Abstract:In the paper, we study approximation properties of the Malmquist-Takenaka-Christov (MTC) system. We show that the discrete MTC approximations converge rapidly under mild restrictions on functions asymptotic at infinity. This makes them particularly suitable for solving semi- and quasi-linear problems containing Fourier multipliers, whose symbols are not smooth at the origin. As a concrete application, we provide rigorous convergence and stability analyses of a collocation MTC scheme for solving the nonlinear Benjamin equation. We demonstrate that the method converges rapidly and admits an efficient implementation, comparable to the best spectral Fourier and hybrid spectral Fourier/finite-element methods described in the literature.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is how to solve the nonlinear Benjamin equation effectively and accurately. Specifically, the authors studied the approximation properties of the Malmquist - Takenaka - Christov (MTC) system and showed that in the case where the function has slight restrictions at infinity, the discrete MTC approximation can converge rapidly. This makes them particularly suitable for solving semilinear and quasilinear problems involving Fourier multipliers, especially those where the symbol is not smooth at the origin.
To be more specific, the nonlinear Benjamin equation proposed in the paper is:
\[
u_t = -\alpha u_x + \beta H[u_{xx}] + \gamma u_{xxx} - \delta (u^2)_x, \quad u(0) = u_0, \quad x \in \mathbb{R},
\]
where \( \alpha, \beta, \gamma, \delta \) are real parameters, and \( H[u](x) = \text{p.v.} \frac{1}{\pi} \int_{\mathbb{R}} \frac{u(y)}{y - x} dy \) is the standard Hilbert transform.
### Main problems and solutions
1. **Limitations of traditional methods**:
- Traditional numerical methods (such as the pseudo - spectral Fourier method, the mixed Fourier/finite - difference method, etc.) usually need to truncate the spatial domain to a large interval \([-L, L]\) and impose periodic boundary conditions. However, due to the jump discontinuity of the Hilbert transform symbol at the origin, the exact solution decays at most algebraically at infinity. This leads to a very large requirement for the truncation parameter \( L \), thus increasing the computational complexity.
2. **Advantages of the MTC system**:
- The functions in the MTC system are orthogonal rational functions and have many attractive computational properties. For example, they are eigenfunctions of the Hilbert transform, behave well under product operations, the differential matrix is diagonal and antisymmetric, and the discrete spectral coefficients can be efficiently calculated by the discrete fast Fourier transform (FFT).
- The paper proved that the convergence speed of the MTC - Fourier expansion is only controlled by the regularity and asymptotic behavior of the function on \(\mathbb{R}\setminus\{0\}\), allowing square - integrable singularities at the origin. This makes the MTC - Fourier approximation exhibit spectral convergence even when the function decays slowly at infinity.
3. **Application to the Benjamin equation**:
- The authors proposed a collocation method based on the MTC basis for solving the nonlinear Benjamin equation. Through strict convergence and stability analysis, it is proved that this method not only has a fast convergence speed but can also be efficiently implemented, which is comparable to the best spectral Fourier and mixed spectral Fourier/finite - element methods described in the literature.
In summary, the main contribution of this paper is to provide a new numerical method, which solves the difficulties encountered by traditional methods in dealing with the nonlinear Benjamin equation, especially when dealing with problems involving Fourier multipliers and having a non - smooth symbol at the origin.