An index formula in connection with meromorphic approximation

DOI10.1007/S11785-012-0249-3MaRDI QIDQ2440925zbMATH OpenOpenAlexFDO

Publication date 20 March 2014

Published in Complex Analysis and Operator Theory (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1103.3906

badly approximable matrix-valued functions best approximation Hankel operator index formula Nehari-Takagi problem superoptimal approximation Toeplitz operator

Mathematics Subject Classification ID

Spaces of vector- and operator-valued functions (46E40) Linear operator methods in interpolation, moment and extension problems (47A57) Toeplitz operators, Hankel operators, Wiener-Hopf operators (47B35)

Abstract: Let

P h i

be a continuous

n i m e s n

matrix-valued function on the unit circle

T

such that the

(k - 1)

th singular value of the Hankel operator with symbol

P h i

is greater than the

k

th singular value. In this case, it is well-known that

P h i

has a unique superoptimal meromorphic approximant

Q

in

H_{(k)}^{i n f t y}

; that is,

Q

has at most

k

poles in the unit disc

m a t h b b D

(i.e. the McMillan degree of

Q

in

m a t h b b D

is at most

k

) and

Q

minimizes the essential suprema of singular values

s_{j} ((P h i - Q) (z e t a))

,

j g e q 0

, with respect to the lexicographic ordering. For each

j g e q 0

, the essential supremum of

s_{j} ((P h i - Q) (z e t a))

is called the

j

th superoptimal singular value of

P h i

of degree

k

. We prove that if

P h i

has

n

non-zero superoptimal singular values of degree

k

, then the Toeplitz operator

T_{P h i - Q}

with symbol

P h i - Q

is Fredholm and has index [ ind T_{Phi-Q}=dimker T_{Phi-Q}=2k+dimmathcal{E}, ] where

m a t h c a l E = x i i n k e r H_{Q} : | H_{P h i} x i |_{2} = | (P h i - Q) x i |_{2}

and

H_{P h i}

denotes the Hankel operator with symbol

P h i

. In fact, this result can be extended from continuous matrix-valued functions to the wider class of

k

-emph{admissible} matrix-valued functions, i.e. essentially bounded

n i m e s n

matrix-valued functions

P h i

on

T

for which the essential norm of the Hankel operator

H_{P h i}

is strictly less than the smallest non-zero superoptimal singular value of

P h i

of degree

k

.

Review text

This paper deals with matrix-valued functions on the unit circle and the associated Hankel and Toeplitz operators. \(\Psi \in L^{\infty} (M_{m,n})\) is said to be very badly approximable if it has a superoptimal approximation in \(H^{\infty}(M_{m,n})\). If \(\Psi\) further is \(0\)-admissible and has \(n\) non-zero superoptimal singular values of degree \(0\), then for the associated Toeplitz operator \(T_{\Psi}\), in Section 4 of this paper, the author establishes the interesting formula \[ \operatorname{ind}T_{\Psi} = \dim\operatorname{ker}T_{\Psi} > 0. \]

Recommendations

Cites work

This page was built for publication: An index formula in connection with meromorphic approximation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2440925)