Error estimates and extrapolation for the numerical solution of Mellin convolution equations

In this paper we consider a quadrature method for the numericalsolution of a second-kind integral equation over the interval,where the integral operator is a compact perturbation of a Mellinconvolution operator. This quadrature method relies upon a singularitysubtraction and transformation technique. Stability and convergenceorder of the approximate solution are well known. We shall derivethe first term in the asymptotics of the error which shows that,in the interior of the interval, the approximate solution convergeswith higher order than over the whole interval. This implieshigher orders of convergence for the numerical calculation ofsmooth functionals to the exact solution. Moreover, the asymptoticsallows us to define a new approximate solution extrapolatedfrom the dilated solutions of the quadrature method over mesheswith different mesh sizes. This extrapolated solution is designedto improve the low convergence order caused by the non-smoothnessof the exact solution even when the transformation techniquecorresponds to slightly graded meshes. Finally, we discuss theapplication to the double-layer integral equation over the boundaryof polygonal domains and report numerical results.