Numerical simulation of an inverse problem: Testing the influence of data

Abstract

In this paper, we consider the iterative algorithm proposed by Kozlov, Mazya and Fomin (KMF) for solving the inverse problem for Laplace equation which consiste to determine the missing conditions on a part of the boundary from the overspecified coditions on the accessible part. Several formulations are discussed according to the measure of the underspecified boundary and the condition on the other part to conclude the relationschip between the data problems and the rate of convergence. Numerical tests are developed with the software FreeFem with smooth and non-smooth domains.