基于时频域特征融合的IR-UWB穿墙雷达人体行为识别方法

doi:10.12305/j.issn.1001-506X.2024.03.10

摘要/Abstract

摘要：

冲激脉冲(impulse radio, IR)超宽带(ultra-wideband, UWB)穿墙雷达因其良好的穿透性和距离分辨率在穿墙人体行为识别领域具有重要作用, 但是常规识别方法仅采用单域特征对行为模式进行描述, 识别准确率不高。针对这一问题, 提出基于时频域特征融合的IR-UWB穿墙雷达人体行为识别算法。首先, 通过杂波抑制及距离补偿方法获取高信噪比的人体行为距离像。其次, 基于距离像提取目标时域特征, 与频域特征进行融合, 构建数据集。最后, 基于支持向量机(support vector machine, SVM)算法对人体行为进行识别。实验结果表明, 所提算法对于IR-UWB穿墙雷达人体行为识别能够达到95%的准确率。

关键词: 人体行为识别, 冲激脉冲超宽带雷达, 特征提取, 支持向量机

Abstract:

The impulse radio (IR) ultra-wideband (UWB) through wall radar plays an important role in the field of through wall human behavior recognition due to its good penetration and range resolution. However, the conventional recognition method, only uses single domain feature to describe the behavior pattern, and the recognition accuracy is not high. Aiming at this problem, an IR-UWB through wall radar human behavior recognition algorithm is proposed based on time-frequency domain feature fusion. Firstly, the human behavior range image with high signal-to-noise ratio is obtained by clutter suppression and distance compensation methods. Secondly, the time domain features of the target are extracted based on the range image. It is fused with frequency domain features to build data set. Finally, human behavior is identified based on support vector machine (SVM) algorithm. Experimental results show that the proposed algorithm can achieve 95% accuracy for human behavior recognition with IR-UWB through wall radar.

Key words: human behavior recognition, impulse radio (IR) ultra-wideband (UWB) radar, feature extraction, support vector machine (SVM)

中图分类号:

TN957

杨德贵, 许道峰. 基于时频域特征融合的IR-UWB穿墙雷达人体行为识别方法[J]. 系统工程与电子技术, 2024, 46(3): 849-858.

Degui YANG, Daofeng XU. Human behavior recognition method of IR-UWB through wall radar based on time-frequency domain feature fusion[J]. Systems Engineering and Electronics, 2024, 46(3): 849-858.

图/表 16

图1

图2

图3

图4

图5

图6

图7

图8

图9

表1

图10

表2

表3

表4

图11

表5

参考文献 31

1	LIU J L , WANG L , FANG J , et al. Multi-target intense human motion analysis and detection using channel state information[J]. Sensors, 2018, 18 (10): 3379- 3386. doi: 10.3390/s18103379
2	AMIN M G . Radar for indoor monitoring: detection, classification, and assessment[M]. London: CRC Press, 2018.
3	王彩云, 姚晨, 吴钇达, 等. 基于改进Dijkstra算法与时频域滤波的雷达目标识别[J]. 系统工程与电子技术, 2022, 44 (10): 3090- 3095. doi: 10.12305/j.issn.1001-506X.2022.10.12
	WANG C Y , YAO C , WU Y D , et al. Radar target recognition based on improved Dijkstra algorithm with time frequency domain filtering[J]. Systems Engineering and Electronics, 2022, 44 (10): 3090- 3095. doi: 10.12305/j.issn.1001-506X.2022.10.12
4	王彩云, 黄盼盼, 李晓飞, 等. 基于AEPSO-SVM算法的雷达HRRP目标识别[J]. 系统工程与电子技术, 2019, 41 (9): 1984- 1989. doi: 10.3969/j.issn.1001-506X.2019.09.10
	WANG C Y , HUANG P P , LI X F , et al. Radar HRRP target recognition based on AEPSO-SVM algorithm[J]. Systems Engineering and Electronics, 2019, 41 (9): 1984- 1989. doi: 10.3969/j.issn.1001-506X.2019.09.10
5	YANG D G , ZHU Z L , ZHANG J C , et al. The overview of human localization and vital sign signal measurement using handheld IR-UWB through-wall radar[J]. Sensors, 2021, 21 (2): 402. doi: 10.3390/s21020402
6	ZHU Z L, YANG D G, LIANG B G. Vital sign signal extraction method based on permutation entropy and EMD algorithm for ultra-wideband radar[C]//Proc. of the 3rd International Conference on Electronic Information Technology and Computer Engineering, 2019.
7	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. doi: 10.12305/j.issn.1001-506X.2022.06.09
	SUN J M , YU S K , SUN J . Radar small sample target recognition method based on meta learning and its improvement[J]. Systems Engineering and Electronics, 2022, 44 (6): 1839- 1845. doi: 10.12305/j.issn.1001-506X.2022.06.09
8	晏媛, 孙俊, 孙晶明, 等. 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. doi: 10.19665/j.issn1001-2400.2021.02.002
	YAN Y , SUN J , SUN J J , et al. Radar few shot target recognition method and application analysis[J]. Systems Engineering and Electronics, 2021, 43 (3): 684- 692. doi: 10.19665/j.issn1001-2400.2021.02.002
9	王兆其. 虚拟人合成研究综述[J]. 中国科学院研究生院学报, 2000, 17 (2): 89- 98.
	WANG Z Q . A review of research on virtual human synthesis[J]. Journal of Graduate School, Academia Sinica, 2000, 17 (2): 89- 98.
10	DORP P , GROEN F C A . Human walking estimation with radar[J]. IEEE Proceedings Radar, Sonar and Navigation, 2003, 150 (5): 356- 365. doi: 10.1049/ip-rsn:20030568
11	DORP P V , GROEN F C A . Feature-based human motion parameter estimation with radar[J]. IET Radar, Sonar & Navigation, 2008, 2 (2): 135- 145.
12	CHEN V C , LI F , HO S S , et al. Micro-Doppler effect in radar: phenomenon, model, and simulation study[J]. IEEE Trans.on Aerospace and Electronic Systems, 2006, 42 (1): 2- 21.
13	ZHU Z L , YANG D G , ZHANG J C , et al. Dataset of human motion status using IR-UWB through wall radar[J]. Journal of Systems Engineering and Electronics, 2021, 32 (5): 1083- 1096. doi: 10.23919/JSEE.2021.000093
14	WU Q , ZHANG Y D , TAO W , et al. Radar based fall detection based on Doppler time frequency signatures for assisted living[J]. IET Radar, Sonar & Navigation, 2015, 9 (2): 164- 172.
15	SUN Z S, WANG J, SUN J P, et al. Parameter estimation method of walking human based on radar micro-Doppler[C]//Proc. of the IEEE Radar Conference, 2017: 567-570.
16	张军. 多站低频雷达运动人体微多普勒特征提取与跟踪技术[D]. 长沙: 国防科技大学, 2017.
	ZHANG J. Technique of human micro-Doppler feature extraction and tracking with multi-station low frequency radar[D]. Changsha: National University of Defense Technology, 2017.
17	OROVIC' I , STANKOVIC' S , AMIN M G . A new approach for classification of human gait based on time-frequency feature representations[J]. Signal Processing, 2011, 91 (6): 1448- 1456. doi: 10.1016/j.sigpro.2010.08.013
18	DING C W , ZHANG L , CHEN H , et al. Human motion re-cognition with spatial-temporal-convLSTM network using dynamic range-Doppler frames based on portable FMCW radar[J]. IEEE Trans.on Microwave Theory Technology, 2022, 70 (11): 5029- 5038. doi: 10.1109/TMTT.2022.3200097
19	DING C W , HONG H , ZOU Y , et al. Continuous human motion recognition with a dynamic range-Doppler trajectory method based on FMCW radar[J]. IEEE Trans.on Geoscience and Remote Sensing, 2019, 57 (9): 6821- 6831. doi: 10.1109/TGRS.2019.2908758
20	DING C W , CHAE R , WANG J , et al. Inattentive driving behavior detection based on portable FMCW radar[J]. IEEE Trans.on Microwave Theory Technology, 2019, 67 (10): 4031- 4041. doi: 10.1109/TMTT.2019.2934413
21	REN L Y , TRAN N , FARNAZ F , et al. Short-time state-space method for micro-Doppler identification of walking subject using UWB impulse Doppler radar[J]. IEEE Trans.on Microwave Theory Technology, 2018, 66 (7): 3521- 3534. doi: 10.1109/TMTT.2018.2829523
22	YAO Y R , LIU W B , ZHANG G , et al. Radar-based human activity recognition using hyperdimensional computing[J]. IEEE Trans.on Microwave Theory Technology, 2022, 70 (3): 1605- 1619. doi: 10.1109/TMTT.2021.3134992
23	SADREAZAMI H , BOLIC M , RAJAN S . Contactless fall detection using time-frequency analysis and convolutional neural networks[J]. IEEE Trans.on Industrial Informatics, 2021, 17 (10): 6842- 6851. doi: 10.1109/TII.2021.3049342
24	KIM Y , LING H . Human activity classification based on micro-Doppler signatures using a support vector machine[J]. IEEE Trans.on Geoscience and Remote Sensing, 2009, 47 (5): 1328- 1337. doi: 10.1109/TGRS.2009.2012849
25	JOKANOVIC' B , Amin M G . Suitability of data representation domains in expressing human motion radar signals[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14 (12): 2370- 2374. doi: 10.1109/LGRS.2017.2765341
26	AMIN M G , AHMAD F , ZHANG Y D , et al. Human gait recognition with cane assistive device using quadratic time-frequency distributions[J]. IET Radar, Sonar & Navigation, 2015, 9, 1224- 1230.
27	孙忠胜, 王俊, 毕严先, 等. 基于广义S变换的多人微多普勒特征分析[J]. 系统工程与电子技术, 2014, 36 (7): 1291- 1297. doi: 10.3969/j.issn.1001-506X.2014.07.11
	SUN Z S , WANG J , BI Y X , et al. Analysis of multi human micro Doppler signatures based on generalized S transform[J]. Systems Engineering and Electronics, 2014, 36 (7): 1291- 1297. doi: 10.3969/j.issn.1001-506X.2014.07.11
28	CHANG C C , LIN C J . LibSVM: a library for support vector machines[J]. ACM Trans.on Intelligent Systems and Technology, 2011, 2 (3): 1- 27.
29	HSU C W , LIN C J . A comparison of methods for multi-class support vector machines[J]. IEEE Trans.on Neural Networks, 2002, 13 (2): 415- 425.
30	杜浩. 基于深度学习的超宽带雷达人体行为辨识研究[D]. 长沙: 国防科技大学, 2020.
	DU H. Research on deep learning-based human behavior recognition in ultra-wideband radar[D]. Changsha: National University of Defense Technology, 2020.
31	蒋留兵, 李骢, 车俐. 超宽带雷达人体动作识别[J]. 电子测量与仪器学报, 2018, 32 (1): 129- 134.
	JIANG L B , LI C , CHE L . Human motion recognition by-ultra-wide band radar[J]. Journal of Electronic Measurement and Instrumentation, 2018, 32 (1): 129- 134.

参数类型	参数值	参数类型	参数值
信号体制	冲激脉冲信号	基线长度/m	0.15
中心频率/MHz	500	等效采样间隔/ps	200
采样点数目	768	脉冲重复周期/s	0.125
模数转换器实时采样率/(KSa/s)	400	脉宽/ns	2

行为	距离像
行走1 m/s
奔跑2.3 m/s
坐下
起立
跌倒
起身

真实	预测
真实	行走	奔跑	起立	坐下	起身	跌倒
行走	47	0	0	0	0	0
奔跑	0	45	0	0	0	0
起立	0	0	52	0	0	0
坐下	0	0	2	45	0	0
起身	0	0	0	0	44	1
跌倒	0	0	0	0	2	46

类型	PPV	TPR	F
行走	100	100	100
奔跑	100	100	100
起立	96.30	100	98.11
坐下	100	95.74	97.83
起身	95.65	97.78	96.70
跌倒	97.87	95.83	96.84

行为	PPV		TPR		F
行为	频域	时域	频域	时域	频域	时域
行走	100	100	95.16	100	97.52	100
奔跑	98.44	100	100	100	99.21	100
起立	91.18	63.04	95.38	44.62	93.23	52.25
坐下	96.72	63.86	92.19	77.94	94.40	70.20
起身	87.88	70.83	95.08	57.63	91.34	63.55
跌倒	92.42	68.49	88.41	86.21	90.37	76.34