On the rational generating functions of Toeplitz matrices
In their 2010 paper [Oper. Theory: Adv. Appl. 199, 25--118 (2010; Zbl 1203.15020)] \textit{G. Heinig} and \textit{K. Rost}, introduced the rational generating functions for an \(n \times n\) Toeplitz matrix and presented a parameterized formula for the rational generating functions with McMillan degree \(n\) when the Toeplitz matrix is nonsingular. Building on this pioneering work, the authors of the present paper study the rational generating functions of an arbitrary nonzero Toeplitz matrix (not necessarily square) with a prescribed allowable McMillan degree.\N\NDenote the monic greatest common divisor of a pair of complex polynomials \(s(z)\) and \(t(z)\) by \(\text{gcd}\big(s(z), t(z)\big)\). If \(r(z)\) is a nonzero rational function, then there exists a pair of nonzero polynomials \(s(z)\) and \(t(z)\) such that \N\[\Nf(z) = \frac{s(z)}{t(z)},\quad\gcd\big(s(z), t(z)\big)=1.\N\]\NThe McMillan degree of the rational function \(r(z)\) is defined by \N\[\N\deg r(z) :=~ \max\big\{\deg\,s(z) \;,\; \deg\,t(z)\big\}.\N\]\NGiven a nonzero Toeplitz matrix \(T := (a_{i-j})_{i,j=0}^{m-1,n-1} \in \mathbb{C}^{m \times n}\) with \({a_0 = a_0^+ + a_0^-}\), the rational generating function problem can be formulated in the following way: find all rational functions \(\varphi(z)\) with prescribed mcMillan degree, whose power series expansions at the origin and at infinity are \N\[\N\varphi(z) = a_0^+ +a_1z + a_2z^2 + \dots + a_{m-1}z^{m-1} + \mathcal{O}(z^m)\quad (z \rightarrow 0),\N\]\Nand \N\[\N\varphi(z) = -a_0^- - \frac{a_{-1}}{z} - \frac{a_{-2}}{z^2} - \dots - \frac{a_{1-n}}{z^{n-1}} + \mathcal{O}(z^{-n})\quad (z \rightarrow \infty),\N\] \Nrespectively. If a rational function \(\varphi(z)\) with \(\deg \varphi(z) = r\) both of the above, then \(\varphi(z)\) is referred to as a rational generating function of \(T\) with McMillan degree \(r\). In this case, \(r\) is referred to as an allowable McMillan degree of \(T\).\N\NTwo questions naturally arise here:\N\begin{itemize}\N\item[1.] Which positive integer is an allowable McMillan degree of \(T\)?\N\item[2.] How to describe the rational generating functions with a prescribed allowable McMillan degree?\N\end{itemize}\N\NIn the present paper, the authors provide complete answers to these two questions in the case where \(T\) is a nonzero Toeplitz matrix by introducing the \textit{characteristic degrees} and the \textit{characteristic polynomial pair of \(T\)}. It is important to emphasize that the characteristic degrees defined here are different from the characteristics of a nonzero Toeplitz matrix as defined and used by Heinig and Rost [loc. cit.].\N\NAt the end of the paper, the authors present two examples to demonstrate the feasibility and efficiency of their proposed method.
- A Schur-based algorithm for computing bounds to the smallest eigenvalue of a symmetric positive definite Toeplitz matrix
- A Variant of the Gohberg–Semencul Formula Involving Circulant Matrices
- Algebraic methods for Toeplitz-like matrices and operators
- An Algorithm for the Inversion of Finite Toeplitz Matrices
- An Optimal Circulant Preconditioner for Toeplitz Systems
- Conjugate Gradient Methods for Toeplitz Systems
- Conjugate-gradient preconditioning methods for shift-variant PET image reconstruction
- Continued Fractions which Correspond to Power Series Expansions at Two Points
- Hankel matrices generated by the Markov parameters of rational functions
- scientific article; zbMATH DE number 3991705 (Why is no real title available?)
- scientific article; zbMATH DE number 3561857 (Why is no real title available?)
- scientific article; zbMATH DE number 3451988 (Why is no real title available?)
- scientific article; zbMATH DE number 2118874 (Why is no real title available?)
- scientific article; zbMATH DE number 3219899 (Why is no real title available?)
- Introduction to Bézoutians
- Kernel structure of block Hankel and Toeplitz matrices and partial realization
- Positive-definite Toeplitz completion in DOA estimation for nonuniform linear antenna arrays. I: Fully augmentable arrays
- Positive-definite Toeplitz completion in DOA estimation for nonuniform linear antenna arrays. II: Partially augmentable arrays
- Stable row recurrences for the Padé table and generically superfast lookahead solvers for non-Hermitian Toeplitz systems
- The Minimum Eigenvalue of a Symmetric Positive-Definite Toeplitz Matrix and Rational Hermitian Interpolation
- Weighted Toeplitz regularized least squares computation for image restoration
- Über Potenzreihen, die im Innern des Einheitskreises beschränkt sind. I, II.
This page was built for publication: On the rational generating functions of Toeplitz matrices
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6876132)