Asymptotic completeness of the wave operators for first order systems with nonsmooth coefficients (Q1910053)

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 861813
Language Label Description Also known as
default for all languages
No label defined
    English
    Asymptotic completeness of the wave operators for first order systems with nonsmooth coefficients
    scientific article; zbMATH DE number 861813

      Statements

      Asymptotic completeness of the wave operators for first order systems with nonsmooth coefficients (English)
      0 references
      0 references
      21 April 1997
      0 references
      The author studies existence and asymptotic completeness of the wave operators for first order systems of the form \[ \partial_t \varphi= E(x) \sum^N_{j=1} \partial_j \bigl(H_jE(x) \varphi \bigr), \quad x\in \Omega\subset \mathbb{R}^N. \tag{1} \] Here \(H_j\), \(j=1, \dots, N\), are symmetric \(M\times M\) matrices and \(E\in L^\infty (\Omega)\). The wave operators can be defined comparing two such problems. Namely, let \[ B^{(0)}_k \varphi= i\sum^N_{j=1} \partial_j (H_j \varphi),\;k=1,2. \tag{2} \] Then \(\varphi \in D(B_k^0)\) if, in addition to (2), it satisfies certain boundary conditions. Let \(\Omega_k\), \(k=1,2\), be the corresponding domains in \(\mathbb{R}^N\) associated with these two operators. Consider the operator \(B_k= E_kB_k^{(0)} E_k\), \(k=1,2\). Then the wave operators are defined formally as follows \[ \Omega^\pm=s\text{-}\lim_{t\to\pm \infty} \exp (it B_2) I_{1,2} \exp (-it B_1) P_{\text{cont}} (B_1). \] Here the natural identification operator \(I_{1,2}\) is defined as follows \(I_{1,2} (\varphi) (x)= \varphi(x)\) if \(x\in \Omega_1 \cap \Omega_2\) and \(I_{1,2} (\varphi) (x)=0\) if \(x\in \Omega_2 \backslash \Omega_1\), \(\varphi \in X_1\). Here \(X_k= L_2 (\Omega_k; \mathbb{C}^M)\), \(k=1,2\). The main result asserts that the wave operators exist and that they are complete. The completeness means that \(\text{ran} \Omega^\pm = X_{\text{cont}} (B_2)\). Here \(X_{\text{cont}} (B_2)\), \(X_{\text{cont}} (B_1)\) are the continuous spaces of \(B_2,B_1\). The main idea of the proof is to obtain the following local energy decay \[ \lim_{t\to\pm\infty} \int_{\Omega_1 \cap\{(x)<vt\}} \bigl(\varphi(t) \bigr)^2 = 0 \] for any \(v\in(0,1)\), \(\varphi \in {\mathcal D} (B_1) \cap X_{\text{cont}} (B_1)\), \(\varphi(t) = e^{-itB_1} \varphi\). This decay estimate is based on a suitable application of the RAGE theorem.
      0 references
      asymptotic completeness
      0 references
      RAGE theorem
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references