Abstract:In this study, the influence of a test vector selection used in subspace migration to reconstruct the shape of a sound-hard arc in a two-dimensional inverse acoustic problem is considered. In particular, a new mathematical structure of imaging function is constructed in terms of the Bessel functions of the order 0, 1, and 2 of the first kind based on the structure of singular vectors linked to the nonzero singular values of a Multi-Static Response (MSR) matrix. This structure indicates that imaging performance of subspace migration is highly related to the unknown shape of arc. The simulation results with noisy data indicate support for the derived structure.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: in the two - dimensional inverse acoustic problem, reconstruct the shape of the sound - hard arc through the subspace migration technique. Specifically, the focus of the research is on the influence of the choice of test vectors on the imaging performance of subspace migration.
### Problem Background
In traditional inverse scattering problems, especially for the imaging of sound - hard arcs, common methods include Newton's iterative method, inverse Fourier transform, hybrid methods, MUSIC algorithm, etc. However, these methods face problems such as high computational complexity and the need for prior information when dealing with multiple arcs. In contrast, non - iterative methods such as MUSIC and subspace migration can be used for multi - arc imaging, but their success depends on the selection of appropriate test vectors and requires knowledge of the unit outer normal vectors of the arcs.
### Research Objectives
This paper aims to explore the mathematical structure of the subspace migration imaging function, especially how to use Bessel functions to describe this structure. By analyzing the relationship between the singular vectors and non - zero singular values of the Multi - Static Response (MSR) matrix, the author hopes to reveal the influence of the choice of test vectors on the imaging results and provide a theoretical basis for selecting the optimal test vectors.
### Main Contributions
1. **Establishment of Mathematical Structure**: The author establishes the mathematical expression of the subspace migration imaging function by introducing Bessel functions (the first - kind Bessel functions of order 0, 1, and 2).
2. **Influence of Test Vectors**: The research shows that the imaging performance is highly dependent on the choice of test vectors. If the test vector is close to the unit outer normal vector of the arc, a better imaging effect can be obtained; conversely, if the test vector is orthogonal to the outer normal vector, the shape of the arc cannot be effectively identified.
3. **Numerical Simulation Verification**: Through numerical simulation, the validity of the proposed mathematical structure is verified, and the imaging results under different test vector selections are shown.
### Formula Summary
According to the content of the paper, the key formulas are as follows:
1. **Expression of Imaging Function**:
- When \( c_n\approx\nu(y_m) \):
\[
F(x)\approx\sum_{m = 1}^M J_0(k|x - y_m|)^2
\]
- When \( c_n=\theta_n \):
\[
F(x)\approx2\sum_{m = 1}^M\left(\frac{x - y_m}{|x - y_m|}\cdot\nu(x_m)\right)^2J_1(k|x - y_m|)^2
\]
- When \( c_n\equiv\xi\) and \(\xi\neq\nu(y_m)\) or \(\xi\neq\theta_n\):
\[
F(x)\approx\sum_{m = 1}^M\left\{(\nu(y_m)\cdot\xi)\left(J_0(k|x - y_m|)-J_2(k|x - y_m|)\right)-2\left(\frac{x - y_m}{|x - y_m|}\cdot\nu(y_m)\right)\left(\frac{x - y_m}{|x - y_m|}\cdot\xi\right)J_2(k|x - y_m|)\right\}^2
\]
2. **Integral Identity**:
- When \(\xi\neq\zeta\):
\[
\frac{1}{N}\sum_{n = 1}^N(\theta_n\cdot\xi)(\theta_n\cdot\zeta)e^{ik\theta_n\cdot x}=\frac{1}{2\pi}\int_{S^1}(\theta\