基于混合信号多域特征和Transformer的干扰识别

doi:10.12305/j.issn.1001-506X.2024.06.32

系统工程与电子技术 ›› 2024, Vol. 46 ›› Issue (6): 2138-2145.doi: 10.12305/j.issn.1001-506X.2024.06.32

• 通信与网络 • 上一篇

基于混合信号多域特征和Transformer的干扰识别

阳鹏飞¹, 何羚¹^,², 王茜¹^,²^,*, 王睿笛¹, 张明志¹

1. 电子科技大学航空航天学院, 四川成都 611731
2. 飞行器集群智能感知与协同控制四川省重点实验室, 四川成都 611731

收稿日期:2023-07-13 出版日期:2024-05-25 发布日期:2024-06-04
通讯作者: 王茜
作者简介:阳鹏飞(1999—), 男, 硕士研究生, 主要研究方向为复合调制信号识别、干扰信号识别、智能抗干扰决策
何羚(1972—), 女, 副教授, 硕士, 主要研究方向为测控通信链路安全防护、信号感知与处理
王茜(1984—), 女, 副教授, 博士, 主要研究方向为信号检测、时频分析
王睿笛(1998—), 男, 硕士研究生, 主要研究方向为多域自适应抗干扰算法
张明志(2000—), 男, 硕士研究生, 主要研究方向为二进制偏移载波调制信号仿真与实现
基金资助:
四川省自然科学基金(2022NSFSC0545);四川省自然科学基金(2023NSFSC0494)

Interference identification based on mixed signal multidomain feature and Transformer framework

Pengfei YANG¹, Ling HE¹^,², Qian WANG¹^,²^,*, Ruidi WANG¹, Mingzhi ZHANG¹

1. School of Aeronautics and Astronautics, University of Electronic Science and Technology of China, Chengdu 611731, China
2. Aircraft Swarm Intelligent Sensing and Cooperative Control Key Laboratory of Sichuan Province, Chengdu 611731, China

Received:2023-07-13 Online:2024-05-25 Published:2024-06-04
Contact: Qian WANG

摘要/Abstract

摘要：

针对无线通信信道易受到蓄意射频信号干扰问题，提出了一种从混合信号中识别干扰类型的方法。通过改进经典Transformer结构, 形成新型网络模型Multidomain-former, 以提取多域特征和识别信号干扰类型。首先, 通过特定的序列划分机制对输入频谱进行预处理, 并通过线性嵌入和位置编码保留原始顺序特征; 其次, 设计了逆傅氏变换和傅氏变换结合的编码模块, 使Multidomain-former能同时提取频域和时域特征。使用通用仪器和收发天线搭建了无线收发信道, 在不同干信比条件下对混合信号频谱进行采集, 得到训练集和测试集。干扰对比实验通过所提Multidomain-former网络模型完成, 并将经典的Transformer结构和其他常见的深度学习模型与所提网络模型进行了对比。对比实验结果表明, 在干信比小于10 dB时, 所提模型性能相较于经典Transformer在识别正确率方面有2%~3%的提升; 在干信比等于-5 dB时, 所提模型以最少参数量和次低计算复杂度获得了比另外5种基准网络高3.0%~9.3%的识别率。

关键词: 混合信号, 多域特征提取, 干扰识别, Transformer, 干信比

Abstract:

Aiming at the vulnerability of wireless communication channels to interference from intentional radio frequercy signals, a method of interference identification from mixed signals is presented. In this work, a novel Transformer model named Multidomain-former is devised to serve as the multidomain features extractor and jamming types recognizer. The proposed model is built with the following characteristics: the original spectral data is firstly preprocessed by a specific sequence partitioning mechanism. Meanwhile, the initial sequence features are reserved by linear embedding and position coding. Secondly, an encoding module which jointly adopting inverse Fourier transform and Fourier transform is designed, by this means Multidomain-former can obtain both frequency domain and time domain features. A real wireless transceiver channel is established utilizing universal instruments and transceiver antennas, and the mixed signal spectrum is collected subject to different jamming-to-signal ratios (JSR) to form training and test data sets. The interference classification experiments are carried out sequentially by the proposed Multidomain-former, in comparison with the classic Transformer and other popular deep learning networks as well. It is shown that Multidomain-former achieves the best performance with the least number of parameters and lower complexity. With the condition of the JSR is less than 10 dB, the probability for correct classification of Multidomain-former is 2%~3% higher than that of classic Transformer. When the JSR is equal to -5 dB, the performance of Multidomain-former is proofed to increase by 3.0%~9.3% on correct classification rate compared with other benchmarks.

Key words: mixed signal, multidomain feature extraction, interference identification, Transformer, jamming-to-signal ratio (JSR)

中图分类号:

TN973.1

阳鹏飞, 何羚, 王茜, 王睿笛, 张明志. 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.

Pengfei YANG, Ling HE, Qian WANG, Ruidi WANG, Mingzhi ZHANG. Interference identification based on mixed signal multidomain feature and Transformer framework[J]. Systems Engineering and Electronics, 2024, 46(6): 2138-2145.

图/表 10

图1

图2

图3

表1

表2

表3

图4

表4

图5

图6

参考文献 17

1	周红平, 马明辉, 吴若无, 等. 基于稀疏表示分类的雷达欺骗干扰识别方法[J]. 系统工程与电子技术, 2022, 44 (9): 2791- 2799.
	ZHOU H P , MA M H , WU R W , et al. Deception jamming recog-nition of radar based on sparse representation classification[J]. Systems Engineering and Electronics, 2022, 44 (9): 2791- 2799.
2	MUGHAL M O , KIM S . Signal classification and jamming detection in wide-band radios using Naive Bayes classifier[J]. IEEE Communications Letters, 2018, 22 (7): 1398- 1401. doi: 10.1109/LCOMM.2018.2830769
3	SHI Y X , LU X J , NIU Y T , et al. Efficient jamming identification in wireless communication: using small sample data driven Naive Bayes classifier[J]. IEEE Wireless Communications Letters, 2021, 10 (7): 1375- 1379. doi: 10.1109/LWC.2021.3064843
4	ZHANG L , REN J , LI T T . Time-varying jamming modeling and classification[J]. IEEE Trans.on Signal Processing, 2012, 60 (7): 3902- 3907. doi: 10.1109/TSP.2012.2193574
5	SHAO G Q , CHEN Y S , WEI Y S . Convolutional neural network-based radar jamming signal classification with sufficient and limited samples[J]. IEEE Access, 2020, 8, 80588- 80598. doi: 10.1109/ACCESS.2020.2990629
6	HAN H , LI W , FENA Z B , et al. Proceed from known to unknown: jamming pattern recognition under open-set setting[J]. IEEE Wireless Communications Letters, 2022, 11 (4): 693- 697. doi: 10.1109/LWC.2021.3140145
7	靳增源, 张晓瀛, 谭思源, 等. 基于集成时频通道注意力的倒残差神经网络干扰识别[J]. 信号处理, 2023, 39 (2): 343- 355.
	JIN Z Y , ZHANG X Y , TAN S Y , et al. Jamming identification based on inverse residual neural network with integrated time-frequency channel attention[J]. Journal of Signal Processing, 2023, 39 (2): 343- 355.
8	SHAO G G , CHEN Y S , WEI Y S . Deep fusion for radar jamming signal classification based on CNN[J]. IEEE Access, 2020, 8, 117236- 117244. doi: 10.1109/ACCESS.2020.3004188
9	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.
10	王李静. 航天测控链路干扰多维感知及决策研究[D]. 成都: 电子科技大学, 2020.
	WANG L J. Research on multidimensional perception and decision-marking of interference in space TT&C links[D]. Chengdu: University of Electronic Science and Technology of China, 2020.
11	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proc. of the Advances in Neural Information Processing Systems, 2017.
12	DOSOVITSHIY A, BEYER L, KOLESNIKOV A, et al. An image is worth 16×16 words: transformers for image recognition at scale[EB/OL]. [2023-05-10]. https://arxiv.org/abs/2010.11929.
13	IANDOLA F N, HAN S, MOSKEWICZ M W, et al. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and < 0.5 MB model size[EB/OL]. [2023-05-10]. https://arxiv.org/abs/1602.07360.
14	MA N, ZHANG X, ZHENG H T, et al. Shufflenet v2: practical guidelines for efficient CNN architecture design[C]//Proc. of the European conference on computer vision, 2018: 116-131.
15	HOWARD A, SANDLER M, CHU G, et al. Searching for MobileNetV3[C]//Proc. of the IEEE/CVF International Conference on Computer Vision, 2019: 1314-1324.
16	TAN M X, LE Q. EfficientNetV2: Smaller models and faster training[C]//Proc. of the International Conference on Machine Learning, 2021: 10096-10106.
17	HOCHREITER S , SCHMIDHUBER J . Long short-term memory[J]. Neural Computation, 1997, 9 (8): 1735- 1780. doi: 10.1162/neco.1997.9.8.1735

仪器设备功能	型号	仪器设备设置
SOI模拟产生	Agilent E4438C	CF=3 GHz, BW=20 MHz, 发射功率固定为P_s(单位: dB)
干扰信号模拟产生	Ceyear 1465D	CF=3 GHz, BW及信号按照典型干扰类型定义产生, 发射功率P_j(单位: dB)可调, 调节范围为JSR(dB)=P_j-P_s, JSR∈{-5, 0, 5, 10, 15, 20}
混合信号频谱分析	Keysight N9030B	CF=3 GHz, Span=50 MHz, RBW=20 kHz, VBW=20 kHz, 平均类型: Log-Power, FFT点数为4 096

信号	信号类型/体制	参数设置
SOI s(t)	直序扩频-跳频-BPSK	载频3 GHz, 跳频带宽100 MHz, gold码扩频码速10 Mcps, 跳频点数2 048, 跳频间隔40 kHz
干扰信号J(t)	MTJ	中心频率3 GHz, 音频数5, 频率间隔2 MHz
	PJ	中心频率3 GHz, 脉冲宽度200 ns, 脉冲周期1 μs
	LFMJ	扫描范围2.99~3.01 GHz, 单次扫描时间10 ms
	PBNJ	中心频率3 GHz, 带宽6 MHz
	PBWJ	中心频率3 GHz, 带宽12 MHz

子帧长度L	参数量(×10⁶)	FLOPS(×10⁶)	平均正确识别率/%
32	0.11	1.16	81.4
48	0.21	2.48	82.0
64	0.34	4.29	83.0
80	0.49	6.59	83.0
96	0.69	9.38	82.4

网络模型名称	参数量(×10⁶)	FLOPS(×10⁶)
SqueezeNet	0.73	6.43
ShuffleNetV2	1.26	11.77
MobileNetV3	1.52	5.48
EfficientNetV2	20.18	237.51
LSTM	1.23	2.29
Multidomain-former	0.34	4.29

[1]	姚震, 杨闯, 纪晓东. 低快拍下混合信号DOA快速估计算法[J]. 系统工程与电子技术, 2024, 46(2): 722-728.
[2]	刘子昌, 白永生, 李思雨, 贾希胜. 基于小波时频图与Swin Transformer的柴油机故障诊断方法[J]. 系统工程与电子技术, 2023, 45(9): 2986-2998.
[3]	邵正途, 许登荣, 徐文利, 王晗中. 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.
[4]	眭海刚, 李嘉杰, 苟国华. 基于异源影像匹配的无人机在线快速定位方法[J]. 系统工程与电子技术, 2023, 45(10): 3008-3015.
[5]	钱坤, 李晨瑄, 陈美杉, 郭继伟, 潘磊. 基于改进Swin Transformer的舰船目标实例分割算法[J]. 系统工程与电子技术, 2023, 45(10): 3049-3057.
[6]	周红平, 马明辉, 吴若无, 许雄, 郭忠义. 基于稀疏表示分类的雷达欺骗干扰识别方法[J]. 系统工程与电子技术, 2022, 44(9): 2791-2799.
[7]	王鹏博, 贺成艳, 杨倩倩, 佟文华. 窄带干扰信号对北斗三号B2信号质量的影响[J]. 系统工程与电子技术, 2022, 44(7): 2286-2292.
[8]	齐栋, 唐敏, 刘成城, 赵拥军. 高斯色噪声下混合信号二维DOA估计方法[J]. 系统工程与电子技术, 2019, 41(10): 2198-2204.
[9]	毛虎, 吴德伟, 卢虎. 对GPS军码信号的高效带限高斯噪声干扰分析[J]. 系统工程与电子技术, 2019, 41(1): 21-26.
[10]	杨立波, 高仕博, 胡瑞光, 韦海萍, 肖利平. 合成孔径雷达相干与非相干干扰性能分析[J]. 系统工程与电子技术, 2018, 40(11): 2444-.
[11]	孙殿星, 陈翔, 万建伟, 王国宏, 吴巍. 基于多特征的密集假目标干扰融合识别与抑制[J]. 系统工程与电子技术, 2018, 40(10): 2207-2215.
[12]	景小荣, 杨洋, 张祖凡, 陈前斌. 均匀线阵混合信源DOA估计与互耦误差自校正[J]. 系统工程与电子技术, 2014, 36(9): 1696-1701.
[13]	范广伟，蔚保国，晁磊，邓志鑫. 卫星导航电磁干扰识别分类器设计[J]. 系统工程与电子技术, 2014, 36(2): 234-238.
[14]	孙闽红, 唐斌. 距离-速度同步拖引欺骗干扰的频谱特征分析[J]. Journal of Systems Engineering and Electronics, 2009, 31(1): 83-85.
[15]	孙闽红, 唐斌. 距离-速度同步拖引欺骗干扰的频谱特征分析[J]. Journal of Systems Engineering and Electronics, 2009, 31(01): 83-85.

基于混合信号多域特征和Transformer的干扰识别

Interference identification based on mixed signal multidomain feature and Transformer framework

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 17

相关文章 15

编辑推荐

Metrics

本文评价