基于协同注意力机制的跨模态特征融合干扰辨识方法

董阳阳; 杨孔婧; 冯家琛

doi:10.12382/bgxb.2025.0725

您当前的位置：

首页 >

文章列表页 >

基于协同注意力机制的跨模态特征融合干扰辨识方法

更新时间：2026-05-06

- 基于协同注意力机制的跨模态特征融合干扰辨识方法
- Co-attention-based Mechanism Cross-modal Feature Fusion for Jamming Signal Identification
- 兵工学报 2026年47卷第4期页码：250725
- 作者机构：
  
  西安电子科技大学电子工程学院，陕西西安 710071
- 作者简介：
  
  通信作者邮箱：yydong@xidian.edu.cn
- 基金信息：
- DOI：10.12382/bgxb.2025.0725
  中图分类号： TN973
- 收稿：2025-08-08，
  
  网络首发：2026-01-22，
  
  纸质出版：2026-04
- 稿件说明：
移动端阅览
董阳阳, 杨孔婧, 冯家琛. 基于协同注意力机制的跨模态特征融合干扰辨识方法[J]. 兵工学报, 2026,47(4):250725.

DONG Yangyang, YANG Kongjing, FENG Jiachen. Co-attention-based Mechanism Cross-modal Feature Fusion for Jamming Signal Identification[J]. Acta Armamentarii, 2026, 47(4): 250725.
董阳阳, 杨孔婧, 冯家琛. 基于协同注意力机制的跨模态特征融合干扰辨识方法[J]. 兵工学报, 2026,47(4):250725. DOI： 10.12382/bgxb.2025.0725.

DONG Yangyang, YANG Kongjing, FENG Jiachen. Co-attention-based Mechanism Cross-modal Feature Fusion for Jamming Signal Identification[J]. Acta Armamentarii, 2026, 47(4): 250725. DOI： 10.12382/bgxb.2025.0725.

摘要

针对现代雷达干扰策略自动化、样式多样化、参数精准化引起的干扰辨识效能下降问题，提出一种基于协同注意力机制的跨模态特征融合干扰辨识模型。该模型通过协同注意力机制融合干扰信号时频模态和距离多普勒模态的深度特征，充分挖掘出不同模态间的互补信息；同时结合迁移学习策略预训练残差神经网络，有效克服小样本训练数据限制，显著提升复杂干扰信号的辨识性能。仿真实验结果表明，所提方法在低干噪比和复杂干扰环境下仍保持较高的识别准确率，干噪比为-3dB时识别准确率不低于96%；在小样本训练条件下，干噪比为-3dB及以上时识别准确率不低于85%，验证了该方法的有效性和鲁棒性。

Abstract

To address the degradation in the jamming identification performance of radar caused by the automation

diversification and parameter precision of modern radar jamming strategies

a cross-modal feature fusion recognition model based on a collaborative attention mechanism is proposed. The deep features from the time-frequency modality and range-Doppler modality of jamming signals are effectively integrated through the collaborative attention mechanism

allowing the complementary information between different modalities to be fully exploited. Meanwhile

a residual neural network is pre-trained using a transfer learning strategy to effectively mitigate the limitations imposed by small-sample training data

leading to significant improvement in the identification performance for complex jamming signals. Simulated results demonstrate that the proposed method maintains a higher recognition accuracy under low jamming-to-noise ratio (JNR) and complex jamming scenarios. A recognition accuracy of no less than 96% is achieved at a JNR of -3dB. The recognition accuracy remains no less than 85% at JNR values of-3dB and above under small-sample training conditions

which validates the effectiveness and robustness of the proposed method.

关键词

Keywords

references

WEI J J, YU L, WEI Y S, et al. A rapid recognition method for radar active jamming based on the Hybrid 3CNN-Transformer model [J].IET Radar, Sonar & Navigation, 2024, 18 (11): 2377-2395.

郭文杰，吴振华，曹宜策，等.多域浅层特征引导下雷达有源干扰多模态对比识别方法[J].雷达学报（中英文），2024，13（5）：1004-1018.

GUO W J, WU Z H, CAO Y C, et al. Multidomain characteristicguided multimodal contrastive recognition method for active radar jamming [J]. Journal of Radars (Bilingual), 2024, 13 (5): 1004-1018. (in Chinese)

WU Z H, WANG T X, CAO Y C, et al. Radar active deception jamming recognition based on Siamese squeeze wavelet attention network [J]. IET Radar, Sonar & Navigation, 2023, 17 (12): 1886-1898.

ZHU C J, LI A, LÜ Q Z, et al. Radar deception jamming recognition based on GCN and Transformer[C]∥Proceedings of the 2024 IEEE International Conference on Signal, Information and Data Processing. Zhuhai, China:IEEE, 2024:1-5.

阳鹏飞，何羚，王茜，等.基于混合信号多域特征和Transformer的干扰识别[J].系统工程与电子技术，2024，46（6）：2138-2145.

YANG P F, HE L, WANG Q, et al. Interference identification based on mixed signal multidomain feature and Transformer framework [J]. Systems Engineering and Electronics, 2024, 46 (6): 2138-2145. (in Chinese)

聂千祁，沙明辉，朱应申.基于改进残差神经网络的雷达信号识别方法[J].系统工程与电子技术，2024，46（10）：3356-3364.

NIE Q Q, SHA M H, ZHU Y S.Radar signal recognition method based on improved residual neural network [J]. Systems Engineering and Electronics, 2024, 46 (10): 3356-3364. (in Chinese)

陈佳音，郭山红，朱海锐，等.基于多特征融合的海面目标智能检测算法[J].现代雷达，2024，46（12）：24-30.

CHEN J Y, GUO S H, ZHU H R, et al. Intelligent detection algorithm for sea surface targets based on multi-feature fusion [J]. Modern Radar, 2024, 46 (12): 24-30. (in Chinese)

LÜ Q Z, FAN H X, LIU J L, et al. Multilabel deep learning-based lightweight radar compound jamming recognition method [J].IEEE Transactions on Instrumentation and Measurement, 2024, 73.DOI:10.1109/TIM.2024.3400337.

ZHONG W F, XIONG H L, HUA Y, et al. TSFANet: temporalspatial feature aggregation network for GNSS jamming recognition [J].IEEE Transactions on Instrumentation and Measurement, 2024, 73.DOI:10.1109/TIM.2024.3375975.

CHEN Z C, ZHU S Q, LI X M, et al. Radar active composite jamming recognition method based on STFT feature extraction [C]∥Proceedings of the 2024 7th International Conference on Information Communication and Signal Processing. Zhoushan, China:IEEE, 2024:582-586.

周东平，阮航，沙明辉，等.基于特征融合的Fast-CNN的复杂波形调制识别[J].系统工程与电子技术，2025，47（7）：2165-2175.

ZHOU D P, RUAN H, SHA M H, et al. Complex waveform modulation recognition based on feature fusion with Fast-CNN [J]. Systems Engineering and Electronics, 2025, 47 (7): 2165-2175. (in Chinese)

WANG J K, WU Z X, OUYANG W H, et al. M2TR:multi-modal multi-scale transformers for deepfake detection[C]∥Proceedings of the 2022 International Conference on Multimedia Retrieval. New York, NY, US:ACM, 2022:615-623.

魏赓力，李凉海，闫海鹏，等.基于多模特征融合的雷达干扰信号识别[J].遥测遥控，2023，44（4）：80-87.

WEI G L, LI L H, YAN H P, et al Radar jamming signal recognition based on multimode feature fusion [J]. Journal of Telemetry, Tracking and Command, 2023, 44 (4): 80-87. (in Chinese)

CHEN H, CHEN H, LEI Z S, et al. Compound jamming recognition based on a dual-channel neural network and feature fusion [J]. Remote Sensing, 2024, 16 (8): 1325.

LUO Z Y, CAO Y H, YEO T S, et al. Few-shot radar jamming recognition network via complete information mining [J]. IEEE Transactions on Aerospace and Electronic Systems, 2024, 60 (3): 3625-3638.

LUO Z Y, CAO Y H, YEO T S, et al. Few-shot radar jamming recognition network via time-frequency self-attention and global knowledge distillation [J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61:1-12.

吕勤哲，全英汇，沙明辉，等.基于集成深度学习的有源干扰智能分类[J].系统工程与电子技术，2022，44（12）：3595-3602.

LÜ Q Z, QUAN Y H, SHA M H, et al Ensemble deep learningbased intelligent classification of active jamming [J]. Systems Engineering and Electronics, 2022, 44 (12): 3595-3602. (in Chinese)

郭立民，黄文青，陈前，等.轻量化的ML-SNet雷达复合干扰识别算法[J].系统工程与电子技术，2025，47（2）：418-427.

GUO L M, HUANG W Q, CHEN Q, et al. Lightweight algorithm of ML-SNet radar compound jamming recognition [J]. Systems Engineering and Electronics, 2025, 47 (2): 418 - 427. (in Chinese)

邵正途，许登荣，徐文利，等.基于LSTM和残差网络的雷达有源干扰识别[J].系统工程与电子技术，2023，45（2）：416-423.

SHAO Z T, XU D R, XU W L, et al. Radar active jamming recognition based on LSTM and residual network [J]. Systems Engineering and Electronics, 2023, 45 (2): 416-423. (in Chinese)

王永，李行健，邓江洲.融合注意力机制的残差神经协同过滤推荐模型[J].运筹与管理，2024，33（10）：201-208.

WANG Y, LI X J, DENG J Z.Residual neural collaborative filtering recommendation based on attention mechanism [J]. Operations Research and Management Science, 2024, 33 (10): 201-208. (in Chinese)

罗皓文，何绍溟，亢有为.一种基于迁移学习的多任务制导算法[J].兵工学报，2024，45（6）：1787-1798.

LUO H W, HE S M, KANG Y W.A multitask guidance algorithm based on transfer learning [J]. Acta Armamentarii, 2024, 45 (6): 1787-1798. (in Chinese)

刘建业.基于注意力与残差机制的深度协同推荐算法及其应用[D].北京：北京交通大学，2024.

LIU J Y.Deep collaborative recommendation algorithm based on attention and residual mechanism and its application [D]. Beijing: Beijing Jiaotong University, 2024. (in Chinese)

赵小强，程伟.基于空间特征交叉融合的轻量级图像超分辨率重建[J].兵工学报，2024，45（4）：1273-1284.

ZHAO X Q, CHENG W.Lightweight image super-resolution reconstruction based on cross-fusion of spatial features [J]. Acta Armamentarii, 2024, 45 (4): 1273-1284. (in Chinese)

浏览量

258

下载量

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

YOLOv11-MFFM-A-OBB：基于多模态特征融合与旋转框优化的船舶尾流迁移学习检测框架

基于改进DeepLabV3+的海上船舶目标分割方法

结合系统辨识和迁移学习的高速旋转弹气动力建模方法

一种基于迁移学习的多任务制导算法

基于迁移学习的供药装置故障诊断方法