Spectrum completion and inverse Sturm-Liouville problems

TL;DR

This paper introduces a novel numerical method to complete spectra and recover the potential in Sturm-Liouville problems using eigenvalue data, Neumann series of Bessel functions, and linear algebra techniques.

Contribution

It presents a new approach for spectrum completion and inverse problem solving without additional information, improving accuracy and efficiency.

Findings

Eigenvalues can be computed with uniform absolute accuracy.

The method effectively completes spectra without extra potential or boundary condition data.

The approach enables accurate potential recovery from two spectra.

Abstract

Given a finite set of eigenvalues of a regular Sturm-Liouville problem for the equation -y{\prime}{\prime}+q(x)y={\lambda}y, the potential q(x) of which is unknown. We show the possibility to compute more eigenvalues without any additional information on the potential q(x). Moreover, considering the Sturm-Liouville problem with the boundary conditions y{\prime}(0)-hy(0)=0 and y{\prime}({\pi})+Hy({\pi})=0, where h, H are some constants, we complete its spectrum without additional information neither on the potential q(x) nor on the constants h and H. The eigenvalues are computed with a uniform absolute accuracy. Based on this result we propose a new method for numerical solution of the inverse Sturm-Liouville problem of recovering the potential from two spectra. The method includes the completion of the spectra in the first step and reduction to a system of linear algebraic equations in the second. The potential q(x) is recovered from the first component of the solution vector. The approach is based on special Neumann series of Bessel functions representations for solutions of Sturm-Liouville equations possessing remarkable properties and leads to an efficient numerical algorithm for solving inverse Sturm-Liouville problems.