Acoustic Technology, Department of Electrical Engineering, DTU Technical University of Denmark, Building 352, Oersteds Plads, DK-2800 Kongens Lyngby, Denmark.
J Acoust Soc Am. 2012 Dec;132(6):3818-25. doi: 10.1121/1.4763988.
In conventional near-field acoustic holography (NAH) it is not possible to distinguish between sound from the two sides of the array, thus, it is a requirement that all the sources are confined to only one side and radiate into a free field. When this requirement cannot be fulfilled, sound field separation techniques make it possible to distinguish between outgoing and incoming waves from the two sides, and thus NAH can be applied. In this paper, a separation method based on the measurement of the particle velocity in two layers and another method based on the measurement of the pressure and the velocity in a single layer are proposed. The two methods use an equivalent source formulation with separate transfer matrices for the outgoing and incoming waves, so that the sound from the two sides of the array can be modeled independently. A weighting scheme is proposed to account for the distance between the equivalent sources and measurement surfaces and for the difference in magnitude between pressure and velocity. Experimental and numerical studies have been conducted to examine the methods. The double layer velocity method seems to be more robust to noise and flanking sound than the combined pressure-velocity method, although it requires an additional measurement surface. On the whole, the separation methods can be useful when the disturbance of the incoming field is significant. Otherwise the direct reconstruction is more accurate and straightforward.
在传统的近场声全息(NAH)中,无法区分阵列两侧的声音,因此,要求所有声源都局限于一侧,并辐射到自由场中。当无法满足此要求时,声场分离技术可用于区分两侧的出入波,从而可以应用 NAH。本文提出了一种基于两层粒子速度测量的分离方法和另一种基于单层压力和速度测量的分离方法。这两种方法都使用具有单独的出入波传递矩阵的等效源公式,以便可以独立地对阵列两侧的声音进行建模。提出了一种加权方案,以考虑等效源和测量表面之间的距离以及压力和速度之间的幅度差异。已经进行了实验和数值研究来检验这些方法。与组合压速法相比,双层速度法对噪声和侧翼声似乎更稳健,尽管它需要额外的测量表面。总的来说,当入射场的干扰很显著时,分离方法可能很有用。否则,直接重建更准确和直接。