An extended GCRD algorithm for parametric univariate polynomial matrices and application to parametric Smith form (Q2674009)

This paper (extension of [\textit{T. Kailath}, Linear systems. Englewood Cliffs, N.J.: Prentice-Hall, Inc. (1980; Zbl 0454.93001)]) introduces an algorithm for computing the extended GCRD of parametric univariate matrices.This entails a discussion of parameters such that for each different value, there is a GCRD. Sections 2 and 3 introduce the problem for univariate polynomial matrices. As one can expect, the computation of GCRD implies the computing a minimal Gröbner basis for the submodule generated by the rows of the given matrices (see Theorems 9 and 10); furthermore, the computation of such a minimal basis allows to develop another algorithm for computed the extended GCRD (see Theorem 12). Section 4 introduce the new contributions. The authors extend the ideas introduced in Section 3, in order to compute the extended GCRD of parametric univariate matrices. Logically, usual Gröbner bases must be replaced by comprehensive Gröbner systems which provide the discussion of parameters. Finally, Section 5 introduced the computation of Smith normal form.