A least-squares finite element method for the Helmholtz equation (Q804263)

The author proposes a new finite element method in order to solve the Helmholtz equation for both the solution and its gradient. He modifies an idea proposed by \textit{G. J. Fix}, \textit{M. D. Gunzburger} and \textit{R. A. Nicolaides} [Comput. Math. Appl. 5, 87-98 (1979; Zbl 0422.65064)] consisting in the transformation of the original problem into a first order system; the system is then dealt with an appropriate Sobolev space, where a convenient bilinear operator is shown to be coercive, thus allowing the application of well-known results leading to a priori error estimates. A numerical example is included to demonstrate the convergence properties of the proposed method.