Determining a magnetic Schrödinger operator from partial Cauchy data (Q2472450)

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 5237972
Language Label Description Also known as
default for all languages
No label defined
    English
    Determining a magnetic Schrödinger operator from partial Cauchy data
    scientific article; zbMATH DE number 5237972

      Statements

      Determining a magnetic Schrödinger operator from partial Cauchy data (English)
      0 references
      0 references
      0 references
      0 references
      22 February 2008
      0 references
      Suppose that \(n\geq 3\) and \(\Omega\subseteq\mathbb{R}^n\) is a bounded domain with smooth boundary. For a real magnetic potential \(A\in C^2(\overline{\Omega},\mathbb{R}^n)\) and a bounded electric potential \(q\in L^\infty(\Omega)\) consider the magnetic Schrödinger operator \[ \mathcal{L}_{A,q}(x,D):=\sum_{j=1}^n(-\text{i}\partial_j+A_j(x))^2+q(x). \] It is assumed that \(0\) is not an eigenvalue of \(\mathcal{L}_{A,q}\colon H^2(\Omega)\cap H^1_0(\Omega)\to L^2(\Omega)\). If \(\nu\) is the exterior unit normal to \(\partial\Omega\) then let us define the Dirichlet to Neumann map (DN map) \[ \mathcal{N}_{A,q}\colon H^{1/2}(\partial\Omega)\to H^{-1/2}(\partial\Omega) \] by \[ \mathcal{N}_{A,q}(f) :=((\partial_\nu+\text{i}A\cdot\nu) \mathcal{L}_{A,q}^{-1}f)| _{\partial\Omega}, \] where \(\mathcal{L}_{A,q}^{-1}\colon H^{1/2}(\partial\Omega)\to H^1(\Omega)\) is the Dirichlet inverse of \(\mathcal{L}_{A,q}\). If \(x_0\in\mathbb{R}^n\backslash\overline{\text{conv}(\Omega)}\) then \[ F(x_0):=\{x\in\partial\Omega\mid(x-x_0)\cdot\nu(x)\leq0\} \] is the front side of \(\partial\Omega\) with respect to \(x_0\). The main result is the following: Suppose that \(\Omega\) is simply connected, \(A_1,A_2\in C^2(\overline{\Omega},\mathbb{R}^n)\) are two real magnetic potentials, \(q_1,q_2\in L^\infty(\Omega)\) are two electric potentials, and \(\mathcal{L}_{A_k,q_k}\) is invertible in the sense mentioned above, for \(k=1,2\). If the images of \(\mathcal{N}_{A_1,q_1}\) and \(\mathcal{N}_{A_2,q_2}\) coincide when restricted to a neighborhood of \(F(x_0)\) for some \(x_0\) as above, then \(A_1\) and \(A_2\) differ only by a gradient and \(q_1=q_2\). In other words, if \(n=3\) the magnetic field and electric potential can be recovered from the values of the images of the DN map on one side of the boundary of the domain. In particular, if \(\Omega\) is strongly starshaped with respect to a point \(x_0\in\partial\Omega\) then the images of the DN map, restricted to a neighborhood of \(x_0\), determine the magnetic field and the electric potential. This work complements the results in [Ann. Math. (2) 165, No. 2, 567--591 (2007; Zbl 1127.35079)].
      0 references
      inverse problem
      0 references
      Dirichlet to Neumann map
      0 references
      magnetic Schrödinger operator
      0 references
      complex geometrical optics solution
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references