Well-posedness of dispersion managed nonlinear Schrödinger equations (Q2685232)

From MaRDI portal





scientific article; zbMATH DE number 7655393
Language Label Description Also known as
default for all languages
No label defined
    English
    Well-posedness of dispersion managed nonlinear Schrödinger equations
    scientific article; zbMATH DE number 7655393

      Statements

      Well-posedness of dispersion managed nonlinear Schrödinger equations (English)
      0 references
      0 references
      0 references
      0 references
      20 February 2023
      0 references
      The authors consider the following dispersion managed nonlinear Schrödinger equation, also known as Gabitov-Turitsyn equation, \[ i\partial_t u +d_{\mathrm{av}}\partial_x^2 u +\int_{\mathbb R}T_r^{-1}\left( P(T_ru)\right)\psi(r)dr,\quad u_{\mid t=0}=u_0,\tag{1} \] where \(x\in \mathbb R\), \(d_{\mathrm{av}}\in \mathbb R\) is a constant which may or may not be zero according to the origin of the model in optics, \(T_r=e^{ir\partial_x^2}\) is the Schrödinger evolution group at time \(r\), \(P(z)=h(|z|)z\) is the nonlinearity, and \(\psi\ge 0\) is a probability density (hence \(\psi\in L^1(\mathbb R)\)), connected to the local periodic dispersion profile in optics. According to various assumptions on the nonlinearity \(h\), the average dispersion \(d_{\mathrm{av}}\) (\(d_{\mathrm{av}}=0\) or \(d_{\mathrm{av}}\not=0\)), and possibly extra integrability conditions on \(\psi\), the authors first show several results regarding the local and the global Cauchy problem (1), for \(u_0\in L^2(\mathbb R)\) or \(u_0\in H^1(\mathbb R)\). The nonlinear term is estimated thanks to (homogeneous) Strichartz estimates, and Hölder inequality. Based on the nonlinear estimates, the Cauchy problem is studied via a fixed point argument based on Duhamel's formula. Globalization is based on the conservation of the mass (\(\|u(t)\|_{L^2}^2\)) and of the energy \begin{align*} E(u(t))&=\frac{d_{\mathrm{av}}}{2}\|\partial_x u(t)\|_{L^2}^2- \iint_{\mathbb R^2} V\left(|T_r(u(t)|\right)dx\psi(r)dr,\\ &\text{where } V(a)=\int_0^aP(s)ds=\int_0^ah(s)sds. \end{align*} In a final section, the authors analyze the set of energy minimizers, \[ S_\lambda^{d_{\mathrm{av}}}=\{f\in X,\quad E(f) = E_\lambda^{d_{\mathrm{av}}},\ \|f\|_{L^2}^2=\lambda\}, \] where \[ E_\lambda^{d_{\mathrm{av}}}=\inf\{ E(f),\ f\in X,\ \|f\|_{L^2}^2=\lambda\}, \] with \(X=L^2\) in the absence of dispersion \(d_{\mathrm{av}}=0\), and \(X=H^1\) otherwise (in which case the signs of \(d_{\mathrm{av}}\) and the nonlinearity \(P\) must be related adequately so the above variational problem is nontrivial). Under yet different assumptions on the nonlinearity and the weight \(\psi\), the authors establish the existence of minimizers (ground states), and their orbital stability under the flow of (1), following the general strategy of Cazenave and Lions.
      0 references
      0 references
      nonlocal NLS
      0 references
      dispersion management
      0 references
      well-posedness
      0 references
      orbital stability
      0 references
      0 references
      0 references

      Identifiers

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