A Distributed Recognition Method for Interference Types based on the Self-attention Mechanism of ViT Network Model

doi:10.12382/bgxb.2025.0127

Abstract

Abstract:

The centralized high-power jamming has the limitations，such as restricted coverage and susceptibility to localization，in modern communication countermeasures.This paper presents a distributed jamming signal recognition method based on an air-sea integrated framework.The proposed method is to use the unmanned aerial vehicles to collect the jamming signals，while a maritime information center conducts multi-source fusion and feature extraction.A multi-stage Vision Transformer (ViT) model with multi-dimensional signal representation，is then designed for the recognition and classification of jamming signals.Experimental results show that the proposed method achieves higher recognition accuracy and efficiency than the conventional methods，while maintaining robustness under low signal-to-noise ratios and diverse interference scenarios.These results demonstrate the method’s adaptability to complex environments and its potential to support rapid decision-making in modern battlefields.This study provides a feasible path for the engineering application of distributed communication jamming technologies and establishes a technical basis for the development of air-sea integrated electronic countermeasure systems.

Key words: distributed communication jamming, air-sea integrated architecture, signal recognition, vision transformer

WANG Xun, WU Zhiqiang, LIU Mingqian. A Distributed Recognition Method for Interference Types based on the Self-attention Mechanism of ViT Network Model[J]. Acta Armamentarii, 2025, 46(S1): 250127-.

Figures/Tables 13

Fig.1 Comparison chart of time-frequency representation

Fig.2 Time-frequency diagram of communication jamming signal

Fig.3 Constellation Diagram of Communication Interference Signal

Fig.4 Structure of ViT network model

Fig.5 Encoder Block structure

Fig.6 Self-attention mechanism used in the ViT network model

Fig.7 Transformer Encoder layer and feedforward neural network structure

Fig.8 Flowchart for interference type recognition based on the self-attention mechanism of ViT network model

Fig.9 Interference Type Recognition Performance Based on the ViT Network Model

Fig.10 Performance Comparison of Confusion Matrices under Different JNRs

Fig.11 Recognition oerformance of communication interference under different frequency offsets

Fig.12 The performance of average signal recognition rate under different channel models

Fig.13 The performance of different communication interference identification methods

References 26

[1]	左浩. 短波通信中分布式干扰的效能分析[J]. 电子测试, 2022, 36(2):55-57.
	ZUO H. Efficacy analysis of distributed interference in shortwave communications[J]. Electronic Test, 2022, 36(2):55-57. (in Chinese)
[2]	定少浒. DRFM有源欺骗干扰识别算法研究[D]. 西安: 西安电子科技大学, 2019.
	DING S H. Research on DRFM active spoofing interference identification algorithm[D]. Xi'an: Xidian University, 2019. (in Chinese)
[3]	WANG Y F, SUN B Y, WANG N. Recognition of radar active-jamming through convolutional neural networks[J]. The Journal of Engineering, 2019, 2019(21):7695-7697.
[4]	韩国强, 赵洋, 纪志强, 等. 基于多特征量的雷达辐射源脉间PRI调制识别[J]. 雷达科学与技术, 2019, 17(5):519-525.
	HAN G Q, ZHAO Y, JI Z Q, et al. Multi-feature volume-based identification of intervein PRI modulation of radar radiation sources[J]. Radar Science and Technology, 2019, 17(5):519-525. (in Chinese)
[5]	张景云, 苏世游, 何新乐, 等. 一种基于全卷积神经网络的雷达PRI调制样式识别方法[J]. 电子信息对抗技术, 2020, 35(3):40-45.
	ZHANG J Y, SU S Y, HE X L, et al. A full convolutional neural network-based radar PRI modulation style recognition method[J]. Electronic Information Countermeasures Technology, 2020, 35(3):40-45. (in Chinese)
[6]	侯欢. 分布式雷达信号识别与识别[D]. 西安: 西安电子科技大学, 2021.
	HOU H. Distributed radar signal recognition and identification[D]. Xi'an: Xidian University, 2021. (in Chinese)
[7]	WANG B, JIANG Z Y, ZHOU H J, et al. Blind recognition method for non-cooperative communication Signals based on aerial computing[J]. IEEE Transactions on Network Science and Engineering, 2024, 11(6):5244-5252.
[8]	CHEN T, TIAN H, LIU Y M, et al. Radar signal intra-pulse modulation recognition based on point cloud network[J]. IEEE Signal Processing Letters, 2025, 32:596-600.
[9]	ZENG Y, ZHANG M, HAN F, et al. Spectrum analysis and convolutional neural network for automatic modulation recognition[J]. IEEE Wireless Communications Letters, 2019, 8(3):929-932.
[10]	AN Z L, ZHANG T Q, SHEN M, et al. Series-constellation feature based blind modulation recognition for beyond 5G MIMO-OFDM systems with channel fading[J]. IEEE Transactions on Cognitive Communications and Networking, 2022, 8(2):793-811.
[11]	LIU M Q, LIU Z L, LU W D, et al. Distributed few-shot learning for intelligent recognition of communication jamming[J]. IEEE Journal of Selected Topics in Signal Processing, 2022, 16(3):395-405.
[12]	LUO Z Q, XIAO W S, ZHANG X Q, et al. RLITNN:a multi-channel modulation recognition model combining multi-modal features[J]. IEEE Transactions on Wireless Communications, 2024, 23(12):19083-19097.
[13]	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J]. Advances in Neural Information Processing Systems (NeurIPS), 2017,6000-6010.
[14]	LECUN Y, BOTTOW L. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11):2278-2324.
[15]	刘明骞, 高晓腾, 李明, 等. 空地协同场景下通信干扰智能识别方法[J]. 电子与信息学报, 2022, 44(3):825-834.
	LIU M Q, GAO X T, LI M, et al. Intelligent identification method of communication interference in air-ground cooperative scenario[J]. Journal of Electronics & Information Technology. 2022, 44(3):825-834. (in Chinese)
[16]	ALEXEY D, LUCAS B, KOLESNIKOV A, et al. An image is worth 16x16 words:Transformers for image recognition at scale[C]//Proceedings of the 9th International Conference on Learning Representations.[S.I.]:[s.n.],2021.
[17]	胡闯. 基于深度神经网络的通信干扰信号分类识别方法研究[D]. 长春: 吉林大学, 2021.
	HU C. Research on communication interference signal classification and identification method based on deep neural network[D]. Changchun: Jilin University, 2021. (in Chinese)
[18]	杨磊, 雷梦尧. 基于时频分析的雷达非匀速运动目标参数估计[J]. 中国民航大学学报, 2023, 41(4):29-36.
	YANG L, LEI M Y. Radar non-uniform motion target parameter estimation based on time-frequency analysis[J]. Journal of Civil Aviation University of China, 2023, 41(4):29-36. (in Chinese)
[19]	李瑞瑞. 相控阵雷达有源干扰智能感知与抗干扰决策方法研究[D]. 成都: 电子科技大学, 2021.
	LI R R. Research on intelligent sensing and anti-jamming decision-making method for phased array radar active jamming[D]. Chengdu: University of Electronic Science and Technology, 2021. (in Chinese)
[20]	赵婷, 张成祥. 一种基于SPWVD-WVD的高质量时频分析方法及ISAR成像应用[J]. 电讯技术, 2023, 63(10):1538-1545.
	ZHAO T, ZHANG C X. A high quality time-frequency analysis method based on SPWVD-WVD and ISAR imaging applications[J]. Telecommunication Engineering, 2023, 63(10):1538-1545. (in Chinese)
[21]	李玉洁, 马子航, 王艺甫, 等. 视觉Transformer(ViT)发展综述[J]. 计算机科学, 2025, 52(1):194-209.
	LI Y J, MA Z H, WANG Y F, et al. Visual Transformer (ViT) development overview[J]. Computer Science, 2025, 52(1):194-209. (in Chinese)
[22]	李翔, 张涛, 张哲, 等. Transformer在计算机视觉领域的研究综述[J]. 计算机工程与应用, 2023, 59(1):1-14. doi: 10.3778/j.issn.1002-8331.2204-0207
	LI X, ZHANG T, ZHANG Z, et al. A review of transformer research in computer vision[J]. Computer Engineering and Applications, 2023, 59(1):1-14. (in Chinese)
[23]	杨思栋. 基于时序图像化与ViT的时间序列分类研究[D]. 徐州: 中国矿业大学, 2023.
	YANG S D. A study of time series classification based on time series imaging with ViT[D]. Xuzhou: China University of Mining and Technology, 2023. (in Chinese)
[24]	冯辅周, 姬龙鑫, 朱俊臻, 等. 基于卷积神经网络的金属结构缺陷红外图像识别[J]. 四川兵工学报, 2021, 42(8):281-286.
	FENG F Z, JI L X, ZHU J Z, et al. Infrared image recognition of metal structure defects based on convolutional neural network[J]. Journal of Ordnance Equipment Engineering. 2021, 42(8):281-286. (in Chinese)
[25]	张姬, 侯进, 陈观业. 基于联合卷积与记忆神经网络的信号自动调制识别[J]. 计算机应用与软件, 2022, 39(8):226-233.
	ZHANG J, HOU J, CHEN G Y. Automatic signal modulation recognition based on joint convolutional and memory neural networks[J]. Computer Applications and Software, 2022, 39(8):226-233. (in Chinese)
[26]	GE R, LI Y, ZHU Y, et al. Method for recognition of communication interference signals under Small-Sample conditions[J]. Applied Sciences, 2024, 14(13):5869.