On the reduction and inversion of the nodal admittance matrix in rational form

A direct method for the reduction or inversion of the nodal admittance matrix in rational form is described. The problem of redundant factor generation is satisfactorily overcome and discussion is given to the further problems which arise in applying the method on a digital computer. The method may be used to obtain one or more network functions as rational functions in the Laplace operator with constant coefficients.