基于信号递归图和卷积宽度学习的小样本辐射源个体识别方法

doi:10.12305/j.issn.1001-506X.2025.09.30

摘要/Abstract

摘要：

针对当前辐射源个体识别方法在小样本条件下容易过学习、识别准确率低的问题，提出一种基于信号递归图和卷积宽度学习的小样本辐射源个体识别方法。该方法将辐射源信号转换为递归图作为宽度学习网络的输入，将辐射源数据时序特征转化为图像空间特征；此外，提出一种卷积宽度学习网络，将宽度学习中特征节点的计算方式由矩阵乘法替换为卷积运算，通过稀疏连接和权值共享减少模型参数数量，从而减轻模型过拟合风险。通过对公开数据集实验，验证了所提算法在少量训练样本数量条件下相较于其他算法有更好的识别性能。

关键词: 递归图, 卷积宽度学习, 小样本, 辐射源个体识别

Abstract:

In order to solve the problem that the current specific emitter identification methods are easy to overfit and have low recognition accuracy under the condition of small samples, a few-shot specific emitter identification method based on signal recurrence plot and convolutional broad learning is proposed. In this method, the emitter signal is converted into a recurrence plot as the input of the broad learning network, and the time series features of the emitter data are transformed into image spatial features. In addition, a convolutional broad learning network is proposed, which replaces the computation method of feature nodes in broad learning from matrix multiplication to convolution operation, and reduces the number of model parameters through sparse joining and weight sharing, thereby reducing the risk of model overfitting. Through experiments on public datasets, it is verified that the proposed algorithm has better recognition performance than other algorithms under the condition of a small number of training samples.

Key words: recurrence plot, convolutional broad learning, few-shot, specific emitter identification

中图分类号:

陈宇鹏, 黄科举, 刘辉, 邝龙坤, 杨俊安. 基于信号递归图和卷积宽度学习的小样本辐射源个体识别方法[J]. 系统工程与电子技术, 2025, 47(9): 3086-3092.

Yupeng CHEN, Keju HUANG, Hui LIU, Longkun KUANG, Junan YANG. Few-shot specific emitter identification method based on signal recurrence plot and convolutional broad learning[J]. Systems Engineering and Electronics, 2025, 47(9): 3086-3092.

图/表 9

图1

图2

图3

表1

图4

表2

图5

表3

图6

参考文献 30

1	胡苏, 马上, 林迪, 等. 基于射频指纹技术的卫星信号个体识别方法[J]. 信息对抗技术, 2022, 1 (1): 55- 61. doi: 10.12399/j.issn.2097-163x.2022.01.005
	HU S, MA S, LIN D, et al. Identification of individual satellite signal based on radio frequency fingerprint technology[J]. Information Countermeasure Technology, 2022, 1 (1): 55- 61. doi: 10.12399/j.issn.2097-163x.2022.01.005
2	ZENG M Y, LIU Z G, WANG Z W. An adaptive specific emitter identification system for dynamic noise domain[J]. IEEE Internet of Things Journal, 2022, 9 (24): 25117- 25135. doi: 10.1109/JIOT.2022.3195450
3	王春升, 王永民, 许华, 等. 基于残差原型网络的辐射源个体识别[J]. 系统工程与电子技术, 2023, 45 (7): 2249- 2258.
	WANG C S, WANG Y M, XU H, et al. Specific emitter identification based on residual prototype network[J]. Systems Engineering and Electronics, 2023, 45 (7): 2249- 2258.
4	MERCHANT K, REVAY S, STANTCHEV G, et al. Deep learning for RF device fingerprinting in cognitive communication networks[J]. IEEE Journal on Selected Topics in Signal Processing, 2018, 12 (1): 160- 167. doi: 10.1109/JSTSP.2018.2796446
5	SABINE A, ALEXANDER C, GERD A. Identification of radar emitter type with recurrent neural networks[C]// Proc. of the Sensor Signal Processing for Defence Conference, 2020.
6	ZHANG Y Z, ZHOU Z N, LI X P. Generative adversarial networks with gramian angular field for handling imbalanced data in specific emitter identification[J]. Signal, Image and Video Processing, 2024, 18 (3): 2929- 2938. doi: 10.1007/s11760-023-02960-x
7	YAN G L, CAI Z X, LIU Y C. Intelligent specific emitter identification using complex-valued convolutional neural network[C]// Proc. of the IEEE 23rd International Conference on Communication Technology, 2023: 1259−1263.
8	曲凌志, 杨俊安, 刘辉, 等. 嵌入注意力机制的通信辐射源个体识别方法[J]. 系统工程与电子技术, 2022, 44 (1): 20- 27. doi: 10.12305/j.issn.1001-506X.2022.01.03
	QU L Z, YANG J A, LIU H, et al. Method for individual identific-ation of communication radiation source embedded in attention mechanism[J]. Systems Engineering and Electronics, 2022, 44 (1): 20- 27. doi: 10.12305/j.issn.1001-506X.2022.01.03
9	LIU Y H, XU H, QI Z S. Specific emitter identification against unreliable features interference based on time-series classification network structure[J]. IEEE Access, 2020, 8, 200194- 200208. doi: 10.1109/ACCESS.2020.3035813
10	GUO L T, LIU C B, LIU Y C. Towards open-set specific emitter identific-ation using auxiliary classifier generative adversarial network and OpenMax[J]. IEEE Trans. on Cognitive Communications and Networking, 2024, 10(6): 2019−2028.
11	桂冠, 陶梦圆, 王诚, 等. 面向特定辐射源识别的小样本学习方法综述[J]. 南通大学学报（自然科学版）, 2023, 22 (3): 1- 16. doi: 10.12194/j.ntu.20220928001
	GUI G, TAO M Y, WANG C, et al. Survey of few-shot learning methods for specific emitter identification[J]. Journal of Nantong University（Natural Science Edition）, 2023, 22 (3): 1- 16. doi: 10.12194/j.ntu.20220928001
12	ZHANG W J, SHI F, ZHANG Q Y, et al. Few-shot specific emitter identification leveraging neural architecture search and advanced deep transfer learning[J]. IEEE Internet of Things Journal, 2024, 11 (8): 30087- 30093.
13	LIU M Q, CHAI Y E, LI M. Transfer learning-based specific emitter identification for ADS-B over satellite system[J]. Remote Sensing, 2024, 16 (12): 2068. doi: 10.3390/rs16122068
14	YANG N, ZHANG B N, DING G R, et al. Specific emitter identification wi-th limited samples: a model-agnostic meta-learning approach[J]. IEEE Communications Letters, 2022, 26 (2): 345- 349. doi: 10.1109/LCOMM.2021.3110775
15	HE B X, WANG F G. Specific emitter identification via sparse Bayesian learning versus model-agnostic meta-learning[J]. IEEE Trans. on Information Forensics and Security, 2023, 18, 3677- 3691. doi: 10.1109/TIFS.2023.3287073
16	HAN G J, WANG W T, XU Z W, et al. Data augmentation and individual identification method for emitters using contour stella image mapping[J]. IEEE Trans. on Cognitive Communications and Networking, 2024, 10 (4): 1253- 1262.
17	ZHANG X X, WANG Y, ZHANG Y B. Data augmentation aided few-shot learning for specific emitter identification[C]//Proc. of the IEEE 96th Vehicular Technology Conference, 2022.
18	WANG C, FU X, WANG Y, et al. Interpolative metric learning for few-shot specific emitter identification[J]. IEEE Trans. on Vehicular Technology, 2023, 72 (12): 16851- 16855. doi: 10.1109/TVT.2023.3326765
19	WANG Y, GUI G, LIN Y. Few-shot specific emitter identification via deep metric ensemble learning[J]. IEEE Internet of Things Journal, 2022, 9 (24): 24980- 24994. doi: 10.1109/JIOT.2022.3194967
20	WU Z T, WANG F G, HE B X. Specific emitter identification via contrastive learning[J]. IEEE Communications Letters, 2023, 27 (4): 1160- 1164. doi: 10.1109/LCOMM.2023.3264431
21	LIU B, YU H Y, DU J P. Specific emitter identification based on self-supervised contrast learning[J]. Electronics, 2022, 11 (18): 2907.
22	CHEN C L, LIU Z L. Broad learning system: an effective and efficient incremental learning system without the need for deep architecture[J]. IEEE Trans. on Neural Networks and Learning Systems, 2018, 29 (1): 10- 24. doi: 10.1109/TNNLS.2017.2716952
23	CHEN C L, LIU Z L. Broad learning system: a new learning paradigm and system without going deep[C]// Proc. of the 32nd Youth Academic Annual Conference of Chinese Association of Automation, 2017: 1271−1276.
24	ZHANG Y, GONG B, WANG Q. BLS-identification: a device fingerprint classification mechanism based on broad learning for internet of things[J]. Digital Communications and Networks, 2024, 10 (3): 728- 739.
25	XU Z W, HAN G J, LIU L, et al. A lightweight specific emitter identification model for IIoT devices based on adaptive broad learning[J]. IEEE Trans. on Industrial Informatics, 2023, 19 (5): 7066- 7075. doi: 10.1109/TII.2022.3206309
26	ZHANG Y B, PENG Y, SUN J L, et al. GPU-Free specific emitter identification using signal feature embedded broad learning[J]. IEEE Internet of Things Journal, 2023, 10 (14): 13028- 13039. doi: 10.1109/JIOT.2023.3257479
27	ZHOU Z Y, WEN J Q, LIU M X, et al. Classifying aquatic organism via evolving residual convolutional neural networks by optimized random vector fu-nctional link[J]. Multimedia Tools and Applications, 2024, 84, 221- 243.
28	陈梅辉. 基于递归图的多元生理时间序列复杂网络研究[D]. 徐州: 中国矿业大学, 2023.
	CHEN M H. Study on complex network of multivariate physiological time series based on recurrence plots[D]. Xuzhou: China University of Mining and Technology, 2023.
29	SAMER H, SAMURDHI K, DANIJELA C. WiSig: a large-Scale WiFi signal dataset for receiver and channel agnostic RF fingerprinting[J]. IEEE Access, 2022, 10, 22808- 22818. doi: 10.1109/ACCESS.2022.3154790
30	姚步泉. 智能通信辐射源个体识别研究[D]. 长沙: 国防科技大学, 2020.
	YAO B Q. Specific emitter identification via artificial intelligence[D]. Changsha: National University of Defense Technology, 2020.

参数	参数值
特征节点窗口个数$o$	8
特征节点个数$n$	8
增强节点个数$m$	3000
正则化系数$\lambda $	10⁻⁶

方法	优化方法	学习率	丢弃率	训练轮数	训练批大小	测试批大小
FCN	Adam	0.0001	0.1	50	100	250
RESNET	Adam	0.0001	0.1	50	100	250
STC-CVCNN	Adam	0.001	0.5	200	64	64

参数	BLN	RP-BLN	FCN	RESNET	SFEBLN	CBLN	RP-CBLN	STC-CVCNN
训练时间	0.59	24.13	2.92	9.61	0.66	0.82	67.10	568.35
识别时间	0.05	0.45	0.33	0.38	0.06	0.18	66.74	1.78

[1]	蔡伟, 王鑫, 蒋昕昊, 杨志勇, 陈栋. 基于解耦的小样本目标检测方法研究[J]. 系统工程与电子技术, 2024, 46(9): 2941-2950.
[2]	王春升, 王永民, 许华, 朱华丽. 基于残差原型网络的辐射源个体识别[J]. 系统工程与电子技术, 2023, 45(7): 2249-2258.
[3]	王东, 崔忠马, 陈文东, 舒勤. 基于模糊递归和最优硬阈值的局部投影降噪算法[J]. 系统工程与电子技术, 2023, 45(3): 621-628.
[4]	孙晶明, 虞盛康, 孙俊. 基于元学习的雷达小样本目标识别方法及改进[J]. 系统工程与电子技术, 2022, 44(6): 1839-1845.
[5]	秦博伟, 蒋磊, 许华, 齐子森. 基于残差生成对抗网络的调制识别算法[J]. 系统工程与电子技术, 2022, 44(6): 2019-2026.
[6]	吕勤哲, 全英汇, 沙明辉, 董淑仙, 邢孟道. 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.
[7]	刘旗, 张新禹, 刘永祥. 基于门控多尺度匹配网络的小样本SAR目标识别[J]. 系统工程与电子技术, 2022, 44(11): 3346-3356.
[8]	马玉凤, 向南, 豆亚杰, 姜江, 杨克巍, 谭跃进. 军事系统工程中的知识图谱应用及研究[J]. 系统工程与电子技术, 2022, 44(1): 146-153.
[9]	史蕴豪, 许华, 郑万泽, 刘英辉. 基于集成学习与特征降维的小样本调制识别方法[J]. 系统工程与电子技术, 2021, 43(4): 1099-1109.
[10]	陈浩, 杨俊安, 刘辉. 基于深度残差适配网络的通信辐射源个体识别[J]. 系统工程与电子技术, 2021, 43(3): 603-609.
[11]	晏媛, 孙俊, 孙晶明, 于俊朋. 雷达小样本目标识别方法及应用分析[J]. 系统工程与电子技术, 2021, 43(3): 684-692.
[12]	曾强, 黄政, 魏曙寰. 融合专家先验知识和单调性约束的贝叶斯网络参数学习方法[J]. 系统工程与电子技术, 2020, 42(3): 646-652.
[13]	吴子龙, 陈红, 雷迎科, 李昕, 熊颢. 基于堆栈式LSTM网络的通信辐射源个体识别[J]. 系统工程与电子技术, 2020, 42(12): 2915-2923.
[14]	潘崇煜, 黄健, 郝建国, 龚建兴, 张中杰. 融合零样本学习和小样本学习的弱监督学习方法综述[J]. 系统工程与电子技术, 2020, 42(10): 2246-2256.
[15]	方章闻, 张金艺, 李科, 姜玉稀. 小样本条件下的通信辐射源半监督特征提取[J]. 系统工程与电子技术, 2020, 42(10): 2381-2389.