基于Deep SVDD的通信信号异常检测方法

doi:10.12305/j.issn.1001-506X.2022.07.29

系统工程与电子技术 ›› 2022, Vol. 44 ›› Issue (7): 2319-2328.doi: 10.12305/j.issn.1001-506X.2022.07.29

基于Deep SVDD的通信信号异常检测方法

康颖^1,², 赵治华^1,², 吴灏^1,^2,*, 李亚星^1,², 孟进^1,²

1. 海军工程大学军用电气科学与技术研究所, 湖北武汉 430033
2. 海军工程大学舰船综合电力技术国防科技重点实验室, 湖北武汉 430033

收稿日期:2021-05-21 出版日期:2022-06-22 发布日期:2022-06-28
通讯作者: 吴灏
作者简介:康颖 (1994—), 男, 博士研究生, 主要研究方向为通信信号处理|赵治华 (1962—), 男, 教授, 博士, 主要研究方向为电磁兼容、电磁领域的分析与计算、自适应干扰对消|吴灏 (1988—), 男, 讲师, 博士, 主要研究方向为通信信号处理、波形设计|李亚星 (1988—), 男, 讲师, 博士, 主要研究方向为通信信号处理|孟进 (1977—), 男, 研究员, 博士, 主要研究方向为电磁干扰与防护
基金资助:
国家杰出青年科学基金(52025072);国家自然科学基金(61801502);国家自然科学基金(61801501);国家自然科学基金(62001497);国防科技重点实验室稳定支持项目(6142217200105);湖北省自然科学基金(ZRMS202001331);海军工程大学发展基金(425317S126)

Deep SVDD-based anomaly detection method for communication signals

Ying KANG^1,², Zhihua ZHAO^1,², Hao WU^1,^2,*, Yaxing LI^1,², Jin MENG^1,²

1. Institute of Military Electrical Science and Technology, Naval University of Engineering, Wuhan 430033, China
2. National Key Laboratory of Science and Technology on Vessel Integrated Power System, Naval University of Engineering, Wuhan 430033, China

Received:2021-05-21 Online:2022-06-22 Published:2022-06-28
Contact: Hao WU

摘要/Abstract

摘要：

针对复杂电子对抗场景中的非理想信道环境, 该文提出了一种基于深度学习的异常检测(anomaly detection, AD)方法。首先, 分析了利用时频同相/正交(in-phase/quadrature, I/Q)采样数据进行AD的可行性; 然后, 设计了深度学习网络架构, 并提出基于深度支持向量描述(deep support vector data description, Deep SVDD)和调制识别的AD方法。仿真及实验结果表明: 相比于经典的单分类检测算法, 该方法检测性能和实时性明显提升, 且在非理想信道环境下表现鲁棒。该方法已在某型号项目原理样机上得到验证, 具有很高应用价值。

关键词: 异常检测, Deep SVDD, 调制识别, 干扰预警

Abstract:

To solve the problem of anomaly detection (AD) of the non-ideal channel in complex electronic countermeasures, a deep learning based method is presented. First, the feasibility of using time-frequency in-phase/quadrature (I/Q) sampling data for anomaly detection (AD) is analyzed. Then, a deep learning network architecture is designed and an AD method based on deep support vector data description (Deep SVDD) and modulation classification is proposed. Simulation and experimental results show that the detection performance and real-time performance of the method are significantly improved compared with the classical algorithms of one-class classification, and the performance is robust in non-ideal channel environment. The method is validated on a sample machine and is of high application value.

Key words: anomaly detection (AD), deep support vector data description (Deep SVDD), modulation classification, interference warning

中图分类号:

TN911.7

康颖, 赵治华, 吴灏, 李亚星, 孟进. 基于Deep SVDD的通信信号异常检测方法[J]. 系统工程与电子技术, 2022, 44(7): 2319-2328.

Ying KANG, Zhihua ZHAO, Hao WU, Yaxing LI, Jin MENG. Deep SVDD-based anomaly detection method for communication signals[J]. Systems Engineering and Electronics, 2022, 44(7): 2319-2328.

图/表 24

图1

图2

图3

图4

图5

图6

图7

表1

图8

图9

表2

表3

图10

表4

图11

图12

表5

图13

表6

图14

图15

图16

图17

图18

参考文献 34

1	吴灏, 周亮, 李亚星, 等. 基于卷积神经网络和稀疏滤波的调制识别方法[J]. 系统工程与电子技术, 2019, 41 (9): 2114- 2121.
	WU H , ZHOU L , LI Y S . Modulation classification based on convolutional neural network and sparse filtering[J]. Systems Engineering and Electronics, 2019, 41 (9): 2114- 2121.
2	刘义, 赵晶, 刘佳楠, 等. 基于作战效能的战场电磁环境分级描述方法[J]. 系统工程与电子技术, 2011, 33 (5): 1059- 1062. doi: 10.3969/j.issn.1001-506X.2011.05.20
	LIU Y , ZHAO J , LIU J N , et al. Novel battlefield electromagnetic environment classification method based on combat efficacy of weapon systems[J]. Systems Engineering and Electronics, 2011, 33 (5): 1059- 1062. doi: 10.3969/j.issn.1001-506X.2011.05.20
3	CHANDOLA V , BANERJEE A , KUMAR V , et al. Anomaly detection: a survey[J]. ACM Computing Surveys, 2009, 41 (3): 1- 58.
4	WANG B , LIU K R . Advances in cognitive radio networks: a survey[J]. IEEE Journal of Selected Topics in Signal Processing, 2010, 5 (1): 5- 23.
5	TANDRA R , SAHAI A . SNR walls for signal detection[J]. IEEE Journal of Selected Topics in Signal Processing, 2008, 2 (1): 4- 17. doi: 10.1109/JSTSP.2007.914879
6	TURIN G L . An introduction to matched filters[J]. IRE Trans.on Information Theory, 1960, 6 (3): 311- 329. doi: 10.1109/TIT.1960.1057571
7	YE Z, GROSSPIETSCH J, MEMIK G. Spectrum sensing using cyclostationary spectrum density for cognitive radios[C]//Proc. of the IEEE Workshop on Signal Processing Systems, 2007.
8	LECUN Y , BENGIO Y , HINTON G . Deep learning[J]. Nature, 2015, 521 (7553): 436- 444. doi: 10.1038/nature14539
9	HINTON G E , OSINDERO S , TEH Y W . A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, (521): 436- 444.
10	MINAEE S , KALCHBRENNER N , CAMBRIA E , et al. Deep learning-based text classification: a comprehensive review[J]. ACM Computing Surveys, 2021, 54 (3): 1- 40.
11	MINAEE S, BOYKOV Y Y, PORIKLI F, et al. Image segmentation using deep learning: a survey[EB/OL]. [2022-05-25]. https://arxiv.org/abs/2001.05566.
12	O'SHEA T J, CLANCY T C, MCGWIER R W. Recurrent neural radio anomaly detection[EB/OL]. [2022-05-25]. https://arxiv.org/abs/1611.00301.
13	PELLACO L, SINGH N, JALDEN J. Spectrum prediction and interference detection for satellite communications[C]//Proc. of the 37th International Communications Satellite Systems Conference, 2019.
14	CONN M A, JOSYULA D. Radio frequency classification and anomaly detection using convolutional neural networks[C]//Proc. of the IEEE Radar Conference, 2019.
15	LI Z J, XIAO Z J, WANG B L, et al. Scaling deep learning models for spectrum anomaly detection[C]//Proc. of the 20th ACM International Symposium on Mobile Ad Hoc Networking and Computing, 2019: 291-300.
16	ZHOU X H , XIONG J , ZHANG X C , et al. A radio anomaly detection algorithm based on modified generative adversarial network[J]. IEEE Wireless Communications Letters, 2021, 10 (7): 1552- 1556. doi: 10.1109/LWC.2021.3074135
17	ZHANG L, HUANG C X, TANG H, et al. DTV radio spectrum anomaly detection based on an improved GAN[C]//Proc. of the 33rd General Assembly and Scientific Symposium of the International Union of Radio Science, 2020.
18	YOKKAMPON U, CHUMKAMON S, MOWSHOWITZ A, et al. Anomaly detection using variational autoencoder with spectrum analysis for time series data[C]//Proc. of the 9th International Conference on Informatics, Electronics & Vision, 2020.
19	TOMA A, KRAYANI A, MARCENARO L, et al. Deep learning for spectrum anomaly detection in cognitive mmWave radios[C]//Proc. of the IEEE 31st Annual International Symposium on Personal, Indoor and Mobile Radio Communications, 2020.
20	GHANNEY Y, AJIB W. Radio frequency interference detection using deep learning[C]//Proc. of the IEEE 91st Vehicular Technology Conference, 2020.
21	XU J, TIAN Y, YUAN S, et al. Noise attention based spectrum anomaly detection method for unauthorized bands[EB/OL]. [2022-05-25]. https://arxiv.org/abs/2104.08517.
22	RAJENDRAN S , MEERT W , LENDERS V , et al. Unsupervised wireless spectrum anomaly detection with interpretable features[J]. IEEE Trans.on Cognitive Communications Networking, 2019, 5 (3): 637- 647. doi: 10.1109/TCCN.2019.2911524
23	RAJENDRAN S , LENDERS V , MEERT W , et al. Crowdsourced wireless spectrum anomaly detection[J]. IEEE Trans.on Cognitive Communications and Networking, 2019, 6 (2): 694- 703.
24	TOMA A , KRAYANI A , FARRUKH M , et al. AI-based abnormality detection at the PHY-layer of cognitive radio by learning generative models[J]. IEEE Trans.on Cognitive Communications and Networking, 2020, 6 (1): 21- 34. doi: 10.1109/TCCN.2020.2970693
25	SCHEIRER W J , ROCHA A D R , SAPKOTA A , et al. Toward open set recognition[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence, 2013, 35 (7): 1757- 1772. doi: 10.1109/TPAMI.2012.256
26	WANG Y , LIU M , YANG J , et al. Data-driven deep learning for automatic modulation recognition in cognitive radios[J]. IEEE Trans.on Vehicular Technology, 2019, 68 (4): 4074- 4077. doi: 10.1109/TVT.2019.2900460
27	RUFF L, VANDERMEULEN R, GOERNITZ N, et al. Deep one-class classification[C]//Proc. of the International Conference on Machine Learning, 2018: 4393-4402.
28	周鑫, 何晓新, 郑昌文. 基于图像深度学习的无线电信号识别[J]. 通信学报, 2019, 40 (7): 114- 125.
	ZHOU X , HE X X , ZHENG C W . Radio signal recognition based on image deep learning[J]. Journal on Communications, 2019, 40 (7): 114- 125.
29	O'SHEA T J, CORGAN J, CLANCY T C. Convolutional radio modulation recognition networks[C]//Proc. of the International Conference on Engineering Applications of Neural Networks, 2016: 213-226.
30	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
31	Analog Devices. AD9361[EB/OL]. [2022-05-25]. www.analog.com/en/products/ad9361.html.
32	KINGMA D P, BA J. ADAM: a method for stochastic optimization[EB/OL]. [2022-05-25]. https://arxiv.org/abs/1412.6980.
33	CHALAPATHY R, MENON A K, CHAWLA S. Anomaly detection using one-class neural networks[EB/OL]. [2022-05-25]. https://arxiv.org/abs/1802.06360.
34	LIU F T, TING K M, ZHOU Z H. Isolation forest[C]//Proc. of the IEEE 8th International Conference on Data Mining, 2008: 413-422.

参数	数值
频率偏移/kHz	-20~20
相位噪声功率/ (dBc/Hz)	-84(频偏为1 kHz) -100(频偏为10 kHz) -105(频偏为100 kHz) -109(频偏为1 MHz)
白噪声/dB	0~30
降采样倍数	70
低通滤波截至频率/kHz	17.5
低通滤波过渡带宽/kHz	150

编号	输入	一维卷积核长度N_C	卷积核个数	残差单元个数	输出
1	4×1 024	1×35	16	1	1×256
2	4×1 024	1×35	16	2	1×256
3	4×1 024	1×35	32	1	1×256
4	4×1 024	1×35	32	2	1×256

编号	AUC	需优化参数(k个)
1	0.978 5	16 795 152
2	0.952 7	16 813 328
3	0.986 3	33 626 144
4	0.997 9	33 698 848

方法	数据样式	输入维度	网络深度(层)
AE-OCSVM	时频谱图	32×32×3	7
AE-IFOREST	时频谱图	32×32×3	7
本文方法	基带IQ	4×1 024	7

方法	预处理	网络推断时间	线型判决时间	总耗时
AE-OCSVM	118	3.6	0.3	121.9
AE-IFOREST	118	3.6	37	155.6
本文方法	0.3	19.2	—	19.5

基于Deep SVDD的通信信号异常检测方法

Deep SVDD-based anomaly detection method for communication signals

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 24

参考文献 34

相关文章 15

编辑推荐

Metrics

本文评价

参数	数值
信道类型	瑞利衰落, 莱斯衰落
频率偏移/kHz	-1~1
发射功率/W	1
ISR/dB	-20~20

[1]	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.
[2]	邵凯, 朱苗苗, 王光宇. 基于生成对抗与卷积神经网络的调制识别方法[J]. 系统工程与电子技术, 2022, 44(3): 1036-1043.
[3]	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.
[4]	刘凯, 张斌, 黄青华. 基于TCNN-BiLSTM网络的调制识别算法[J]. 系统工程与电子技术, 2020, 42(8): 1841-1849.
[5]	刘凯, 赵梦伟, 黄青华. 基于近似熵的Multi-h CPM调制识别算法[J]. 系统工程与电子技术, 2020, 42(3): 698-703.
[6]	李晨, 杨俊安, 刘辉. 基于信息熵和GA-ELM的调制识别算法[J]. 系统工程与电子技术, 2020, 42(1): 223-229.
[7]	吴灏, 周亮, 李亚星, 郭宇, 孟进. 基于卷积神经网络和稀疏滤波的调制识别方法[J]. 系统工程与电子技术, 2019, 41(9): 2114-2121.
[8]	张一迪, 王培志, 陆起涌, 张建秋. 异常数据恒虚警检测的非参数方法[J]. 系统工程与电子技术, 2019, 41(5): 964-971.
[9]	郝云飞, 刘章孟, 郭福成, 张敏. 基于生成对抗网络的信号调制方式的开集识别[J]. 系统工程与电子技术, 2019, 41(11): 2619-2624.
[10]	郭鹏程, 刘峥, 罗丁利. 基于两次异常检测的雷达HRRP杂波稳健目标识别方法[J]. 系统工程与电子技术, 2019, 41(10): 2221-2226.
[11]	谭晓衡, 褚国星, 张雪静, 杨扬. 基于高阶累积量和小波变换的调制识别算法[J]. 系统工程与电子技术, 2018, 40(1): 171-177.
[12]	付俊强, 李蓉, 赵成林, 李斌. 基于贝叶斯序贯推理的自适应调制识别算法[J]. 系统工程与电子技术, 2015, 37(12): 2860-2864.
[13]	靳晓艳，周希元. 一种最大似然调制识别的快速算法[J]. Journal of Systems Engineering and Electronics, 2013, 35(3): 615-618.
[14]	杨发权，李赞，李红艳，郝本建，潘忠显. 基于蜂群算法和神经网络的通信调制识别方法[J]. 系统工程与电子技术, 2013, 35(10): 2186-2191.
[15]	李一兵, 葛娟, 林云. 基于熵特征和支持向量机的调制识别方法[J]. Journal of Systems Engineering and Electronics, 2012, 34(8): 1691-1695.