瞬态信号的小波变换波达方向估计  

Direction of arrival estimation of transient signals using wavelet transform

作　　者：李冉冉 张君 鲍明[2] 胡小青[2] 管鲁阳[2] 许耀华[1] LI Ranran;ZHANG Jun;BAO Ming;HU Xiaoqing;GUAN Luyang;XU Yaohua(College of Electronic Information Engineering,Anhui University Hefei 230601;Institue of Acoustics,Chinese Academy of Sciences Beijing 100190;College of Automation,Northwestern Polytechnical University Xi'an 710129)

机构地区：[1]安徽大学电子信息工程学院,合肥230601 [2]中国科学院声学研究所,北京100190 [3]西北工业大学自动化学院,西安710129

出　　处：《声学学报》2019年第1期77-85,共9页Acta Acustica

基　　金：国家自然科学基金项目(61501448);中国科学院战略性先导专项项目(XDA06020201)资助

摘　　要：针对瞬态信号方位估计问题,提出了基于连续小波变换的多重信号分类测向算法(CWT_MUSIC)。首先由信号特征确定小波尺度参数,构造Morlet小波,对信号进行小波变换,利用获得的小波变换系数建立多分辨时频阵列信号模型,并据此模型设计基于子空间的MUSIC算法以实现瞬态信号的波达方向估计;然后对该算法的多分辨与误差性能进行分析,最后仿真实验和实际爆炸试验验证了所提出的CWT_MUSIC算法能有效地提高空间谱的分辨率和DOA估计性能。A multiple signal classification algorithm based on the continuous wavelet transform(CWT_MUSIC) is proposed to estimate the Direction Of Arrival(DOA) of the transient signal. Firstly, the scale parameter of the wavelet is determined according to the signal characteristics, and the transient signals is processed using the Morlet wavelet transform constructed by those scale parameters. The wavelet coefficients are used to establish a multi-resolution timefrequency array signal model that is introduced into the MUSIC algorithm to perform DOA estimation of the transient signal. Secondly, the multi-resolution and error performance of this algorithm are analyzed. Finally, some simulations and outfield explosion experiments are conducted. Both results showed that the proposed CWT_MUSIC algorithm can improve the spatial spectral resolution and the DOA estimation performance.

关 键 词：连续小波变换 波达方向估计 瞬态信号 MUSIC算法 MORLET小波 估计性能 方位估计 尺度参数 

分 类 号：TN911.7[电子电信—通信与信息系统]