解析声能流Capon空间谱估计

Capon spatial spectrum estimation based on analytic acoustic energy flux

文章中把解析函数理论用于声矢量信号分析,提出了一个新概念"解析声能流",给出了单矢量传感器输出的解析声能流表达式。并将这个新概念用于描述声矢量阵输出,得到了一种新的矢量阵输出模型"伪解析声能流"。利用该模型,推导了声矢量阵的Capon空间谱估计的表达式。理论分析和计算机仿真表明:1、解析声能流的实部和虚部具有正交的偶极子指向性,并且可以用移相的方法来旋转其主极大方向;2、在平面波和二维质点振速条件下,解析声能流将矢量叠加统一描述在它的相位中,从而可以综合描述声场的相位干涉和矢量结构;3、基于"伪解析声能流"模型,传统的Capon空间谱估计算法无需改动就可以直接用于声矢量信号处理,与传统的Capon空间谱估计算法相比,具有全空间无模糊定向能力、更低的旁瓣和对短时信号有更强的处理能力。

The theory of analytic function is applied to analyze acoustic vector signals, and a new concept of analytic acoustic energy flux (AAEF) is brought forward in this paper. The AAEF expression for a single hydrophone output is derived. By applying the new concept to describe the output of an acoustic vector array, a new model of pseudo analytic acoustic energy flux (PAAEF) is established. Using this model, an algorithm of Capon spatial estimation of the acoustic vector array is derived. The following conclusions can be drawn based on a theoretical analysis and computer simulations : (1) Real and imaginary parts of AAEF have mutually orthogonal dipole directivity, which can be rotated by shifting the phase of pressure. (2) Under the condition of plane wave incidence and two-dimensional particle vibration, AAEF describes both phase interference and vector addition of a sound field by only using its phase. (3) Based on the PAAEF model, the algorithms of Capon spatial spectral estimation can be directly used to process vector signals with unambiguous source azimuth finding capability, low side-lobes and efficient processing power for short duration signals.