Wavelet–Galerkin solutions of one dimensional elliptic problems

Abstract Daubechies wavelet bases are used for numerical solution of partial differential equations of one dimension by Galerkin method. Galerkin bases are constructed from Daubechies functions which are compactly supported and which constitute an orthonormal basis of L 2 ( R ) . Theoretical and numerical results are obtained for elliptic problems of second order with different types of boundary conditions. Optimal error estimates are also obtained. Comparison of solutions with simple finite difference method suggests that for this class of problems, the present method will provide a better alternative to other classical methods. The methodology can be generalized to multidimensional problems by taking care of some technical facts.