基于SVD和MPSO-SVM的光纤周界

doi:10.3969/j.issn.1001-506X.2020.08.02

摘要/Abstract

摘要：

针对光纤振动信号有噪声干扰、识别信号类型准确率不高且识别时间长的问题,提出了基于奇异值分解(singular value decomposition, SVD)和改进粒子群优化支持向量机(modified particle swarm optimization support vector machine, MPSO-SVM)的识别方法。首先,采用SVD对信号去噪,根据奇异值序列二阶差分谱单边极小值原则确定信号重构秩阶次。其次,提取振动信号特征,利用串行特征融合(serial feature fusion, SFF)方法组建特征向量组。最后,利用MPSO-SVM进行分类识别,提高识别精度和算法效率。采用实测信号进行验证,结果表明,信噪比有明显提升,信号平均识别率较粒子群优化支持向量机(particle swarm optimization support vector machine, PSO-SVM)提升5%。该方法较传统神经网络识别方法有较好的效果,具有实际应用价值。

关键词: 光纤光学, 信号识别, 奇异值分解, 支持向量机, 粒子群优化算法

Abstract:

A recognition method based on singular value decomposition (SVD) and modified particle swarm optimization support vector machine (MPSO-SVM) is proposed for the optical fiber vibration signals with noise interference, low accuracy and long recognition time. Firstly, SVD is used to denoise the signal, and the rank order of signal reconstruction is determined according to the single-side minimum principle of second-order difference spectrum of singular value sequence. Secondly, the vibration signal features are extracted and a set of feature vectors is constructed by means of serial feature fusion (SFF). Finally, MPSO-SVM is used for classification and recognition to improve the accuracy and efficiency of the algorithm. The measured signal is used for verification. The results show that the signal to noise ratio is significantly improved, and the average recognition rate is 5% higher than that of PSO-SVM. This method performs better than the traditional neural network recognition method and has the practical application value.

Key words: fiber optics, signal recognition, singular value decomposition (SVD), support vector machine (SVM), particle swarm optimization (PSO)

中图分类号:

TP212

马愈昭, 王强强, 王瑞松, 熊兴隆. 基于SVD和MPSO-SVM的光纤周界[J]. 系统工程与电子技术, 2020, 42(8): 1652-1661.

Yuzhao MA, Qiangqiang WANG, Ruisong WANG, Xinglong XIONG. Optical fiber perimeter vibration signal recognition based on SVD and MPSO-SVM[J]. Systems Engineering and Electronics, 2020, 42(8): 1652-1661.

图/表 19

图1

图2

图3

表1

图4

图5

图6

表2

表3

图7

图8

图9

表4

表5

表6

表7

表8

图10

表9

参考文献 26

1	裴丽, 翁思俊, 吴良英, 等. 光纤激光传感系统的研究进展[J]. 中国激光, 2016, 43 (7): 7- 13.
	PEI L , WENG S J , WU L Y , et al. Progress in optical fiber laser sensing system[J]. Chinese Journal of Lasers, 2016, 43 (7): 7- 13.
2	叶青, 潘政清, 王照勇, 等. 相位敏感光时域反射仪研究和应用进展[J]. 中国激光, 2017, 44 (6): 7- 20.
	YE Q , PAN Z Q , WANG Z Y , et al. Research and application progress of phase sensitive optical time domain reflectometer[J]. Chinese Journal of Lasers, 2017, 44 (6): 7- 20.
3	饶云江. 长距离分布式光纤传感技术研究进展[J]. 物理学报, 2017, 66 (7): 158- 176.
	RAO Y J . Research progress of long-distance distributed optical fiber sensing technology[J]. Acta Physica Sinica, 2017, 66 (7): 158- 176.
4	ALLWOOD G , WILD G , HINCKLEY S . Optical fiber sensors in physical intrusion detection systems: a review[J]. IEEE Sensors Journal, 2016, 16 (14): 5497- 5509.
5	MAHMOUD S S , KATISFOLIS J . Real-timedistributed fiber optic sensor for security systems: performance, event classification and nuisance mitigation[J]. Photonic Sensors, 2012, 2 (3): 225- 236.
6	QADEER M A, ISLAM J, RIZVI S, et al. Design and implementation of a fiber optic perimeter intrusion detection system[C]//Proc.of the International Bhurban Conference on Applied Sciences & Technology, 2016: 48-50.
7	XIN L , JIN B Q , BAI Q , et al. Distributed fiber-optic sensors for vibration detection[J]. Sensors, 2016, 16 (8): 1164.
8	JIANG J F , AN J C , LIU K , et al. A fast positioning algorithm for the asymmetric dual Mach-Zehnder int erferometric infrared fiber vibration sensor[J]. Infrared Physics & Technology, 2017, 85, 359- 363.
9	MA C Y , LIU T G , LIU K , et al. Long range distributed fiber vibration sensor using an asymmetric dual Mach-Zehnder interferometers[J]. Journal of Lightwave Technology, 2016, 34 (9): 2235- 2239.
10	TAO W , LIU T G , LIU K , et al. An EMD-based filtering algorithm for the fiber-optic SPR sensor[J]. IEEE Photonics Journal, 2016, 8 (3): 1- 8.
11	查翔, 倪世宏, 张鹏. 一类非线性信号去噪的奇异值分解有效迭代方法[J]. 电子与信息学报, 2015, 37 (6): 1330- 1335.
	ZHA X , NI S H , ZHANG P . Effective iteration method of a class of nonlinear signal denoising based on singular value decomposition[J]. Journal of Electronics & Information Technology, 2015, 37 (6): 1330- 1335.
12	周长林, 钱志升, 王勤民, 等. 基于PSO-SVM方法的电源线传导泄漏信号识别与还原[J]. 电子与信息学报, 2018, 40 (9): 2206- 2211.
	ZHOU C L , QIAN Z S , WANG Q M , et al. Recognition and reconstruction of conduction leakage signal via power line based on PSO-SVM method[J]. Journal of Electronics & Information Technology, 2018, 40 (9): 2206- 2211.
13	熊兴隆, 张琬童, 李猛, 等. 基于局部均值分解和串行特征融合的光纤周界振动信号识别[J]. 光学学报, 2019, 39 (2): 43- 52.
	XIONG X L , ZHANG W T , LI M , et al. Recognition of intrusion signal based on LMD and SFF on optical fiber perimeter warning system[J]. Acta Optica Sinica, 2019, 39 (2): 43- 52.
14	熊兴隆, 张琬童, 冯磊, 等. 基于多重分形谱的光纤周界振动信号识别[J]. 光子学报, 2019, 48 (2): 56- 67.
	XIONG X L , ZHANG W T , FENG L , et al. Optical fiber perimeter vibration signal recognition based on multifractal spectrum[J]. Acta Photonica Sinica, 2019, 48 (2): 56- 67.
15	蒋立辉, 刘杰生, 熊兴隆, 等. 光纤周界入侵信号特征提取与识别方法的研究[J]. 激光与红外, 2017, 47 (7): 906- 913.
	JIANG L H , LIU J S , XIONG X L , et al. Research on intrusion signal extraction and recognition of optical fiber sensor perimeter[J]. Laser & Infrared, 2017, 47 (7): 906- 913.
16	LI Z X , DAI W X . Local mean decomposition combined with SVD and application in telemetry vibration signal processing[J]. Applied Mechanics and Materials, 2013, 347-350, 854- 858.
17	王建国, 李健, 刘颖源. 一种确定奇异值分解降噪有效秩阶次的改进方法[J]. 振动与冲击, 2014, 33 (12): 176- 180.
	WANG J G , LI J , LIU Y Y . An improved method for determining effective order rank of SVD denoising[J]. Journal of Vibration and Shock, 2014, 33 (12): 176- 180.
18	严强强, 盛守照, 周俊, 等. 一种改进奇异值分解降噪方法研究[J]. 电光与控制, 2018, 25 (9): 22- 25, 41.
	YAN Q Q , SHENG S Z , ZHOU J , et al. A noise reduction method based on improved singular value decomposition[J]. Electronics Optics & Control, 2018, 25 (9): 22- 25, 41.
19	AGRAWAL S , AGRAWAL J , KAUR S , et al. A comparative study of fuzzy PSO and fuzzy SVD-based RBF neural network for multi-label classification[J]. Neural Computing & Applications, 2016, 29 (1): 1- 12.
20	CHEN Z F, QIAN P D. Application of PSO-RBF neural network in network intrusion detection[C]//Proc.of the 3rd International Symposium on Intelligent Information Technology Application, 2009: 362-364.
21	IMAM T, TICKLE K. Class information adapted kernel for support vector machine[M]. Neural information processing. models and applications. Berlin: Springer, 2010.
22	LIU K , TIAN M , LIU T G , et al. A high-efficiency multiple events discrimination method in optical fiber perimeter security[J]. Journal of Lightwave Technology, 2015, 33 (23): 4885- 4890.
23	MA P F , LIU K , JIAN J F , et al. Probabilistic event discrimination algorithm for fiber optic perimeter security systems[J]. Journal of Lightwave Technology, 2018, 36 (1): 2069- 2075.
24	BANDT C , POMPE B . Permutation entropy: a natural complexity measure for time series[J]. Physical Review Letters, 2002, 88 (17): 174102- 0.
25	陈怀忠, 何仁初, 史桂丽. 基于改进PSO算法的自动配棉工艺参数优化设计[J]. 纺织学报, 2014, 35 (6): 142- 147.
	CHEN H Z , HE R C , SHI G L . Parameter optimization design for automatic cotton assorting based on improved PSO algo rithm[J]. Journal of Textile Research, 2014, 35 (6): 142- 147.
26	CLERC M. The swarm and the queen: towards a deterministic and adaptive particle swarm optimization[C]//Proc.of the Congress on Evolutionary Computation, 1999: 1951-1957.

激光器功率/mW	线宽/kHz	采样频率/(M/s)	采样个数
20	10	5	100 000

重构秩阶次	SNR/dB	RMSE
5	39.191	0.000 58
6	37.704	0.000 69
7	36.408	0.000 81

去噪方法	振动类型	平均SNR/dB	平均RMSE
小波分析	无扰动	38.320	0.000 52
	跑跳	41.996	0.000 54
	攀爬	31.010	0.001 4
	敲击	20.808	0.003 7
互补集合经验模态分解	无扰动	11.206	0.015
	跑跳	12.612	0.018 5
	攀爬	14.972	0.007 4
	敲击	11.156	0.011 2
SVD	无扰动	46.304	0.000 21
	跑跳	40.467	0.000 27
	攀爬	37.367	0.001 3
	敲击	24.125	0.002 8

识别方法	进化200次	进化100次	进化50次	进化20次
PSO-SVM	100	83.3	70	56.6
MPSO-SVM	100	100	100	73.3

识别方法	进化200次	进化100次	进化50次	进化20次
PSO-SVM	10.361	4.952	2.458	1.014
MPSO-SVM	9.732	4.832	2.392	0.989