Approximation of the Hilbert Transform on the real line using Hermite zeros

From MaRDI portal





The Hilbert transform \(H(G,t)\) is defined by the integral \[ H(G,t):= \int_{\mathbb{R}} {G(x)\over x-t} dx= \lim_{\varepsilon\to 0^+} \int_{|x-t|\geq\varepsilon} {G(x)\over x-t} dx,\quad t\in\mathbb{R} \] provided that the limit on the right-hand side exists. The approximations and numerical evaluations of \(H(G,t)\) are of great interest to mathematicians. Here, the authors approximate \(H(G,t)\) on the real line by using the zeros of Hermite polynomials. They give a Gauss quadrature rule and a product quadrature rule. They obtain a formula having greater computational cost than the one of the Gauss-type quadrature formula, but it proves to be more useful in collocation methods for solving singular integral equation on the real line. The authors also establish convergence conditions for their procedure in Theorem 3.2 and also propose a simple algorithm when the parameter \(t\) is large. Extensive numerical testing and comparisons with other procedures are performed and all results confirm the theoretical estimates. Some important examples are given in the last section.



Cites work


Cited in
(20)


Describes a project that uses

Uses Software






This page was built for publication: Approximation of the Hilbert Transform on the real line using Hermite zeros

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4529711)