新自适应阈值的小波滤波在被动声源定位中应用

Application of Wavelet Filter with a New Adaptive Threshold in Passive Sound Source Localization

在被动声源定位研究领域,广义互相关法是应为最为广泛的一种定位方法。而基于小波滤波的广义互相关法克服了传统广义互相关法需要信号和噪声先验知识的局限性;很好地提高了相关噪声条件下的非平稳信号的定位精度。针对小波阈值滤波提出一种新的自适应阈值以获得更好的去噪效果,提高基于广义互相关法的声源定位精度。最后,仿真比较了几种自适应阈值去噪效果以说明提出的自适应阈值的优越性,并进行定位实验验证基于新自适应阈值小波滤波的被动声源定位精度得到极大改善。

In the field of passive sound source localization, the generahzed cross - correlation method is one of the most widely used methods. The GCC method based on the wavelet filter overcomes the limitations of the traditional GCC method which requires prior knowledge of signal and noise, and im- proves the localization accuracy of the non - stationary signal under the condition of correlated noise. A new adaptive threshold for wavelet threshold filter was proposed in this paper to achieve better denoising effect and improve the localization accuracy of the sound source based on the GCC method. Finally, the de- noising effects of several adaptive thresholds were simulated to illustrate the superiority of the proposed a- daptive threshold. Finally, the localization experiment results showed that the localization accuracy of pas- sive sound source based on the wavelet filter with the proposed adaptive threshold was improved signifi- cantly.