Abstract
Analytical solutions to acoustic scattering problems involving nonspherical shapes, such as spheroids and disks, have long been known and have many applications. However, these solutions require special functions that are not easily computable. For this reason, their asymptotic forms are typically used since they are more readily available. We explore these solutions and provide computational software for calculating their nonasymptotic forms, which are accurate over a wide range of frequencies and distances. This software, which runs in MATLAB, computes the solutions to acoustic scattering problems involving spheroids and disks by semi-analytical means, and is freely available from our webpage.
Introduction
In computational acoustics, numerical solvers like the Boundary Element Method (BEM) are essential for simulating sound interaction with complex geometries. However, as approximations, these methods require an analytical "gold standard" for validation. While solutions for spherical objects are well-established, nonspherical geometries such as spheroids and disks present significant mathematical challenges. Although the underlying theory has existed for over a century, it relies on spheroidal wave functions that are notoriously difficult to compute accurately. Consequently, researchers often resort to simplified asymptotic approximations limited to specific frequency ranges or distances. This work addresses this gap by providing a robust, non-asymptotic computational framework implemented in MATLAB.
Methodology
The authors utilize a semi-analytical approach based on the separation of variables in prolate and oblate spheroidal coordinate systems. This technique decomposes the Helmholtz equation—the fundamental governing equation for acoustic waves—into three independent ordinary differential equations.
To handle the resulting spheroidal wave functions, the software suite is divided into two primary libraries:
spheroidal: A high-precision library for computing the underlying wave functions.scattering: A library that applies these functions to solve physical scattering problems, expanding incident fields (such as point sources or plane waves) to calculate the scattered field.
A significant contribution of this framework is the inclusion of Robin boundary conditions. These conditions represent surfaces with properties between perfectly "soft" (Dirichlet) and perfectly "hard" (Neumann), offering greater physical realism. Furthermore, the framework elegantly treats infinitely thin disks as "degenerate" oblate spheroids, allowing a unified mathematical treatment for both 3D and 2D-like shapes.
Results and Validation
The software demonstrates exceptional precision and versatility. Visualizations of plane waves scattering off various geometries at different wavenumbers ($k=10$ and $k=25$) reveal the transition from simple scattering to complex interference patterns as frequency increases.
Validation is a core focus of this work. As shown in the "Check on Boundary Conditions," the numerical error in satisfying the prescribed boundary conditions is consistently between $10^{-12}$ and $10^{-15}$. This level of accuracy is near the limit of double-precision floating-point arithmetic, confirming the software's reliability as a benchmark for other numerical methods.
The software's public release on GitHub includes 18 routines covering various scattering scenarios, supported by precomputed tables of wave functions to enhance computational efficiency.
Conclusion
This MATLAB-based suite provides a vital resource for the acoustics community, offering a "ground truth" for validating modern numerical solvers. While currently focused on single-object scattering and dependent on the MATLAB environment, it fills a long-standing gap in accessible, high-precision tools for spheroidal wave functions. Future work may involve extending the framework to multiple interacting objects and porting the libraries to high-performance languages like C++ or Python to support larger-scale simulations.
Summary updated