Multidimensional integral operators with homogeneous-difference kernels (Q414050)

From MaRDI portal
Revision as of 16:27, 19 February 2024 by RedirectionBot (talk | contribs) (‎Changed an Item)
scientific article
Language Label Description Also known as
English
Multidimensional integral operators with homogeneous-difference kernels
scientific article

    Statements

    Multidimensional integral operators with homogeneous-difference kernels (English)
    0 references
    0 references
    10 May 2012
    0 references
    The author considers multidimensional integral operators with homogeneous-difference kernels. The usual singular integrals homogeneous of degree \(-n\) are well understood, but for some differential equations arising in mechanics, the author wishes to consider a more general class defined as follows. \noindent In the space \(L_p(\mathbb R^{n + m}), 1 \leq p \leq \infty\), the operator is \[ (A \phi) (x, t) = \int_{\mathbb R^n} \int_{\mathbb R^m} k(x, y, t - s) \phi(y,s)\, dy\, ds, \] \noindent where \(x, y \in \mathbb R^n, t \in \mathbb R^m\), and the kernel \(k\), defined on \(\mathbb R^n \times \mathbb R^n \times \mathbb R^m\), satisfies three conditions. \noindent (1) It is homogeneous of degree \(-n\) with respect to the variables \(x\) and \(y\): \[ k(\alpha x, \alpha y, t) = \alpha^{-n} k(s, y, t), \;\;\;\;\;\text{for all}\;\alpha > 0. \] \noindent (2) It is invariant under the rotation group \(SO(n)\) with respect to the variables \(x\) and \(y\): \[ k(\omega(x), \omega(y), t) = k(x, y, t), \;\;\;\;\;\text{for all}\;\omega \in SO(n). \] \noindent (3) It is integrable with respect to the \(y\) and \(t\) variables: \[ \gamma := \int_{\mathbb R^n} \int_{\mathbb R^M} |k(e_1, y, t) | |y|^{-n/p}\, dy\, dt < \infty, \;\;e_1 = (1, 0, \ldots, 0). \] \noindent The author shows that the operator \(A\) is bounded on \(L_p(\mathbb R^{n +m})\) with the estimate \[ || A \phi||_p \leq \gamma || \phi ||_p. \] \noindent He introduces an appropriate notion of symbols (requiring the use of spherical harmonics) and shows that \(\lambda I - A\) is invertible on \(L_p \) if and only if these symbols do not vanish on an appropriate set.
    0 references
    singular integral operator
    0 references
    boundedness
    0 references
    resolvent
    0 references

    Identifiers