Condition of orthogonality of spectral matrix function
Results concerning a (2,2) system of ordinary differential equations due to M. G. Krejn are extended to a corresponding \((2m,2m)\) system [\textit{M. G. Krejn}, Dokl. Akad. Nauk SSSR 105, 637-640 (1955; Zbl 0068.28502)]. The solution of the Krein system leads to a one parameter set of functions in \(z\), \(P(x,z)\) \(x\geq 0\), \(-\infty< z<\infty\) which are orthogonal on \((-\infty, \infty)\). The analogous discrete problem due to A. N. Kolmogorov is also extended. In this problem, \(P_n(z)\) the solution of a discrete system, is a sequence of polynomials orthogonal on the unit circle. Both of these systems correspohd to stationary stochastic processes which are respectively continuous and discrete in time. Two necessary and sufficient condition are given in both cases for the spectral matrix function to be orthogonal. These conditions also assure that the corresponding stationary processes are singular. The conditions reduce to the Krein and Kolmogorov conditione when \(m=1\). For the entire collection see [Zbl 0959.00055].
- To the spectral theory of Krein systems
- Krein orthogonal entire matrix functions and related Lyapunov equations: a state space approach
- Krein systems
- The inverse problem for orthogonal Krein matrix functions
- A necessary and sufficient condition for the existence of the spectral matrix of a differential system
This page was built for publication: Condition of orthogonality of spectral matrix function
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2717372)
