Inverse Schrödinger scattering in three dimensions (Q4040917)
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 194126
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Inverse Schrödinger scattering in three dimensions |
scientific article; zbMATH DE number 194126 |
Statements
5 June 1993
0 references
inverse scattering theory
0 references
scattering amplitudes
0 references
Gel'fand-Levitan's equation
0 references
\({\bar \partial }\)-approach
0 references
Jost function
0 references
S-matrix
0 references
reconstruction
0 references
0.91448456
0 references
0 references
0.9059906
0 references
0.90588814
0 references
Inverse Schrödinger scattering in three dimensions (English)
0 references
This book presents some of the results in the direct and inverse scattering theory for Schrödinger's equation in three dimensions. The first two sections which form half of the book, deal with scattering solutions. Just one inverse problem is discussed in the book: given the scattering amplitude \(A(\theta',\theta,k)\) \(\forall\theta'\), \(\theta \in S^ 2\), \(\forall k>0\), find \(V(x)\), the potential, which satisfies some technical conditions of smoothness and decay. Discussion of the inverse problem in section 2 is based on a Marchenko (M) equation and the related Riemann-Hilbert problem. Assuming that the given function \(A(\theta',\theta,k)\) is a scattering amplitude corresponding to a \(V(x)\) from a certain class the author derives a Marchenko equation, a solution of which allows one to recover \(V(x)\). A number of necessary conditions for \(A(\theta',\theta,k)\) to be a scattering amplitude are given, a characterization of the class of scattering amplitudes due to Ramm and Weaver is given, a functional is given the global minimum of which is attained at a solution to M equation, the Jost function is introduced, stability of the reconstruction of \(V\) towards small perturbations of \(A(\theta',\theta,k)\) is briefly discussed, and other topics are mentioned. Sections 3-5 discuss the regular solution, an analogue of Gel'fand-Levitan's equation (GL), and the \({\bar\partial}\)-approach in a particular case. Sections 6 and 7 discuss the approach to the solution of the inverse scattering problem used by Faddeev and the recent results based on the \({\bar\partial}\)-approach. The book contains a bibliography (which is far from being complete), an index of symbols, a subject index, and bibliographical notes at the end of sections. There are some important points, that the reader should have in mind, of which the following are worth mentioning: 1) The stability study in {\S} 2.7 is conditional: it is done under the assumption that the Jost function exists and it is not known, in general, if this assumption holds. Moreover, the additional assumption (2.100) implies that the perturbed scattering amplitude A differs from the unperturbed one by \(O(| k|^{-1})\) as \(k\to +\infty\). This assumption is not practical: one cannot assume that the error measurements decays as \(O(| k|^{-1})\) as \(k\to \infty\). Under such assumption one can find the potential exactly via the high-frequency Born's inversion. Therefore the material on pp. 81-84 is not very useful. 2) The reader should be aware that many results in the book are conditional: they are valid under assumptions which may or may not be true (e.g., that equation M is uniquely solvable, that the Jost function with required properties exists, etc.; see e.g. Lemma 2.4.6, Theorems 3.2.4, 4.1.3, 4.1.5, etc.). 3) The argument on p. 62 in the proof of Theorem 2.4.4 is not clear: the equation \(G_ 1(\alpha)=-G_ 2(-\alpha)\) does not seem to follow from the assumption. However, the conclusion of Theorem 2.4.4 is correct and is an immediate consequence of the unitarity of the S-matrix: the assumption of Theorem 2.4.4 implies that \(A^*A=0\), so \(A=0\). By the same reason cA, where \(c\neq 1\) is a real constant, is not the scattering amplitude if A is. 4) The inverse scattering problem studied in the book consists of the three problems: uniqueness, reconstruction, and stability. The first problem is easy and was solved long ago. The material presented in the book on the stability problem ({\S} 2.7) does not solve the problem theoretically or practically. The construction procedure suggested by the author is based upon equation (2.13), p. 48. This is a three-dimensional Fredholm integral equation over an infinite region. Its kernel depends on a vector parameter x running through all of \({\mathbb{R}}^ 3\). Therefore one has to solve a continuum of multidimensional integral equations over an infinite region in order to solve the inverse problem by the author's method. A serious additional complicational comes from the fact that it is not known that the operator in these equations is boundedly invertible even in the case of exact data. While in this case the equations are solvable for the special right-hand side, it is not established that the operator has a bounded inverse. There are also complications when there are bound states. The reconstruction method described in section 7 is even less tractable. Thus, the reader should recognize that the study of the 3D inverse scattering problem with the data depending on five variables \(\theta,\theta'\in S^ 2\) and \(k\in {\mathbb{R}}_+=(0,\infty)\), is far from being complete (except in the case of small potentials for which the reconstruction and characterization problems are solved, see also remark 8 below). 5) The proof of the central result, Lemma 2.4.1, p. 64, is incorrect. There it is still an open question if the potential constructed in (2.55) produces the same scattering amplitude with which we started. The proofs of Lemma 2 on pp. 65 and 66 are also incorrect. Among minor points one can mention: 6) On p. 120 line -4, an equation is missing and in formula (6.7) \(x_ 0\) is not defined. 7) On p. 31 in the last formula in the proof of Lemma 1.2.1 a factor is missing. 8) The characterization result of Ramm and Weaver on p. 62 is misstated: the last statement should read: ``Furthermore, there is at most one solution of (2.13) which satisfies (B) and the coefficient A in (1.2) is identical with the given function \(A(\theta ',\theta,k)''\). This has been proved by the reviewer [``Characterization of the scattering data in multidimensional inverse scattering problems'', in the book ``Inverse problems: An interdisciplinary study'' (1987), 153--167, ed. \textit{P. Sabatier}]. A characterization of the fixed-energy scattering data has been given by the reviewer [J. Math. Anal. Appl. 134, No. 1, 211--253 (1988; Zbl 0632.35076)]. An algorithmically verifiable characterization of the class of scattering amplitudes for small potentials is given by the reviewer [Appl. Math. Lett. 3, 61--65 (1990; Zbl 0711.35148)]. 9) Lemma 2.1.1 is attributed to [Fa 56]. In fact the result belongs to Šnol (1954) [see e.g., \textit{Yu. M. Berezanskii}, Transl., II. Ser., Am. Math. Soc. 35, 167--235 (1964); translation from Tr. Mosk. Mat. O.-va 7, 3--62 (1958; Zbl 0125.33501)]. This book covers material a good part of which was not available in book form. It will be useful for those who are interested in inverse scattering.
0 references