The focal point method for solving systems of linear equations

From MaRDI portal





Using the fact that any equation of a linear system describes a hyperplane, a method is described that starting from a solution to one equation, by successively adding equations one by one, produces a solution for the large system. This method called focal point method makes use of the zero structure and sparseness of the system. An error analysis and an operation count is given. A comparison with a linear equation solver of the IMSL-package, testing the algorithm on a Hilbert matrix (which is not sparse) is given. One would wish to see comparisons with sparse matrix solvers on sparse matrices (for which this algorithm was claimed to be made) and also a more detailed description on how to cope with breakdowns of the algorithm.











This page was built for publication: The focal point method for solving systems of linear equations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1060531)