Abstract:In the present paper we consider Neumann Laplacians on singular domains of the type "rooms and passages" or "combs" and we show that, in typical situations, the essential spectrum can be determined from the geometric data. Moreover, given an arbitrary closed subset S of the non-negative reals, we construct domains Q = Q(S) such that the essential spectrum of the Neumann Laplacian on R is just this set S. 'PI 1991 Academnc PEW, Inc
“…The results of [13] indicate that such domains might not have discrete spectrum. The exclusion of these domains from our consideration does not seem to be an essential limitation because lip domains with shapes significantly different from the square are not likely to have multiple second eigenvalues.…”
Section: Ii) ("Hot Spots Conjecture" ) For Every Lip Domain Every Nementioning
confidence: 92%
“…However, by pathwise uniqueness of solutions to (13), under P , Z(u, T, ·) and Z(u, 0, ·) are equal in law, and therefore so are U (u, T, ·) and U (u, 0, ·). Consequently, g(r) = Ef (U (r, 0, t)) = P t (r, f ).…”
Section: We Will Use the Following Standard Fact If φ(H ·) Is Indepmentioning
“…The results of [13] indicate that such domains might not have discrete spectrum. The exclusion of these domains from our consideration does not seem to be an essential limitation because lip domains with shapes significantly different from the square are not likely to have multiple second eigenvalues.…”
Section: Ii) ("Hot Spots Conjecture" ) For Every Lip Domain Every Nementioning
confidence: 92%
“…However, by pathwise uniqueness of solutions to (13), under P , Z(u, T, ·) and Z(u, 0, ·) are equal in law, and therefore so are U (u, T, ·) and U (u, 0, ·). Consequently, g(r) = Ef (U (r, 0, t)) = P t (r, f ).…”
Section: We Will Use the Following Standard Fact If φ(H ·) Is Indepmentioning
Abstract.We obtain the first term in the asymptotic expansion of the eigenvalues of the Laplace operator in a typical dumbbell domain in E2 . This domain consists of two disjoint domains Í2L, iV* joined by a channel Re of height of the order of the parameter e . When an eigenvalue approaches an eigenvalue of the Laplacian in ClL uClR , the order of convergence is £ , while if the eigenvalue approaches an eigenvalue which comes from the channel, the order is weaker: e| lne| . We also obtain estimates on the behavior of the eigenfunctions.
“…This in turn only occurs if there are boundary points where the Wiener series (for the boundary) converges [55,28]. For the Neumann case the situation is more complicated [34,26,33]. In particular, there are domains with arbitrary closed continuous Neumann spectrum [26].…”
mentioning
confidence: 99%
“…For the Neumann case the situation is more complicated [34,26,33]. In particular, there are domains with arbitrary closed continuous Neumann spectrum [26]. We therefore restrict ourselves in this paper to domains (and, later, manifolds) where conditions (2.1.1) and (2.1.2) are valid.…”
Abstract. We use heat kernels or eigenfunctions of the Laplacian to construct local coordinates on large classes of Euclidean domains and Riemannian manifolds (not necessarily smooth, e.g. with C α metric). These coordinates are bi-Lipschitz on embedded balls of the domain or manifold, with distortion constants that depend only on natural geometric properties of the domain or manifold. The proof of these results relies on estimates, from above and below, for the heat kernel and its gradient, as well as for the eigenfunctions of the Laplacian and their gradient. These estimates hold in the non-smooth category, and are stable with respect to perturbations within this category. Finally, these coordinate systems are intrinsic and efficiently computable, and are of value in applications.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.