Computation of canonical forms for single input linear systems over Hermite rings
- 1 IMUVA-Mathematics Research Institute, Universidad de Valladolid
-
2
Universidad de León
info
-
3
Universidad de Valladolid
info
- Galindo Pastor, Carlos (coord.)
- Gimenez, Philippe (coord.)
- Hernando Carrillo, Fernando (coord.)
- Monserrat Delpalillo, Francisco José (coord.)
- Moyano-Fernández, Julio José (coord.)
Editorial: Servei de Comunicació i Publicacions ; Universitat Jaume I
ISBN: 978-84-19647-46-7
Año de publicación: 2023
Páginas: 99-102
Congreso: Encuentro de Álgebra Computacional y Aplicaciones (17. 2022. Castelló de la Plana)
Tipo: Aportación congreso
Resumen
Let R be a commutative ring with the property that unimodular vectors can be completed to invertible matrices. Such a ring is called Hermite in the sense of Lam [4].In this note we construct a canonical form for matrix pairs (A, b), where A ∈ Rn×n and b ∈ Rn×1, under the feedback equivalence relation: (A, b) and (A, b) are equivalent if and only if A = P AP −1 + P bK and b = P b, for some matrices P ∈ GLn(R) and K ∈ R1×n.When R is a principal ideal domain, the canonical form is easily computed by using standard Hermite and Smith normal forms. When R is a polynomial ring K[x1, . . . , xt ], with K a field, the previously obtained canonical form remains valid, and can be determined by means of effective calculations. Our procedure consists mainly of elementary operations, combined with an adaptation of the currently available algorithms to solve theunimodular completion problem, used in the context of the Quillen-Suslin’s theorem that solves Serre’s conjecture.