Abstract:Systems of Prony type appear in various signal reconstruction problems such as finite rate of innovation, superresolution and Fourier inversion of piecewise smooth functions. We propose a novel approach for solving Prony-type systems, which requires sampling the signal at arithmetic progressions. By keeping the number of equations small and fixed, we demonstrate that such "decimation" can lead to practical improvements in the reconstruction accuracy. As an application, we provide a solution to the so-called Eckhoff's conjecture, which asked for reconstructing jump positions and magnitudes of a piecewise-smooth function from its Fourier coefficients with maximal possible asymptotic accuracy -- thus eliminating the Gibbs phenomenon.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is how to improve the accuracy of signal reconstruction by improving the solution method of Prony - type systems, especially for the Fourier reconstruction problem of piecewise - smooth functions. Specifically, the goals of the paper include:
1. **Propose a new method for solving Prony - type systems**: By sampling the signal at arithmetic - progression positions (i.e., "decimation"), keeping the number of equations small and fixed, thus avoiding the numerical instability of these systems to a certain extent and improving the reconstruction accuracy.
2. **Solve the Eckhoff conjecture**: The Eckhoff conjecture raises the problem of reconstructing the jump positions and magnitudes of piecewise - smooth functions from Fourier coefficients. Traditional methods can only achieve "half - precision", while the decimation technique proposed in this paper can achieve complete asymptotic precision, thus completely eliminating the Gibbs phenomenon.
3. **Improve the Fourier reconstruction accuracy of piecewise - smooth functions**: By introducing the decimation technique, the paper shows how to reconstruct the jump positions and magnitudes of piecewise - smooth functions with the maximum possible asymptotic accuracy, thus solving the problem of accuracy degradation caused by the Gibbs phenomenon.
### Formula summary
- **Prony system equation**:
\[
m_k=\sum_{j = 1}^{K}c_jz_j^k, \quad c_j,z_j\in\mathbb{C}, \quad k\in\mathbb{N}
\]
- **More general model**:
\[
f(x)=\sum_{j = 1}^{K}\sum_{\ell=0}^{\ell_j - 1}b_{\ell,j}\delta^{(\ell)}(x - x_j), \quad b_{\ell,j}\in\mathbb{R}, \quad x_j\in[-\pi,\pi]
\]
- **Form after Fourier transformation**:
\[
m_k=\sum_{j = 1}^{K}z_j^k\sum_{\ell=0}^{\ell_j - 1}c_{\ell,j}k^\ell, \quad c_{\ell,j}\in\mathbb{C}, \quad |z_j| = 1
\]
- **Error estimate**:
\[
|\Delta c_{i,j}|\leq C(i,\ell_j)\left(\frac{2}{\delta_p}\right)^R\left(\frac{1}{2}+\frac{R}{\delta_p}\right)^{\ell_j}t^{\ell_j - i}p_i\left(1+\frac{|c_{i - 1,j}|}{|c_{\ell_j - 1,j}|}\right)\epsilon
\]
\[
|\Delta z_j|\leq\frac{2}{\ell_j!}\left(\frac{2}{\delta_p}\right)^R\frac{1}{|c_{\ell_j - 1,j}|}p^{-\ell_j}\epsilon
\]
where \(\delta_p=\min_{i\neq j}|z_j^p - z_i^p|\), and \(C(i,\ell_j)\) is an explicit constant.
### Conclusion
By using the decimation technique, the paper not only improves the solution accuracy of Prony - type systems, but also solves the key problems in the Eckhoff conjecture, realizes the high - precision Fourier reconstruction of piecewise - smooth functions, and eliminates the influence of the Gibbs phenomenon. This result has important application values in signal processing, image reconstruction and other fields.