报告详情
  1. 活动首页
  2. 报告清单
  3. 报告详情
CWD-AMNet: A Lightweight Multi-Scale Attention Network for Radar Intra-Pulse Modulation Recognition Based on CWD Time-Frequency Images
编号：168 访问权限：仅限参会人 更新：2026-10-19 18:05:54 浏览：5次 张贴报告

报告开始：暂无开始时间（Asia/Shanghai）

报告时间：暂无持续时间

所在会场：[暂无会议] [暂无会议段]

暂无文件

摘要
Radar intra-pulse modulation recognition is an important task in electronic reconnaissance and radar signal analysis. Time-frequency representations provide an effective way to describe non-stationary radar waveforms, but directly applying general-purpose convolutional neural networks to radar time-frequency images may lead to insufficient utilization of signal-specific characteristics. In this paper, a lightweight recognition network named CWD-AMNet is proposed for radar intra-pulse modulation recognition based on Choi-Williams distribution (CWD) time-frequency images. The proposed method uses MobileNetV2 as the backbone and introduces two task-oriented improvements: coordinate attention and multi-scale feature fusion. Coordinate attention enhances the perception of time and frequency directional information, while multi-scale fusion combines local time-frequency details with high-level semantic features. Experiments on a self-constructed radar intra-pulse modulation dataset show that the proposed network improves recognition accuracy over the MobileNetV2 baseline while maintaining a lightweight structure.
关键词
radar signal recognition,intra-pulse modulation,Choi-Williams distribution,MobileNetV2,Coordinate attention,multi-scale feature fusion
报告人
Enzhong Bai
Student Harbin Institute of Technolog

稿件作者
Enzhong Bai Harbin Institute of Technolog
yin hongtao Harbin Institute of Technology
发表评论
验证码 看不清楚，更换一张
全部评论
重要日期

  • 会议日期

    11月06日

    2026

    至

    11月08日

    2026

  • 10月15日 2026

    初稿截稿日期

主办单位
IEEE Instrumentation and Measurement Society
承办单位
Sichuan University
联系方式
登录查看完整联系方式
历届会议
移动端
在手机上打开
小程序
打开微信小程序
客服
扫码或点此咨询