On a nonlocal non-linear inverse scattering problem
Saumyajit Das, Susovan Pramanik · 2026
Read On a nonlocal non-linear inverse scattering problem by Saumyajit Das, Susovan Pramanik (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-22592.txt.
Introduction
Das and Pramanik study an inverse scattering problem in three dimensions. The setting is the nonlinear fractional Helmholtz equation with cubic nonlinearity. An inverse problem works backward from observed effects to a hidden cause. Here the goal is to recover a compactly supported potential from the scattering amplitude. That is, they ask what the wave pattern far away reveals about the hidden term. The work sets up this nonlocal problem and examines the recovery task at its core. It adds to the theory of inverse problems for fractional equations.
Catalog record
| CREATOR | Saumyajit Das, Susovan Pramanik |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31515 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.22592
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On a nonlocal non-linear inverse scattering problem Saumyajit Das, Susovan Pramanik Abstract . In this article, we study the inverse scattering problem for the nonlinear fractional Helmholtz equation with cubic nonlinearity in three dimensions, where we recover a compactly supported potential from scattering amplitude. 1. Introduction The inverse scattering problem concerns the recovery of unknown potentials or media from scattering amplitude measurements, with applications in quantum optics for modeling nonlocal photon-atom interactions and geophysical electromagnetics for anomalous responses in self-similar formations. Physically, these arise in wave propagation through heterogeneous media exhibiting nonlocal and nonlinear effects, such as fractional elasticity models. For classical background, we refer readers to [ 3 , 16 , 17 ] . Beyond physics, inverse scattering exhibits mathematical richness. For detailed study, we refer to [ 6 , 2 , 4 , 15 , 25 ] . In recent years, studies have examined the nonlinear Helmholtz equation in the scattering of electromagnetic waves through localized nonlinear Kerr media [ 1 , 7 , 27 ] . Here, the nonlinear Helmholtz equation arises from reducing Maxwell’s equations for a linearly polarized electric field after eliminating the magnetic field. In [ 27 ] , the following equation emerges: -\Delta u-k^{2}u=\rho 1_{\Omega}|u|^{2}u\quad\text{in }\mathbb{R}^{n}, where \Omega is the support of the nonlinear Kerr medium and \rho is the Kerr constant. Motivated by the above, in this article, we consider the fractional Helmholtz equation with potential and cubic nonlinearity ( 1.1 ). In particular, when s=1 , it reduces to the Helmholtz equation, which frequently arises in scattering theory and quantum mechanics [ 22 , 3 ] ; the fractional Helmholtz equation also appears in geophysical electromagnetics [ 26 , 9 ] and quantum optics to describe the nonlocal characteristics of photon-atom interactions [ 11 , 13 ] . Furthermore, when k=1 and s=1 , the equation ( 1.1 ) becomes Ginzburg-Landau equation (for details see [ 10 , 8 ] ). In this article, for k=1 , our model ( 1.1 ) exactly becomes fractional Ginzburg-Landau equation. Motivated by its physical relevance across diverse phenomena in physics, we study the inverse scattering problem for the fractional Helmholtz equation to recover the compactly supported potential from the measurement of scattering amplitude. In particular we study the following Helmholtz equation: for j=1,2 (1.1) \displaystyle(-\Delta)^{s}u_{j}-k^{2s}u_{j}=Q_{j}(x)|u_{j}|^{2}u_{j},\quad\mbox{in}\ \mathbb{R}^{d}, associated with the following Sommerfield radiation condition: (1.2) \displaystyle\lim\limits_{R\to\infty}\frac{1}{R}\int_{1\leq\|x\|\leq R}|D^{s}_{x}u_{j}^{sc}-k^{2s}u_{j}^{sc}\hat{x}|^{2}\,{\rm{d}}x=0, where \hat{x}=\frac{x}{\|x\|} , D_{x}^{s}:=\mathcal{F}^{-1}|\xi|^{s}\mathcal{F} and u_{i}^{sc} is the scattering part of the solution u . For j=1,2 , the potential Q_{i}(x)\in\mathrm{L}^{\infty}(\mathbb{R}^{d}) is assumed to be compactly supported in \Omega\subset{\mathbb{R}}^{d} and s\in(\frac{d}{d+1},1) . The solution u_{j} can be expressed as: for j=1,2 (1.3) \displaystyle u_{j}=u_{j}^{\rm{in}}+u_{j}^{\rm{sc}}, where u_{j}^{\rm{in}} satisfy the initial wave equation (1.4) \displaystyle(-\Delta)^{s}u_{j}^{\rm{in}}-k^{2s}u_{j}^{\rm{in}}=0,\quad\mbox{in}\ {\mathbb{R}}^{d}. given by the Herglotz wave function u_{j}^{\rm{in}}=\int_{\mathbb{S}^{d-1}}e^{ikx\cdot\theta}g(\theta){\rm{d}}\theta, for any function g\in\mathrm{L}^{2}(\mathrm{S}^{d-1}) , whereas u_{j}^{\rm{sc}} satisfy the scattering part of the solution, given by (1.5) \displaystyle(-\Delta)^{s}u_{j}^{\rm{sc}}-k^{2s}u_{j}^{\rm{sc}}=Q_{j}|u_{j}|^{2}u_{j},\qquad\mbox{in}\ \mathbb{R}^{d}. We would like to note that the initial wave solutions are the same for both the equations, i.e., u_{1}^{\rm{in}}=u_{2}^{\rm{in}} . Thanks to the Stein-Tomas theorem [ 24 , 23 ] , we have the following estimate (1.6) \displaystyle\|u_{i}^{\rm{in}}\|_{\mathrm{L}^{2\frac{d+1}{d-1}}({\mathbb{R}}^{d})}\leq C_{ST}\|g\|_{\mathrm{L}^{2}(\mathbb{S}^{d-1})},\quad\mbox{for}\ j=1,2, where u_{i}^{\rm{in}} is given by ( 1.4 ) and C_{ST} is a positive constant depending on the dimension. The following existence result we borrow from [ 22 , Theorem 1.3] : Theorem 1.1 ( [ 22 ] ) . Let u satisfies (1.7) \displaystyle(-\Delta)^{s}u-k^{2s}u=Q(x)|u|^{t-1}u,\qquad\mbox{in}\ {\mathbb{R}}^{d}. where k>0 and Q(x)\in\mathrm{L}^{\infty}({\mathbb{R}}^{d}) is compactly supported. Let the initial wave function is given by u^{\rm{in}}=\int_{\mathbb{S}^{d-1}}e^{ikx\cdot\theta}g(\theta)\,{\rm{d}}\theta, where \|g\|_{\mathrm{L}^{2}({\mathbb{R}}^{d})}\leq\epsilon for some \epsilon>0 . Furthermore, assume s\in\left(\frac{d}{d+1},1\right)\ \text{and}\ \frac{d-1+4s}{d-1}<t<\frac{d+1}{d+1-4s}. Then, there exists a unique solution ‘ u ’ to the equation ( 1.7 ), satisfying u\in\mathrm{L}^{q}({\mathbb{R}}^{d}),\ \text{such that}\ \|u\|_{\mathrm{L}^{q}({\mathbb{R}}^{d})}\leq a, provided u satisfy the following Sommerfield radiation condition: \displaystyle\lim\limits_{R\to\infty}\frac{1}{R}\int_{1\leq\|x\|\leq R}|D^{s}_{x}u^{sc}-k^{2s}u^{sc}\hat{x}|^{2}\,{\rm{d}}x=0. Here a>0 some positive constant depeding on \epsilon and the exponent ‘ q ’ satisfies \frac{d(t-1)}{2s}<q<\frac{2dt}{d+1}. In particular, for the spatial dimension d=3 and t=3 , we obtain that u\in\mathrm{L}^{4}({\mathbb{R}}^{d}). Remark 1.2 . The uniqueness follows from the Sommerfeld radiation condition, formulated through the local estimate in [ 22 , Theorem 1.2] . Additionally, when s=1 , k=1 , ( 1.1 ) reduces to the Ginzburg–Landau equation studied in [ 10 ] . Our nonlinear term in ( 1.1 ) derives from the Ginzburg–Landau equation. Building on the work of Gutiérrez in [ 10 ] , the authors of [ 22 ] study its fractional version. To state our result, let us define the scattering amplitude (or far-field pattern ) by (see Section 2.2 for details) (1.8) u^{\infty}(\hat{x})=\int_{\mathbb{R}^{d}}e^{-\mathsf{i}k\hat{x}\cdot y}Q(y)|u(y)|^{2}u(y)\,dy,\quad\hat{x}\in\mathbb{S}^{d-1}. Now the question is: can one recover the potential from a priori knowledge of the scattering amplitude u^{\infty} ? Here we solve this problem in dimension d=3 . In particular, we have following result: Theorem 1.3 . Let s\in\left(\frac{4}{5},\frac{3}{2}\right) and k>k_{0}>0 . For j=1,2 , let Q_{j}\in\mathrm{L}^{\infty}(\Omega) and compactly supported. Let u_{j} satisfies: for j=1,2 (-\Delta)^{s}u_{j}-k^{2s}u_{j}=Q_{j}|u_{j}|^{2}u_{j}\quad\text{ in }\mathbb{R}^{3}. The solution u_{j} is given by u_{j}=u_{j}^{\rm{in}}+u_{j}^{\rm{sc}}, where u_{j}^{\rm{in}} is the incident wave and u_{j}^{\rm{sc}} is the scattering wave as defined in ( 1.4 ) and ( 1.5 ). Let the scattering amplitude corresponding to the scattering waves are the same i.e., for all (k,\hat{x},\theta)\in(k_{0},+\infty)\times\mathbb{S}^{2}\times\mathbb{S}^{2},\,k_{0}>0 , (1.9) u^{\infty}_{1}(k,\hat{x},\theta)=u^{\infty}_{2}(k,\hat{x},\theta), where the scattering amplitude as defined in ( 1.8 ). Then Q_{1}=Q_{2}. Remark 1.4 . Due to the lack of resolvent estimates for all fractional exponents s (see Proposition 2.3 ), and the restriction arising from the Sobolev embedding (see Lemma 3.3 ), we restrict our analysis to s\in\left(\frac{4}{5},\frac{3}{2}\right) . For other values of the fractional exponent s , the problem remains open. Remark 1.5 . Note that here we vary all three variables (k,\hat{x},\theta)\in(k_{0},\infty)\times\mathbb{S}^{d-1}\times\mathbb{S}^{d-1} in the scattering amplitude to determine the potential Q . This corresponds to data of dimension 2d-1 . A more challenging question arises when k is fixed and only (\hat{x},\theta)\in\mathbb{S}^{d-1}\times\mathbb{S}^{d-1} are varied: can the potential Q still be uniquely recovered? In this case, the dimension of the data is 2d-2 . This problem is related to the Ginzburg-Landau equation when k=1 is fixed. In the local case, i.e., when s=1 , this problem has been studied in [ 8 ] . To the best of our knowledge, the corresponding scattering problem for the fractional Ginzburg-Landau equation remains open, which corresponds to the case k=1 in ( 1.1 ). We state the open problem below. Open problem: Ginzburg-Landau equation \bullet For j=1,2 , let Q_{j}\in\mathrm{L}^{\infty}(\Omega) and compactly supported. Let u_{j} satisfies: for j=1,2 (-\Delta)^{s}u_{j}-u_{j}=Q_{j}|u_{j}|^{2}u_{j}\quad\text{ in }\mathbb{R}^{3}. The solution u_{j} is given by u_{j}=u_{j}^{\rm{in}}+u_{j}^{\rm{sc}}, where u_{j}^{\rm{in}} is the incident wave and u_{j}^{\rm{sc}} is the scattering wave as defined in ( 1.4 ) and ( 1.5 ). Let the scattering amplitude corresponding to the scattering waves are the same i.e., for all (\hat{x},\theta)\in\mathbb{S}^{2}\times\mathbb{S}^{2} , u^{\infty}_{1}(1,\hat{x},\theta)=u^{\infty}_{2}(1,\hat{x},\theta). where the scattering amplitude as defined in ( 1.8 ). Then Q_{1}=Q_{2}. In recent years, the study of inverse scattering has been quite dynamic. Potential recovery using far-field data in the classical case was studied in [ 20 ] . For the fractional case, the authors in [ 5 ] recently studied the inverse scattering problem for the fractional Schrödinger operator. For the fractional Helmholtz equation, we refer to [ 19 ] . For the nonlinear and semilinear cases with local operators, we refer the reader to [ 18 , 8 , 12