Application of Improved Cross Power Spectrum Phase Method to Acoustic Source Localization of Thunder
-
摘要: 互相关函数法是一种常用的时延估计方法,但在低信噪比和闪电存在多分叉的情况下,该方法的效果不理想。针对这一问题,该文选取广州地区2009年8月24日19:08:04和2012年5月5日08:01:00的两次闪电过程的雷声记录,结合高速摄像资料,比较了互相关函数法和改进的互功率谱相位法在雷声声源定位中的应用效果。首先,利用互相关函数法和改进的互功率谱相位法计算声源点的方向信息,定位结果与高速摄像照片对比具有较好的一致性,证明了定位方法的可靠性,对比结果表明:在低信噪比环境和多分叉闪电情况下,改进的互功率谱相位法能够获得比互相关函数法更好的定位效果。最后给出了综合利用改进的互功率谱相位法计算得到的方向信息和声源点与观测点的距离信息得到的雷声声源三维定位结果。
-
关键词:
- 信噪比;
- 多分叉闪电;
- 雷声;
- 改进的互功率谱相位法
Abstract: The thunder sound source is located by using single station lightning channel three-dimensional imaging system. Different time delaying estimation method is used, among which the cross correlation function method and the cross power spectrum phase method are mainly introduced. Considering the high frequency noise produced by the lightning process, reflection reverberation caused when the thunder reaches the ground, all kinds of noise is superimposed on the thunder signal (such as collection circuit itself noise), and an improved cross power spectrum phase delay estimation method is raised.Station imaging system is composed of microphone array and data acquisition card. A large amount of reliable data is collected by the system on the roof of Guangdong Provincal Meteorological Bureau building since 2009. Two thunder processing records in Guangzhou are selected, combined with the high speed camera data, and the cross correlation function method and the improvement cross power spectrum phase method are compared in the application of acoustic localization of thunder. First, the time difference of thunder signal reaching the different microphones is calculated by using the cross correlation function method and the improved cross power spectrum phase method. Then, the sound source azimuth and elevation angle information are solved by the array geometry. Comparing with the two-dimensional photographs observed by high speed camera, the imaging result is in good agreement, showing good reliability. As the cross correlation function method algorithm is based on the amplitude correlation, it can't distinguish the array arriving simultaneously from different sound source. Because its noise immunity is weak, the discrete imaging point can't depict the channel shape better. Instead, the improved cross power spectrum phase method calculates time delay by phase difference, and it has strong noise immunity and intensive imaging point, so it does better in discerning the branch channel. The contrast result indicates that the improved cross power spectrum phase method is better than the cross correlation function method for the low signal to noise ratio environments and multi-forked lightning. Finally, three-dimensional thunder source is obtained through the direction information and distance of thunder calculated with improved cross power phase method.The application of single station microphone array imaging system reduces the environmental requirements and costs. Although the accuracy of results is low, for close range, multi-branch, multi-ground lightning, the single station microphone array provides a simple and practical three-dimensional observation programs, and it has application prospects for small scale lightning monitoring, early warning and research. A real-time processing of three-dimensional imaging thunder sound source system is in plan, based on a single station of microphone array, and it will play its unique role in the three-dimensional structure of the lightning research. -
图 6 闪电F1203的定位结果 (a) 高速摄像拍摄到的图像 (镜头焦距14 mm),(b) 高速摄像拍摄到的图像 (镜头焦距24 mm),(c) 两种方法定位的声源点与二维闪电通道的对比
Fig. 6 The location results of lightning flash F1203(a) the high speed images of lightning F1203 (lens focal length is 14 mm), (b) the high speed images of lightning F1203(lens focal length is 24 mm), (c) the comparison between thunder source locations using two methods and two-dimensional channels
表 1 常用的加权函数
Table 1 Common weighting functions
编号 名称 加权函数 1 互相关函数法 (CCF) 1 2 互功率谱相位法 (CSP) Ψ12(ω)=1 3 幅度平方加权互功率谱相位法 (ASW-CSP) Ψ12(ω)=|G12(ω)|2 -
[1] 王道洪, 郄秀书, 郭昌明.雷电与人工引雷.上海:上海交通大学出版社, 2000:1-21. [2] 张义军, 周秀骥.雷电研究的回顾和进展.应用气象学报, 2006, 17(6):829-834. doi: 10.11898/1001-7313.20060619 [3] 马明, 吕伟涛, 张义军.1997—2006年我国雷电灾情特征.应用气象学报, 2008, 19(4):393-400. doi: 10.11898/1001-7313.20080402 [4] 刘欣生.雷电物理及人工引发雷电研究十年进展与展望.高原气象, 1999, 18(3):266-272. http://www.cnki.com.cn/Article/CJFDTOTAL-GYQX199903001.htm [5] Few A A.Acoustic Radiations from Lightning//Handbook of Atmospheric Electrodynamics.Florida:CRC Press, 1995:1-31. [6] 张义军, 孟青, 马明, 等.闪电探测技术发展和资料应用.应用气象学报, 2006, 17(5):611-620. doi: 10.11898/1001-7313.20060504 [7] 张文娟, 孟青, 吕伟涛, 等.时间差闪电监测网的误差分析和布局优化.应用气象学报, 2009, 20(4):402-410. doi: 10.11898/1001-7313.20090403 [8] Few A A.Lightning channel reconstruction from thunder measurement.J Geophys Res, 1970, 75:7515-7523. [9] MacGorman D R.Lightning Location in a Storm with Strong Wind Shear.Texas:Department of Space Physics and Astronomy, Rice University, 1978:1-83. [10] Akiyama H, Ichino K, Horii K.Channel reconstruction of trigge-red lightning flashes with bipolar currents from thunder measurements.Journal of Geophysical Research:Atmospheres (1984—2012), 1985, 90(D6):10674-10680. doi: 10.1029/JD090iD06p10674 [11] Arechiga R O, Johnson J B, Edens H E, et al.Acoustic localization of triggered lightning.Journal of Geophysical Research:Atmospheres (1984—2012), 2011, 116, D09103, doi: 10.1029/2010JD015248. [12] Johnson J B, Arechiga R O, Thomas R J, et al.Imaging thunder.Geophys Res Lett, 2011, 38, L19807, doi: 10.1029/2011GL049162. [13] 章涵, 王道洪, 吕伟涛, 等.基于雷声到达时间差的单站闪电通道三维定位系统.高原气象, 2012, 31(1):209-217. http://cdmd.cnki.com.cn/Article/CDMD-85101-1011117401.htm [14] Qiu S, Zhou B H, Shi L H.Synchronized observations of cloud-to-ground lightning using VHF broadband interferometer and acoustic arrays.Journal of Geophysical Research:Atmospheres (1984—2012), 2012, 117, D19204, doi: 10.1029/2012JD018542. [15] Few A A, Teer T L.The accuracy of acoustic reconstructions of lightning channels.J Geophys Res, 1974, 79(33):5007-5011. doi: 10.1029/JC079i033p05007 [16] 李俊, 张义军, 吕伟涛, 等.一次多回击自然闪电的高速摄像观测.应用气象学报, 2008, 19(4):401-411. doi: 10.11898/1001-7313.20080403 [17] 李俊, 吕伟涛, 张义军, 等.一次多分叉多接地的空中触发闪电过程.应用气象学报, 2010, 21(1): 95-100. doi: 10.11898/1001-7313.20100113 [18] 朱广信, 陈彪, 金蓉.基于传声器阵列的声源定位.电声技术, 2003(1): 34-37. http://cdmd.cnki.com.cn/Article/CDMD-10287-2007194606.htm [19] 林志斌, 徐柏龄.基于传声器阵列的声源定位.电声技术, 2004(5): 19-23. http://cdmd.cnki.com.cn/Article/CDMD-10287-2007194606.htm [20] 马晓红, 陆晓燕, 殷福亮.改进的互功率谱相位时延估计方法.电子与信息学报, 2004, 26(1):53-59. http://www.cnki.com.cn/Article/CJFDTOTAL-DZYX200401008.htm [21] Allen J B, Berkley D A.Image method for efficiently simulating small-room acoustics.The Journal of the Acoustical Society of America, 1979, 65:943-950. doi: 10.1121/1.382599 [22] Omologo M, Svaizer P.Acoustic Source Location in Noisy and Reverberant Environment Using CSP Analysis.Proc ICASSP, Atlanta, Georgia, USA, 1996, 2:921-924. [23] Omologo M, Svaizer P.Acoustic Event Localization Using a Cr-osspower-spectrum Phase Based Technique.Proc ICASSP, AdeIaide, South Australia, 1994:273-276. [24] 陆大絟.随机过程及其应用分析.北京:清华大学出版社, 1986: 367-376. [25] 常建平, 李海林.随机信号分析.北京:科学出版社, 2006:123-134. [26] 张景川, 袁萍, 欧阳玉花.雷声在大气中传播的吸收衰减特性研究.物理学报, 2010, 59(11):8287-8392. doi: 10.7498/aps.59.8287 [27] 欧阳玉花, 袁萍, 贾向东, 等.用信号处理技术及传播理论还原雷声频谱.物理学报, 2013, 62(8), 084303, doi: 10.7498/aps.62.084303. [28] 赵真, 侯自强.广义相位谱延时估计.声学学报, 1985, 10(4):201-215. http://www.cnki.com.cn/Article/CJFDTOTAL-XIBA198504000.htm [29] 杜功焕, 朱哲民, 龚秀芬, 等.声学基础.南京:南京大学出版社, 2001: 186-188.