Derivative generation from multivariate scattered data by functional minimization (Q1070756)

Many multivariate interpolation schemes require as data values of derivatives. If they are not available, they have to be generated suitably. Derivative values may be constructed by minimizing \(\int (D^ ku)^ 2dx\) over a suitable space subject to interpolation of given Lagrangian data. An example illustrates that the procedure with \(k=2\) or 3 is more suitable than \(k=1\) or \(k=4\). In the computations, the Bézier-Bernstein form of multivariate polynomials is used.