A Criterion for the choice of the interpolation kernel in Smoothed Particle Hydrodynamics

We study the problem of the appropriate choice of the interpolating kernel to be used in the evaluation of gradients of functions. Such interpolation technique is often used in applications, e.g. it is typical for Smoothed Particle Hydrodynamics (SPH). We propose a minimization procedure for selecting kernels in n-dimensions, in the class of regular, normalizable, symmetric functions having finite moments up to a sufficiently high order; the method is valid when the kernel width is position-dependent and allows to recover conservation laws at the same order of approximation that SPH, as interpolation technique, has when the kernel size is constant.