Decrease in the solutions of the first mixed problem for the wave equation in domains with infinite boundaries (Q1594058)

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 1557395
Language Label Description Also known as
default for all languages
No label defined
    English
    Decrease in the solutions of the first mixed problem for the wave equation in domains with infinite boundaries
    scientific article; zbMATH DE number 1557395

      Statements

      Decrease in the solutions of the first mixed problem for the wave equation in domains with infinite boundaries (English)
      0 references
      28 January 2001
      0 references
      Let \(\Omega \subset \mathbb{R}^n\) \((n\geq 2)\) be an unbounded domain with a smooth boundary \(\Gamma\) and the closure of \(\Omega\) does not contain the origin and \(Q= \mathbb{R}_+\times \Omega\). Consider the following problem: \[ u_{tt}-\Delta u =0,\quad (t,x)\in Q, \tag{1} \] \[ u(0,x)=f(x), u_t(0,x)=g(x), \tag{2} \] \[ u|_{\Gamma}=0. \tag{3} \] Let \(2\leq m_1<m\leq n-1\), \(0<\alpha_1\leq\dots\leq\alpha_{m_1-1}<\alpha_{m_1}= \alpha_m<\alpha_{m+1}=\dots+\alpha_n=1\), \(\phi(x)=\alpha_1x^2_1+\dots+ \alpha_nx^2_n\). A domain \(\Omega\) is called of class \(D_{\varphi}\) if \((\nu,\nabla \varphi(x))\leq 0, x\in \Gamma\), where \(\nu\) is the outward unit vector normal to \(\Gamma\). A domain \(\Omega\) is called of class \(G_{\varphi}\) if \(\Omega \in D_{\varphi}\) and there exists \(C_1>0\) and \(C_2>0\) such that \(x_1^2+\dots+x_m^2\leq C_1|x|^{2\alpha_m}\), \(x_1^2+\dots+x_{m_1-1}\leq C_2|x|^{2\alpha_{m_1-1}}\), \(x\in \Omega\). Denote by \(\Pi_{\varphi,t_0}=\{(t,x):t-t_0=(\varphi)^{1/2}\}\), \(\nabla_{\Pi}u=\nabla u- N_{\varphi}\partial u/\partial N_{\varphi}\), \(N_{\varphi}=(1,-\nabla \varphi/2(\varphi)^{1/2}/(1+|\nabla|^2/4\varphi)^{1/2}\), \(E(u)=\int_{\Omega} r^{2\alpha_m+1}(|u|^2+|\nabla u|^2+|\Delta u|^2) dx\), \(r=|x|\). The main result announced by the author is contained in the following theorem. Theorem: Suppose \(n\geq 32\), \[ \frac{1}{2}+\frac{4(4n+(1-\alpha_1)(2(n+4)+C_1(1+2C_1)))}{n^2+2}<\alpha_m<1, \] \[ C_2(1+C_1)<(\alpha_1-1/2)(1-\alpha_m) \] and \(\Omega \in G_{\varphi}\). Then \[ \int_{Q}|\nabla_{\Pi}u|^2\frac{dt dx}{r^{2\alpha_m-1}}\leq C_3(E(f)+ E(\Delta f)+E(g)), \] where \(C_3\) is a constant dependent on \(n, \alpha_1, \alpha_{m_1-1}, \alpha_m, C_1\) and \(C_2\).
      0 references
      0 references
      Helmholtz equation
      0 references
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references