Construction of analogs of the Lyapunov equation for a matrix polynomial (Q4940593)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 1410148
Language Label Description Also known as
default for all languages
No label defined
    English
    Construction of analogs of the Lyapunov equation for a matrix polynomial
    scientific article; zbMATH DE number 1410148

      Statements

      Construction of analogs of the Lyapunov equation for a matrix polynomial (English)
      0 references
      0 references
      2 March 2000
      0 references
      Gauss-Hurwitz problem
      0 references
      Lyapunov equation
      0 references
      matrix-function
      0 references
      Let a matrix-function of dimension \( p \times p\) be given which is uniquely defined and analytical in the domain \( \Lambda \) with \( \det F(\lambda) \not \equiv 0\) and let \( f(\lambda , \mu) \) be uniquely defined and an analytical function of two variables in the domain \( \Lambda \times {\overline \Lambda } , \) with \( f(\lambda , {\overline \mu}) \equiv {\overline f}(\mu , {\overline \lambda })\) for \( (\lambda , {\overline \mu}) \in \Lambda \times {\overline \Lambda } . \) Denote by \( \sigma_{0} \) a part of the spectrum \( \sigma (F) , \) which consists of \( n \) zeros of the function \( \det {F( \lambda)} \) and determine in the complex plane the following sets: \[ \Lambda^{+} = \{ \lambda : f(\lambda , {\overline \lambda })>0 \}, \quad \Lambda^{-} = \{ \lambda : f(\lambda , {\overline \lambda })<0 \} , \quad \Lambda^{0} = \{ \lambda : f(\lambda , {\overline \lambda })=0 \} . \] The problem is to estimate the numbers \( n^{+}, n^{-}, n^{0} \) of points of the set \( \sigma_{0} \) belonging to the sets \( \Lambda^{+}, \Lambda^{-} \) and \( \Lambda^{0} \) respectively and to find a criterion of inclusion \( \sigma_{0} \subset \Lambda^{+}\) in particular. This problem is a natural generalization of the Gauss-Hurwitz problem on distribution of roots of characteristic polynomials for autonomous systems relatively to the imaginary axis. To solve the problem, the author develops a method which is based on a construction and studying the linear matrix equation similar to the Lyapunov equation and which has the following form \( M_{f}X=Y, \) where \( X \) and \( Y \) are Hermitian matrices and \( M_{f} \) is a linear operator. In the case when the set \( \sigma_{0} \) is isolated by a closed contour \( \gamma\), the operator \( M_{f} \) has the following form \[ M_{f}=- {{1}\over{4\pi^{2}}} \oint_{\gamma} \oint_{{\overline \gamma}} f(\lambda , {\overline \mu }) D_{\lambda } X D_{\mu} ^{*} d \lambda d {\overline \mu}, \] with \( D_{\lambda } = F'(\lambda) F^{-1}(\lambda), \) and the total numbers of points of the set \( \sigma_{0} \) can be determined by the formula \[ n=\text{tr} \Delta , \quad \Delta = {{1}\over{2\pi i}} \oint_{\gamma} D_{\lambda } d \lambda . \] Depending on properties of solutions to the equation \( M_{f}X=Y \) and on the form of the operator \( M_{f} \) some theorems on location of points of the set \( \sigma_{0} \) are proved.
      0 references

      Identifiers