小波变换应用于NQR信号处理
|
摘要
爆炸物探测技术在反恐、维护社会安全和战场保障等诸多领域有重要意义,提高爆炸物探测的效率,即提高探测的准确度、降低虚警率、减少探测时间,是爆炸物探测技术研究发展的方向。
在诸多爆炸物探测方法中,核电四极矩共振(NQR,Nuclear Quadrupole Resonance)技术依靠探测爆炸物所含元素本质特性而日益受到世人关注。与其他探测方法相比,NQR技术的优势体现在其不但可以判断爆炸物的存在,而且可以确定爆炸物的种类。然而由于NQR信号极其微弱,很易湮灭在噪声和干扰中不易探测,这是制约NQR技术应用的难点。要将NQR技术应用于实际的爆炸物检测中,就必须解决NQR信号的处理问题。
为了提取淹没在噪声和干扰背景中的NQR信号,人们采用了多种检测算法,典型算法包括：NQR信号的相干累加、谱分析、自适应滤波、奇异值分解等。然而这些算法对于某些特殊爆炸物探测效率低,实时性不高,限制了NQR技术的应用。所以进一步研究更高效的算法十分必要。
小波变换是一种时频分析方法,可以根据信号和噪声的不同特性在不同尺度上进行非线性滤波,在改善信噪比的同时,既有很高的时间(位置)分辨率,而且对信号的形式不敏感。这是传统的滤波方法所无法比拟的,小波变换法特别适用于弱信号的检测和定位。所以可以利用小波阀值去噪的方法对NQR信号进行处理。
本文在详细研究了NQR技术原理的基础上,根据NQR信号自身的特性,提出了将小波变换应用于NQR信号处理方法。实验证明,该处理方法有效抑制了振铃拖尾的干扰,缩短了探测时间,提高了探测效率。
Explosive detection technology is important in many fields. It can be used to fight against terrorism, protect security of society, protect battlefield, etc.Improving the efficiency of explosive detection means improving the accuracy of detection, reducing false alarm rate and reducing detection time are the development direction of the explosive detection technology.
In a number of explosive detection methods, the nuclear quadrupole resonance (NQR, Nuclear Quadrupole Resonance) technique gets more and more attention by the people because it relys on detecting of elements contained in the explosive. Compared with other detection methods, the advantage of NQR technology is that it can not only determine the presence of explosives, but also determine the types of explosive. However, the NQR signal is extremely weak and it is easy to be submerged in the noise and interference.That is the restriction of NQR technology. If we want to make use of NQR technology in the actual project, we must solve the problems in the NQR signal processing.
To extract the NQR signal from the context of noise and interference, it has used a variety of detection algorithms, typical algorithms include:NQR signal of coherent accumulation, spectral analysis, adaptive filtering. However, these algorithms are lack of low detection efficiency, and searching more efficient algorithm is very necessary.
Wavelet transform is a time-frequency analysis mothod, it can take nonlinear filtering according to the different characteristics of signal and noise in different scales, so that it can improve signal to noise ratio. Wavelet transform mothod have high time(position) resolution, and is not sensitive to the signal form.This is the traditional filtering methods can not achieve, the wavelet transform method is suitable for weak signal detection and signal location particularly. So we can use the threshold method of wavelet transform to processing the NQR signal.
In this paper, research the principle of NQR technology, a wavelet transform method is used in NQR signal processing based on the characteristics of NQR signal. Experiments show that the processing algorithm can effectively eliminate background noise, interference suppression of the ring, and increase the detection efficiency.
在诸多爆炸物探测方法中,核电四极矩共振(NQR,Nuclear Quadrupole Resonance)技术依靠探测爆炸物所含元素本质特性而日益受到世人关注。与其他探测方法相比,NQR技术的优势体现在其不但可以判断爆炸物的存在,而且可以确定爆炸物的种类。然而由于NQR信号极其微弱,很易湮灭在噪声和干扰中不易探测,这是制约NQR技术应用的难点。要将NQR技术应用于实际的爆炸物检测中,就必须解决NQR信号的处理问题。
为了提取淹没在噪声和干扰背景中的NQR信号,人们采用了多种检测算法,典型算法包括：NQR信号的相干累加、谱分析、自适应滤波、奇异值分解等。然而这些算法对于某些特殊爆炸物探测效率低,实时性不高,限制了NQR技术的应用。所以进一步研究更高效的算法十分必要。
小波变换是一种时频分析方法,可以根据信号和噪声的不同特性在不同尺度上进行非线性滤波,在改善信噪比的同时,既有很高的时间(位置)分辨率,而且对信号的形式不敏感。这是传统的滤波方法所无法比拟的,小波变换法特别适用于弱信号的检测和定位。所以可以利用小波阀值去噪的方法对NQR信号进行处理。
本文在详细研究了NQR技术原理的基础上,根据NQR信号自身的特性,提出了将小波变换应用于NQR信号处理方法。实验证明,该处理方法有效抑制了振铃拖尾的干扰,缩短了探测时间,提高了探测效率。
Explosive detection technology is important in many fields. It can be used to fight against terrorism, protect security of society, protect battlefield, etc.Improving the efficiency of explosive detection means improving the accuracy of detection, reducing false alarm rate and reducing detection time are the development direction of the explosive detection technology.
In a number of explosive detection methods, the nuclear quadrupole resonance (NQR, Nuclear Quadrupole Resonance) technique gets more and more attention by the people because it relys on detecting of elements contained in the explosive. Compared with other detection methods, the advantage of NQR technology is that it can not only determine the presence of explosives, but also determine the types of explosive. However, the NQR signal is extremely weak and it is easy to be submerged in the noise and interference.That is the restriction of NQR technology. If we want to make use of NQR technology in the actual project, we must solve the problems in the NQR signal processing.
To extract the NQR signal from the context of noise and interference, it has used a variety of detection algorithms, typical algorithms include:NQR signal of coherent accumulation, spectral analysis, adaptive filtering. However, these algorithms are lack of low detection efficiency, and searching more efficient algorithm is very necessary.
Wavelet transform is a time-frequency analysis mothod, it can take nonlinear filtering according to the different characteristics of signal and noise in different scales, so that it can improve signal to noise ratio. Wavelet transform mothod have high time(position) resolution, and is not sensitive to the signal form.This is the traditional filtering methods can not achieve, the wavelet transform method is suitable for weak signal detection and signal location particularly. So we can use the threshold method of wavelet transform to processing the NQR signal.
In this paper, research the principle of NQR technology, a wavelet transform method is used in NQR signal processing based on the characteristics of NQR signal. Experiments show that the processing algorithm can effectively eliminate background noise, interference suppression of the ring, and increase the detection efficiency.
引文
[1]房旭民,张国进,高攀,等.基于Apollo信号控制台的NQR炸药探测方法.波谱学杂志.2004,21(3)：295-303
[2]吴招才,刘天佑.地震数据去噪中的小波方法.地球物理学进展.2008,23(2)：493-499
[3]周新建,张国进,高攀,等.NQR探测系统振铃拖尾抑制方法研究.核电子学与探测技术.2008,28(3)：477-480
[4]房旭民,徐政.一种探测炸药的核四极矩共振/电磁感应方法.同济大学学报,2004,32(5)：632-635
[5]程晓晟,朱佳政.利用核电四极矩共振检测行李包裹中的爆炸物.物理学报,1998,27(7)：423-425
[6]张永利.核四极矩共振信号处理方法研究.青岛,中国电波所,[硕士毕业论文]2008
[7]娄扬.基于TMS320C6713的炸药探测系统信号处理：[硕士学位论文].青岛：中国海洋大学,2008
[8]倪宏伟,房旭民,等.地雷探测技术.北京：国防工业出版社,2003.
[9]肖立志.核磁共振成像测井与岩石核磁共振及应用.北京：科学出版社,2003
[10]何忠伟,房旭民,徐政.核电四级矩共振探测技术研究.核电子学与探测技术.2003,23(6)：604-608
[11]M. Ostafin, B. Nogaj. Detection of Plastic Explosives in Luggage with N Nuclear Quadrupole Resonance Spectroscopy[J]. Applied Magnetic Resonance,2000,19:571 -578
[12]刘兆和,吴腾芳,顾文斌等.核电四极矩共振爆炸物探测原理与实验.解放军理工大学学报(自然科学版),2006,7(1)：59-63
[13]李志强,金余恒.基于核电四极矩共振技术的爆炸物检测系统的数据处理及信号识别算法[J].核电子学与探测技术,2006,11(6)：586-590
[14]原普,毛云志,郭华民.爆炸物检测中微弱NQR信号的处理.电波科学学报,2006,21(2)：204-207
[15]夏佑林,叶朝辉.14N核四极共振自旋系统自旋-晶格驰豫时间的测量.波谱学杂志,1994,11(4)：327-335
[16]Rebort M.Deas, Ian A.burch and Daniel M.port. The detection of RDX and TNT mine like targets by nuclear quadrupole resonance. Proc. SPIE VOl.4742, p.482-489
[17]M.Ostafin, B. Nogaj. Detection of Plastic Explosives in Luggage with N Nuclear Quadrupole Resonance Spectroscopy. Applied Magetic Resonance,2000,571-578
[18]郝国欣,金燕波,周新建,等.NQR爆炸物探测系统中减少振铃拖尾影响的几种措施.工兵装备研究,2007
[19]陈春,张国进,郭华民,等.利用核电四极矩共振技术探测爆炸物.波谱学杂志,2005, 22(4)：367-372
[20]丁玉美,阔永红,高新波.数字信号处理.西安：西安电子科学技术大学出版社,2003
[21]姚天任,孙洪.现代数字信号处理[M].武汉：华中科技大学出版社,1999
[22]张贤达.现代信号处理.北京：清华大学出版社,1996
[23]胡广书.数字信号处理-理论、算法与实现[M].北京：清华大学出版社,2003
[24]徐科军.信号分析与处理[M].北京：清华大学出版社,2006.
[25]赵振维,娄扬,金燕波,等.基于自适应滤波技术的NQR信号处理[J].电波科学学报,2008,23(3)：429-424
[26]林耀荣.自适应滤波理论及其在回波消除中的应用研究：[博士学位论文].广州：华南理工大学电子与通信工程系,1999
[27]谷源涛.LMS算法收敛性能研究及应用：[博士学位论文].北京：清华大学电子系,2003
[28]田宝凤.基于胎儿心率信号提取的自适应去噪算法的研究：[硕士学位论文].长春：吉林大学电子系,2004
[29]Hector F.Cancino-De Greiff,Rubersy Ramos-Garcia,Juan V.Lorenzo-Ginori,Signal De-Noising in Magnetic Resonance Spectroscopy Using Wavelet Transforms.
[30]张永利,孟宪礼.基于奇异值分解的NQR信号去噪分析[J].信息技术,2008,8：47-49
[31]葛哲学,沙威.小波分析理论与MATLAB R2007实现,2007
[32]潘泉,张磊,孟晋丽.小波滤波方法及应用.北京：清华大学出版社,2005
[33]潘泉,孟晋丽,张磊,等.小波滤波方法及应用.电子与信息学报.2007,29(1)：236-242
[34]吕瑞兰.小波阀值去噪的性能分析及基于能量元的小波阀值去噪方法研究.[硕士学位论文].杭州：浙江大学,2003
[35]张哗.信号时频分析及应用[M].哈尔滨,哈尔滨工业大学出版社,2006
[36]Coifman R R,Donoho D L.Translation-invariant de-noising. In:Wavelets in Statistic of Lecture Notes in statistics 103.New York:Sprinter-Verlag,1994.125-150
[37]Lang M,et al. Noise reduction using an undecimated discrete wavelet transform. IEEE Signal Processing Letters,1996,3(1):10-12
[38]Nason G P, Silverman B W. The stationary wavelet transform and some statistical applications. Lecture Notes in Statistics,1995,103:281-299
[39]Barrall G A,Derbyka,Drew A J. Development of quadrupole resonance confirmation system. Pro SPIE,2004,5415(2):1256-1267
[40]J. M. McCool and B. Widrow, Principles and Applications of Adaptive Filters:A Tutorial Review.1980 IEEE ISCAS pp.1143-1157
[41]胡昌华,张军波,夏军,张伟,基于MATLAB的系统分析与设计小波分析,西安：西安电子科技大学出版社,1999
[42]张志涌,等.精通MATLAB6.5版教程.北京航天航空大学出版社,2003
[2]吴招才,刘天佑.地震数据去噪中的小波方法.地球物理学进展.2008,23(2)：493-499
[3]周新建,张国进,高攀,等.NQR探测系统振铃拖尾抑制方法研究.核电子学与探测技术.2008,28(3)：477-480
[4]房旭民,徐政.一种探测炸药的核四极矩共振/电磁感应方法.同济大学学报,2004,32(5)：632-635
[5]程晓晟,朱佳政.利用核电四极矩共振检测行李包裹中的爆炸物.物理学报,1998,27(7)：423-425
[6]张永利.核四极矩共振信号处理方法研究.青岛,中国电波所,[硕士毕业论文]2008
[7]娄扬.基于TMS320C6713的炸药探测系统信号处理：[硕士学位论文].青岛：中国海洋大学,2008
[8]倪宏伟,房旭民,等.地雷探测技术.北京：国防工业出版社,2003.
[9]肖立志.核磁共振成像测井与岩石核磁共振及应用.北京：科学出版社,2003
[10]何忠伟,房旭民,徐政.核电四级矩共振探测技术研究.核电子学与探测技术.2003,23(6)：604-608
[11]M. Ostafin, B. Nogaj. Detection of Plastic Explosives in Luggage with N Nuclear Quadrupole Resonance Spectroscopy[J]. Applied Magnetic Resonance,2000,19:571 -578
[12]刘兆和,吴腾芳,顾文斌等.核电四极矩共振爆炸物探测原理与实验.解放军理工大学学报(自然科学版),2006,7(1)：59-63
[13]李志强,金余恒.基于核电四极矩共振技术的爆炸物检测系统的数据处理及信号识别算法[J].核电子学与探测技术,2006,11(6)：586-590
[14]原普,毛云志,郭华民.爆炸物检测中微弱NQR信号的处理.电波科学学报,2006,21(2)：204-207
[15]夏佑林,叶朝辉.14N核四极共振自旋系统自旋-晶格驰豫时间的测量.波谱学杂志,1994,11(4)：327-335
[16]Rebort M.Deas, Ian A.burch and Daniel M.port. The detection of RDX and TNT mine like targets by nuclear quadrupole resonance. Proc. SPIE VOl.4742, p.482-489
[17]M.Ostafin, B. Nogaj. Detection of Plastic Explosives in Luggage with N Nuclear Quadrupole Resonance Spectroscopy. Applied Magetic Resonance,2000,571-578
[18]郝国欣,金燕波,周新建,等.NQR爆炸物探测系统中减少振铃拖尾影响的几种措施.工兵装备研究,2007
[19]陈春,张国进,郭华民,等.利用核电四极矩共振技术探测爆炸物.波谱学杂志,2005, 22(4)：367-372
[20]丁玉美,阔永红,高新波.数字信号处理.西安：西安电子科学技术大学出版社,2003
[21]姚天任,孙洪.现代数字信号处理[M].武汉：华中科技大学出版社,1999
[22]张贤达.现代信号处理.北京：清华大学出版社,1996
[23]胡广书.数字信号处理-理论、算法与实现[M].北京：清华大学出版社,2003
[24]徐科军.信号分析与处理[M].北京：清华大学出版社,2006.
[25]赵振维,娄扬,金燕波,等.基于自适应滤波技术的NQR信号处理[J].电波科学学报,2008,23(3)：429-424
[26]林耀荣.自适应滤波理论及其在回波消除中的应用研究：[博士学位论文].广州：华南理工大学电子与通信工程系,1999
[27]谷源涛.LMS算法收敛性能研究及应用：[博士学位论文].北京：清华大学电子系,2003
[28]田宝凤.基于胎儿心率信号提取的自适应去噪算法的研究：[硕士学位论文].长春：吉林大学电子系,2004
[29]Hector F.Cancino-De Greiff,Rubersy Ramos-Garcia,Juan V.Lorenzo-Ginori,Signal De-Noising in Magnetic Resonance Spectroscopy Using Wavelet Transforms.
[30]张永利,孟宪礼.基于奇异值分解的NQR信号去噪分析[J].信息技术,2008,8：47-49
[31]葛哲学,沙威.小波分析理论与MATLAB R2007实现,2007
[32]潘泉,张磊,孟晋丽.小波滤波方法及应用.北京：清华大学出版社,2005
[33]潘泉,孟晋丽,张磊,等.小波滤波方法及应用.电子与信息学报.2007,29(1)：236-242
[34]吕瑞兰.小波阀值去噪的性能分析及基于能量元的小波阀值去噪方法研究.[硕士学位论文].杭州：浙江大学,2003
[35]张哗.信号时频分析及应用[M].哈尔滨,哈尔滨工业大学出版社,2006
[36]Coifman R R,Donoho D L.Translation-invariant de-noising. In:Wavelets in Statistic of Lecture Notes in statistics 103.New York:Sprinter-Verlag,1994.125-150
[37]Lang M,et al. Noise reduction using an undecimated discrete wavelet transform. IEEE Signal Processing Letters,1996,3(1):10-12
[38]Nason G P, Silverman B W. The stationary wavelet transform and some statistical applications. Lecture Notes in Statistics,1995,103:281-299
[39]Barrall G A,Derbyka,Drew A J. Development of quadrupole resonance confirmation system. Pro SPIE,2004,5415(2):1256-1267
[40]J. M. McCool and B. Widrow, Principles and Applications of Adaptive Filters:A Tutorial Review.1980 IEEE ISCAS pp.1143-1157
[41]胡昌华,张军波,夏军,张伟,基于MATLAB的系统分析与设计小波分析,西安：西安电子科技大学出版社,1999
[42]张志涌,等.精通MATLAB6.5版教程.北京航天航空大学出版社,2003