Inverse Spectral Problems for Linked Vibrating Systems and Structured Matrix Polynomials

Abstract: We show that for a given set

L a m b d a

of

n k

distinct real numbers

l a m b d a_{1}, l a m b d a_{2}, l d o t s, l a m b d a_{n k}

and

k

graphs on

n

nodes,

G_{0}, G_{1}, l d o t s, G_{k - 1}

, there are real symmetric

n i m e s n

matrices

A_{s}

,

s = 0, 1, l d o t s, k

, such that the matrix polynomial

A (z) := A_{k} z^{k} + c d o t s + A_{1} z + A_{0}

has

L a m b d a

as its spectrum, the graph of

A_{s}

is

G_{s}

for

s = 0, 1, l d o t s, k - 1

, and

A_{k}

is an arbitrary positive definite diagonal matrix. When

k = 2

, this solves a physically significant inverse eigenvalue problem for linked vibrating systems (see Corollary 5.3).

This page was built for publication: Inverse Spectral Problems for Linked Vibrating Systems and Structured Matrix Polynomials