The equivalent refraction index for the acoustic scattering by many small obstacles: with error estimates

TL;DR

This paper demonstrates that the acoustic scattering by many small obstacles can be approximated by an equivalent refraction index, with explicit error estimates, useful for designing acoustic materials with desired properties.

Contribution

It introduces a new method to approximate the farfield scattering by numerous small obstacles with an effective refraction index, including explicit error bounds.

Findings

Convergence of scattered waves to an effective refraction index as obstacle size decreases.

Explicit error estimates for the approximation when obstacles are similar and distribution is Hölder continuous.

Application potential in designing acoustic materials with tailored refraction properties.

Abstract

Let $M$ be the number of bounded and Lipschitz regular obstacles $D_j, j:=1, ..., M$ having a maximum radius $a$, $a<<1$, located in a bounded domain $\Omega$ of $\mathbb{R}^3$. We are concerned with the acoustic scattering problem with a very large number of obstacles, as $M:=M(a):=O(a^{-1})$, $a\rightarrow 0$, when they are arbitrarily distributed in $\Omega$ with a minimum distance between them of the order $d:=d(a):=O(a^t)$ with $t$ in an appropriate range. We show that the acoustic farfields corresponding to the scattered waves by this collection of obstacles, taken to be soft obstacles, converge uniformly in terms of the incident as well the propagation directions, to the one corresponding to an acoustic refraction index as $a\rightarrow 0$. This refraction index is given as a product of two coefficients $C$ and $K$, where the first one is related to the geometry of the obstacles (precisely their capacitance) and the second one is related to the local distribution of these obstacles. In addition, we provide explicit error estimates, in terms of $a$, in the case when the obstacles are locally the same (i.e. have the same capacitance, or the coefficient $C$ is piecewise constant) in $\Omega$ and the coefficient $K$ is H$\ddot{\mbox{o}}$lder continuous. These approximations can be applied, in particular, to the theory of acoustic materials for the design of refraction indices by perforation using either the geometry of the holes, i.e. the coefficient $C$, or their local distribution in a given domain $\Omega$, i.e. the coefficient $K$.