基于频带分解和距离加权的单矢量水听器浅海被动测距方法研究∗

摘要

A novel method is proposed for the passive source range estimation based on union processing of pressure and particle horizontal velocity. Autocorrelation functions’ warping spectra of pressure and particle velocities have frequency invariability. The spectra of the warped autocorrelation functions of pressure and particle horizontal velocity have the same line spectrum feature, while the spectrum of the warped autocorrelation function of particle vertical velocity possesses both line and broadband spectrum features. Moreover, the warped autocorrelation function’s spectrum of particle vertical velocity has more peaks, and the peak width is broader than those of pressure and particle horizontal velocity. In this paper, source ranges are estimated based on frequency band decomposition and distance weighting, and a guided source with a known range is employed to provide the crucial frequency invariant features. Experimental data in shallow water with an iso-speed profile are used to verify the approach which can reasonably estimate source ranges with the relative errors of the source ranging basically less than 7%. This method can effectively reduce the mainlobe width and background level of the cost function, and can significantly improve the resolution of source range estimation, compared with the results of conventional source ranging approach that uses a single pressure hydrophone.%本文针对浅海环境下中低频宽带脉冲声源被动测距问题，提出了一种声压和水平振速联合处理的被动测距方法。在浅海波导中，声压和质点振速的自相关函数卷绕(warping)谱具有稳定的频率特征。声压和水平振速的自相关卷绕谱具有相同的准线谱特征，垂直振速自相关卷绕谱具有宽谱叠加线谱的特征，与声压自相关卷绕谱相比，其尖峰个数比更多，且尖峰宽度更宽。利用引导源，本文提出了基于频带分解和距离加权的声压和水平振速联合被动测距方法。利用该方法对2008年冬季青岛海域综合实验中单矢量水听器接收的气枪信号进行处理，结果表明，该方法能够实现气枪声源的有效测距。与传统单声压水听器被动测距方法相比，该方法可以有效减小代价函数的主瓣宽度、提高测距精度。