Quantitative unique continuation for operators with partially analytic coefficients. Application to approximate control for waves (Q1737975)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Quantitative unique continuation for operators with partially analytic coefficients. Application to approximate control for waves
scientific article

    Statements

    Quantitative unique continuation for operators with partially analytic coefficients. Application to approximate control for waves (English)
    0 references
    0 references
    0 references
    0 references
    24 April 2019
    0 references
    In this nice article under review, the authors first prove quantitative estimates associated to the unique continuation theorems for operators with partially analytic coefficients of \textit{D. Tataru} [Commun. Partial Differ. Equations 20, No. 5--6, 855--884 (1995; Zbl 0846.35021)]; J. Math. Pures Appl. (9) 78, No. 5, 505--521 (1999; Zbl 0936.35038)], \textit{L. Robbiano} and \textit{C. Zuily} [Invent. Math. 131, No. 3, 493--539 (1998; Zbl 0909.35004)] and \textit{L. Hörmander} [Prog. Nonlinear Differ. Equ. Appl. 32, 179--219 (1997; Zbl 0907.35002)]. The authors provide local stability estimates that can be propagated, leading to global ones. Then, they specify those results to the wave operator on a Riemannian manifold \(M\) with boundary. For this operator, they also prove Carleman estimates and local quantitative unique continuation from and up to the boundary \(\partial M\). This allows them to obtain a global stability estimate from any open subset \(\Gamma\) of \(M\) or \(\partial M\), with the optimal time and dependence on the observation. As a first application, the authors compute a sharp lower estimate of the intensity of waves in the shadow of an obstacle. They also provide the cost of approximate controllability on the compact manifold \(M\): for any \(T > 2 \sup_{x\in M}\operatorname{dist}(x, 0)\), they can drive any \(H^1_0L^2\) data in time \(T\) to an \(\epsilon\)-neighborhood of zero in \(L^2 \times H^{-1}\), with a control located in \(\Gamma\), at cost \(e^{C/\epsilon}\). They finally obtain related results for the Schrödinger equation.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    stability estimates
    0 references
    Carleman estimates
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references