉变换网络的调制识别方法

基于分数低阶极坐标图和视觉变换网络的调制识别方法

孟水仙 朱明波 刘喆 周嘉晨 尚睿 栾声扬

无线电工程2025,Vol.55Issue(6):1215-1222,8.
无线电工程2025,Vol.55Issue(6):1215-1222,8.DOI:10.3969/j.issn.1003-3106.2025.06.008

基于分数低阶极坐标图和视觉变换网络的调制识别方法

Automatic Modulation Classification Based on Fractional Lower-Order Polar Plane and Vision Transformer

孟水仙 1朱明波 2刘喆 1周嘉晨 2尚睿 2栾声扬2

作者信息

  • 1. 内蒙古自治区无线电监测站,内蒙古 呼和浩特 010011
  • 2. 江苏师范大学 电气工程及自动化学院,江苏 徐州 221116
  • 折叠

摘要

Abstract

Automatic Modulation Classification(AMC)serves as a critical link between signal detection and demodulation in communication systems,thus playing an essential role.Existing methods commonly assume that the noise follows a Gaussian distribution.However,impulsive noise with sharp peaks is prevalent in practical environment,leading to the performance degradation of traditional approaches due to mismatch of the noise model.To address this issue,a novel AMC method is proposed,which utilizes the newly proposed Fractional Lower-Order Polar Plane(FLOPOLA)as the input feature and employs the newly designed Vision Transformer(ViT)as the lightweight feature extractor and classifier.Monte-Carlo experimental results indicate that the proposed method achieves high classification accuracy under various impulsive noise conditions.Results further demonstrate that the proposed FLOPOLA can effectively suppress the outliers in the impulsive noise and the designed ViT can effectively achieve feature extraction and classification.The proposed method can effectively solve the AMC problem under impulsive noise conditions and has potential application value for scenarios demanding both high accuracy and low power consumption.

关键词

Alpha稳定分布/自动调制识别/分数低阶统计量/极坐标图

Key words

Alpha-stable distribution/AMC/fractional lower-order statistics/polar plane

分类

信息技术与安全科学

引用本文复制引用

孟水仙,朱明波,刘喆,周嘉晨,尚睿,栾声扬..基于分数低阶极坐标图和视觉变换网络的调制识别方法[J].无线电工程,2025,55(6):1215-1222,8.

基金项目

国家自然科学基金(61801197) (61801197)

江苏省自然科学基金(BK20181004) (BK20181004)

江苏省高等学校自然科学研究基金(18KJB510012) (18KJB510012)

2022年江苏高校青蓝工程 National Natural Science Foundation of China(61801197) (61801197)

Jiangsu Provincial Natural Science Foundation of China(BK20181004) (BK20181004)

Natural Science Research Project of Jiangsu Higher Education Institutions(18KJB510012) (18KJB510012)

2022 Qinglan Project of Jiangsu Universities ()

无线电工程

1003-3106

访问量0
|
下载量0
段落导航相关论文