pith. sign in

arxiv: 2606.05977 · v1 · pith:YGOKQ2ZXnew · submitted 2026-06-04 · 🧮 math.NA · cs.NA

Direct reconstruction for acoustic inverse Born scattering

Nuutti Hyv\"onen , Lisa Sch\"atzle This is my paper

Pith reviewed 2026-06-28 00:19 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords inverse scatteringBorn approximationZernike polynomialsfar-field operatorHelmholtz equationdirect reconstructionacoustic imagingtriangular decomposition
0
0 comments X

The pith

Choosing Zernike expansion functions decouples the inverse Born scattering problem into independent triangular systems solvable by forward substitution.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes an explicit reconstruction method for recovering a compactly supported scatterer from far-field data in the acoustic inverse scattering problem under the Born approximation. By expanding both the contrast and the far-field operator in a Zernike basis chosen to make the system matrix triangular, the coefficients of the contrast can be found directly without solving a coupled linear system. This approach overcomes the nonlinearity and ill-posedness in the weak scattering regime by turning the problem into a sequence of independent forward substitutions for each angular frequency. A sympathetic reader would care because it provides a fast, direct alternative to iterative methods for linearized inverse problems, with potential extension to regularized nonlinear cases as shown in numerics.

Core claim

The central claim is that an appropriate choice of expansion functions in the Zernike basis allows the system matrix relating the far-field data to the contrast to decouple into separate infinite triangular systems for each spatial angular frequency. These systems can then be solved independently by forward substitutions to obtain the expansion coefficients of the contrast directly from those of the far-field data.

What carries the argument

Triangular Zernike decompositions of the far-field operator that decouple the inverse problem into independent angular-frequency subsystems.

If this is right

  • The reconstruction expresses the contrast expansion coefficients explicitly in terms of the far-field data coefficients.
  • Each angular frequency subsystem for the contrast is solved independently without coupling to other frequencies.
  • The method can be applied to full nonlinear far-field data when paired with an adequate regularization technique.
  • No iterative solver is required for the linearized Born problem once the basis expansions are fixed.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The independent subsystems could be computed in parallel for different angular frequencies to improve efficiency on large problems.
  • The same basis choice may transfer to other linearized inverse problems that admit similar operator representations, such as certain tomography settings.
  • Stability under data noise could be analyzed by bounding the growth of entries during the forward substitutions in each triangular block.

Load-bearing premise

The far-field operator admits a representation in the chosen Zernike basis such that the system matrix is triangular and decoupled by angular frequency.

What would settle it

A direct computation of the system matrix entries for the selected Zernike expansions on a known contrast, checking whether the matrix is exactly triangular with zeros above the diagonal blocks and whether forward substitution recovers the exact coefficients.

Figures

Figures reproduced from arXiv: 2606.05977 by Lisa Sch\"atzle, Nuutti Hyv\"onen.

Figure 2.1
Figure 2.1. Figure 2.1: The natural non-uniform sampling pattern (blue crosses) for the Fourier data of the contrast q in the definition (2.6) of the Born far field pattern u∞ B for 20 equiangular illumination directions d (red crosses) and observation directions xb. the basic theory on Hilbert–Schmidt integral operators yields (see, e.g., [29]) kTBqk 2 HS(L2(S1)) ≤ κ 4 Z S1 Z S1 [PITH_FULL_IMAGE:figures/full_fig_p005_2_1.png] view at source ↗
Figure 3.1
Figure 3.1. Figure 3.1: The real parts of the basis functions (Ψj,k)j∈{0,...,2N},k∈{0,...,N−⌈j/2⌉} for κR = 2.5 and N = 3. First row: Ψ0,0, Ψ0,1, Ψ0,2 , Ψ0,3. Second row: Ψ1,0, Ψ1,1, Ψ1,2 , Ψ2,0. Third row: Ψ2,1, Ψ2,2, Ψ3,0 , Ψ3,1. Fourth row: Ψ4,0, Ψ4,1, Ψ5,0 , Ψ6,0. The color scale is the same in all subfigures. Solving the resulting finite-dimensional systems F j,N c j,N = a j,N , j ∈ {−2N, . . . , 2N}, (4.2) in place of the… view at source ↗
Figure 3.2
Figure 3.2. Figure 3.2: The real parts of the basis functions (Ψj,k)j∈{0,...,2N},k∈{0,...,N−⌈j/2⌉} for κR = 10 and N = 3. First row: Ψ0,0, Ψ0,1, Ψ0,2 , Ψ0,3. Second row: Ψ1,0, Ψ1,1, Ψ1,2 , Ψ2,0. Third row: Ψ2,1, Ψ2,2, Ψ3,0 , Ψ3,1. Fourth row: Ψ4,0, Ψ4,1, Ψ5,0 , Ψ6,0. The color scale is the same in all subfigures. as elements of a matrix, we obtain Qj,N :=   R j 0 (r0) R j 1 (r0) · · · R j N−⌈j/2⌉ (r0) R j 0 (r1) R j 1 (r1… view at source ↗
Figure 4.1
Figure 4.1. Figure 4.1: Left: Expansion coefficients (am,n)|m|,|n|≤N that are taken into account in the in￾version (cf. (3.21)) after introducing the truncation index N (blue), with a representative diagonal (am,m−j )|m|,|m−j|≤N corresponding to the angular index j = −2 highlighted (red). Right: Orthonormal￾ization error (4.3) in the Gram–Schmidt process for different values of κR as functions of the truncation index N. {5, 10,… view at source ↗
Figure 5.1
Figure 5.1. Figure 5.1: Exact contrast in Example 5.1 (left) and in Example 5.2 (right). we compare the performance of our method to the recently introduced low-rank method for solving the inverse Born scattering problem [30] and with the MATLAB’s built-in NUFFT, as it is described in [12]. We assume the ability to sample the the Born far field data at 2L ∈ N equiangular illumi￾nation and measurement directions. That is, we ass… view at source ↗
Figure 5.2
Figure 5.2. Figure 5.2: Example 5.1. Reconstructed contrast from the exact Born far field data for different truncation indices. Top left: N = 5 (too small). Top right: N = 10 (too small). Middle left: N = 15 (too small). Middle right: N = 20 (slightly too small). Bottom left: N = 29 (optimal). Bottom right: N = 31 (too large). for m, n = 1, . . . , 2L. We choose L = 125 and truncate the series at 250 terms. For expanding the c… view at source ↗
Figure 5.3
Figure 5.3. Figure 5.3: Example 5.1. Left: Relative L 2 reconstruction error εrel as a function of the truncation index N with exact Born far field data. The optimal choice N = 29 is marked by a vertical line. Right: The best and worst relative L 2 reconstruction errors εrel as functions of the absolute noise level p over 20 runs with different realizations of noise [PITH_FULL_IMAGE:figures/full_fig_p017_5_3.png] view at source ↗
Figure 5.4
Figure 5.4. Figure 5.4: Example 5.1. Worst reconstructed contrasts in the sense of relative L 2 error from noisy Born far field data over 20 runs at the noise levels p = 20 (left) and p = 80 (right). information, whereas choosing it too large results in dominance of the orthonormalization error introduced by the Gram–Schmidt process. Interestingly, the region in which the contrast is re￾constructed accurately expands gradually … view at source ↗
Figure 5.5
Figure 5.5. Figure 5.5: Example 5.2. Relative L 2 reconstruction error εrel as a function of the wave number κ with the exact Born far field data and the exact full far field data as the inputs for the proposed method. The relative Frobenius-based approximation error induced by replacing the Born far field data with the full far field data is also shown for reference (cf. (5.2)). reasonably even when multiple scattering effects… view at source ↗
Figure 5.6
Figure 5.6. Figure 5.6: Example 5.2. Reconstructed contrast from the exact Born far field data (left) and the exact full far field data (right). Top: κ = 11 (the best reconstruction from the full far field data, cf [PITH_FULL_IMAGE:figures/full_fig_p020_5_6.png] view at source ↗
Figure 5.7
Figure 5.7. Figure 5.7: Example 5.3. Real part (left) and imaginary part (right) of the exact contrast function. field data to obtain reconstructions ranging in quality from good to reasonable depending on the extent of multiple scattering effects. According to our numerical experiments, the proposed method compares favorably to other algorithms designed for solving the inverse medium Born scattering problem, in particular, pro… view at source ↗
Figure 5.8
Figure 5.8. Figure 5.8: Example 5.3. Real part (left) and imaginary part (right) of reconstructed contrast function for wave number κ = 30 from full far field data with 20% of additive noise. Top: our proposed method. Middle: the low-rank method [30]. Bottom: MATLAB’s built-in NUFFT. 24 [PITH_FULL_IMAGE:figures/full_fig_p024_5_8.png] view at source ↗
Figure 5.9
Figure 5.9. Figure 5.9: Example 5.3. Real part (left) and imaginary part (right) of reconstructed contrast function for wave number κ = 60 from full far field data with 20% additive noise. Top: our proposed method. Middle: the low-rank method [30]. Bottom: MATLAB’s built-in NUFFT. 25 [PITH_FULL_IMAGE:figures/full_fig_p025_5_9.png] view at source ↗
read the original abstract

We consider the inverse medium scattering problem for the Helmholtz equation in two dimensions, i.e., the task to recover a compactly supported penetrable two-dimensional scatterer from full knowledge of the associated far field data or, equivalently, the far field operator. Although this problem is uniquely solvable, it is severely ill-posed and nonlinear. In the regime of weak scattering, the Born approximation yields a linearized relation between the contrast and the far field data, thus overcoming the second difficulty. This linear setting allows to build on recent work on linearized electrical impedance tomography, which relies on triangular Zernike decompositions, to derive an explicit reconstruction formula that expresses the expansion coefficients of the contrast in terms of those of the far field data. By choosing the expansion functions appropriately, the resulting system matrix decouples into separate (infinite) triangular systems for the spatial angular frequencies in the contrast. Consequently, each of these systems can be solved independently by performing forward substitutions. Our numerical experiments indicate that this approach, combined with an adequate regularization method, remains effective even when applied to full nonlinear far field data beyond the Born regime.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript derives an explicit reconstruction formula for the contrast in the 2D acoustic inverse scattering problem under the Born approximation. Building on triangular Zernike decompositions from prior EIT work, it expands the contrast and far-field data in Zernike polynomials and asserts that an appropriate choice of basis functions decouples the linear system into independent infinite triangular systems indexed by angular frequency m, each solvable by forward substitution. Numerical experiments are reported to indicate that the approach, with regularization, remains effective when applied to full nonlinear far-field data.

Significance. If the claimed triangular decoupling holds for the acoustic Born operator, the work would supply a direct, non-iterative reconstruction procedure that is computationally attractive and extends the EIT technique to a new setting. The explicit formula and the numerical indication of utility beyond the linear regime are positive features. The absence of an error analysis or explicit matrix construction, however, limits the immediate strength of the contribution.

major comments (2)
  1. [Abstract and derivation section] Abstract (paragraph on triangular Zernike decompositions) and the derivation of the reconstruction formula: the central claim that the system matrix is block-diagonal in angular frequency m and strictly triangular in the radial index n rests on the assertion that the far-field operator admits a Zernike representation with the required vanishing properties. The acoustic Born kernel is the Fourier transform of the contrast sampled on the Ewald circle and involves phase factors e^{ik(d-d')·x}; unlike the EIT linearized operator, no explicit identity or ordering argument is supplied showing that the radial integrals ∫ R_n^m(r) J_m(kr) r dr vanish for n' > n (or the reverse). This property is load-bearing for the forward-substitution procedure and must be verified by direct computation of the matrix elements.
  2. [Numerical experiments] Numerical experiments section: the claim that the method 'remains effective even when applied to full nonlinear far field data' is supported only by qualitative statements; no quantitative error tables, condition-number estimates for the triangular systems, or verification that the computed matrix is indeed triangular are provided. This weakens the supporting evidence for the practical utility asserted in the abstract.
minor comments (2)
  1. The manuscript would benefit from an explicit statement of the ordering chosen for the Zernike indices (n,m) that produces the triangular structure.
  2. A brief comparison table or figure showing the reconstructed contrast for both Born and nonlinear data would improve clarity of the numerical results.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive feedback on our manuscript. We address each major comment below and will revise the paper accordingly to strengthen the presentation.

read point-by-point responses

  1. Referee: [Abstract and derivation section] Abstract (paragraph on triangular Zernike decompositions) and the derivation of the reconstruction formula: the central claim that the system matrix is block-diagonal in angular frequency m and strictly triangular in the radial index n rests on the assertion that the far-field operator admits a Zernike representation with the required vanishing properties. The acoustic Born kernel is the Fourier transform of the contrast sampled on the Ewald circle and involves phase factors e^{ik(d-d')·x}; unlike the EIT linearized operator, no explicit identity or ordering argument is supplied showing that the radial integrals ∫ R_n^m(r) J_m(kr) r dr vanish for n' > n (or the reverse). This property is load-bearing for the forward-substitution procedure and must be verified by direct computation of the matrix elements.

    Authors: We agree that the triangular structure must be verified explicitly for the acoustic Born operator rather than relying solely on analogy with the EIT case. In the revised manuscript we will add a direct computation (in an appendix or dedicated subsection) of the relevant radial integrals involving the product of Zernike radial functions and the Bessel functions arising from the Fourier transform on the Ewald circle. This calculation will confirm the vanishing property for n' > n and thereby justify the block-diagonal and strictly triangular form of the system matrix. revision: yes

  2. Referee: [Numerical experiments] Numerical experiments section: the claim that the method 'remains effective even when applied to full nonlinear far field data' is supported only by qualitative statements; no quantitative error tables, condition-number estimates for the triangular systems, or verification that the computed matrix is indeed triangular are provided. This weakens the supporting evidence for the practical utility asserted in the abstract.

    Authors: We accept the referee's observation that the numerical evidence is currently qualitative. The revised numerical experiments section will include quantitative L² reconstruction error tables for several test contrasts, reported condition numbers of the triangular subsystems, and explicit numerical checks (e.g., matrix-entry norms or sparsity patterns) confirming that the assembled matrices are indeed triangular to machine precision. These additions will provide stronger support for the claimed practical utility beyond the Born regime. revision: yes

Circularity Check

0 steps flagged

No significant circularity; extends EIT triangularity independently to acoustic Born case

full rationale

The derivation builds on cited prior linearized EIT work for the triangular Zernike property but produces a new explicit reconstruction formula specific to the acoustic far-field operator under Born approximation. The decoupling into angular-frequency blocks is asserted to follow from basis choice in the acoustic kernel, without any reduction of the target formula to a self-referential definition, fitted parameter renamed as prediction, or load-bearing self-citation chain. The result remains independently verifiable against the acoustic integral operator and is not equivalent to its inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The claim rests on standard properties of Zernike polynomials permitting triangular structure and on the domain assumption that the Born approximation linearizes the scattering map; no free parameters or invented entities are introduced in the abstract.

axioms (2)
  • standard math Zernike polynomials admit a triangular decomposition of the relevant far-field operator
    Invoked when stating that appropriate choice of expansion functions yields decoupled triangular systems
  • domain assumption Born approximation yields a linearized relation between contrast and far-field data
    Explicitly used to overcome nonlinearity of the inverse problem

pith-pipeline@v0.9.1-grok · 5747 in / 1231 out tokens · 43608 ms · 2026-06-28T00:19:30.332909+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

32 extracted references · 26 canonical work pages

  1. [1]

    Audibert and H

    L. Audibert and H. Haddar. A generalized formulation of t he linear sampling method with exact characterization of targets in terms of farfield measuremen ts. Inverse Problems , 30(3):035011, 20, 2014. doi:10.1088/0266-5611/30/3/035011

    work page doi:10.1088/0266-5611/30/3/035011 2014

  2. [2]

    Autio, H

    A. Autio, H. Garde, M. Hirvensalo, and N. Hyvönen. Linear ization-based direct reconstruc- tion for eit using triangular zernike decompositions. Inverse Probl. Imaging , 19(3), 2024. doi:10.3934/ipi.2024040

    work page doi:10.3934/ipi.2024040 2024

  3. [3]

    A. H. Barnett. Aliasing error of the exp(β √ 1 − z2) kernel in the nonuniform fast Fourier transform. Appl. Comput. Harmon. Anal. , 51:1–16, 2021. doi:10.1016/j.acha.2020.10.002

    work page doi:10.1016/j.acha.2020.10.002 2021

  4. [4]

    exponential of semicircle

    A. H. Barnett, J. Magland, and L. af Klinteberg. A paralle l nonuniform fast Fourier transform library based on an “exponential of semicircle” kernel. SIAM J. Sci. Comput. , 41(5):C479–C504,

  5. [5]

    doi:10.1137/18M120885X

    work page doi:10.1137/18m120885x

  6. [6]

    Bürgel, K

    F. Bürgel, K. S. Kazimierski, and A. Lechleiter. Algorit hm 1001: IPscatt – A matlab toolbox for the inverse medium problem in scattering. ACM Trans. Math. Software , 45(4):Art. 45, 20, 2019. doi:10.1145/3328525

    work page doi:10.1145/3328525 2019

  7. [7]

    Cakoni, D

    F. Cakoni, D. Colton, and H. Haddar. Inverse scattering theory and transmission eigenvalues , volume 98 of CBMS-NSF Regional Conference Series in Applied Mathematics . Society for Industrial and Applied Mathematics (SIAM), Philadelphia, PA, 2023. doi:10.1137/1.9781611974461. 21

    work page doi:10.1137/1.9781611974461 2023

  8. [8]

    Colton and A

    D. Colton and A. Kirsch. A simple method for solving inver se scattering problems in the resonance region. Inverse Problems, 12(4):383–393, 1996. doi:10.1088/0266-5611/12/4/003

    work page doi:10.1088/0266-5611/12/4/003 1996

  9. [9]

    Colton and R

    D. Colton and R. Kress. Eigenvalues of the far field operat or for the Helmholtz equation in an ab- sorbing medium. SIAM J. Appl. Math. , 55(6):1724–1735, 1995. doi:10.1137/S0036139993256114

    work page doi:10.1137/s0036139993256114 1995

  10. [10]

    Colton and R

    D. Colton and R. Kress. Inverse acoustic and electromagnetic scattering theory , volume 93 of Applied Mathematical Sciences. Springer, fourth edition, 2019. doi:10.1007/978-3-030-30351-8

    work page doi:10.1007/978-3-030-30351-8 2019

  11. [11]

    Colton and P

    D. Colton and P. Monk. The inverse scattering problem fo r time-harmonic acoustic waves in an inhomogeneous medium. Quart. J. Mech. Appl. Math. , 41(1):97–125, 1988. doi:10.1093/qjmam/41.1.97

    work page doi:10.1093/qjmam/41.1.97 1988

  12. [12]

    https://dlmf.nist.gov/, Release 1.1.11 of 2023- 09-15

    NIST Digital Library of Mathematical Functions . https://dlmf.nist.gov/, Release 1.1.11 of 2023- 09-15. F. W. J. Olver, A. B. Olde Daalhuis, D. W. Lozier, B. I. S chneider, R. F. Boisvert, C. W. Clark, B. R. Miller, B. V. Saunders, H. S. Cohl, and M. A. M cClain, eds. URL: https://dlmf.nist.gov/

    2023

  13. [13]

    Dutt and V

    A. Dutt and V. Rokhlin. Fast Fourier transforms for none quispaced data. SIAM J. Sci. Comput. , 14(6):1368–1393, 1993. doi:10.1137/0914081

    work page doi:10.1137/0914081 1993

  14. [14]

    Garde and M

    H. Garde and M. Hirvensalo. Linearized calderón proble m: Reconstruction of unbounded perturba- tions in three dimensions. SIAM J. Appl. Math. , 85(1):210–223, 2025. doi:10.1137/24M1649162

    work page doi:10.1137/24m1649162 2025

  15. [15]

    Garde and N

    H. Garde and N. Hyvönen. Linearized calderón problem: R econstruction and lipschitz stability for infinite-dimensional spaces of unbounded perturbations. SIAM J. Math. Anal. , 56(3):3588–3604,

  16. [16]

    doi:10.1137/23M1609270

    work page doi:10.1137/23m1609270

  17. [17]

    Griesmaier, M

    R. Griesmaier, M. Hanke, and J. Sylvester. Far field spli tting for the Helmholtz equation. SIAM J. Numer. Anal. , 52(1):343–362, 2014. doi:10.1137/120891381

    work page doi:10.1137/120891381 2014

  18. [18]

    Griesmaier and B

    R. Griesmaier and B. Harrach. Monotonicity in inverse m edium scattering on unbounded domains. SIAM J. Appl. Math. , 78(5):2533–2557, 2018. doi:10.1137/18M1171679

    work page doi:10.1137/18m1171679 2018

  19. [19]

    Griesmaier and L

    R. Griesmaier and L. Schätzle. Far field operator splitt ing and completion in inverse medium scattering. Inverse Problems, 40(11):Paper No. 115010, 32, 2024. doi:10.1088/1361-6420/ad7c77

    work page doi:10.1088/1361-6420/ad7c77 2024

  20. [20]

    Griesmaier and L

    R. Griesmaier and L. Schätzle. Far field operator splitt ing by principal component pursuit. CRC 1173 Preprint 2025/53, Karlsruhe Institute of Technology, dec 2025. doi:10.5445/IR/1000188646

    work page doi:10.5445/ir/1000188646 2025

  21. [21]

    Griesmaier and J

    R. Griesmaier and J. Sylvester. Far field splitting by it eratively reweighted ℓ1 minimization. SIAM J. Appl. Math. , 76(2):705–730, 2016. doi:10.1137/15M102839X

    work page doi:10.1137/15m102839x 2016

  22. [22]

    Griesmaier and J

    R. Griesmaier and J. Sylvester. Uncertainty principle s for inverse source problems, far field splitting, and data completion. SIAM J. Appl. Math. , 77(1):154–180, 2017. doi:10.1137/16M1086157

    work page doi:10.1137/16m1086157 2017

  23. [23]

    Kilgore, S

    K. Kilgore, S. Moskow, and J. C. Schotland. Inverse Born series for scalar waves. J. Comput. Math. , 30(6):601–614, 2012. doi:10.4208/jcm.1205-m3935

    work page doi:10.4208/jcm.1205-m3935 2012

  24. [24]

    A. Kirsch. Characterization of the shape of a scatterin g obstacle using the spectral data of the far field operator. Inverse Problems, 14(6):1489–1512, 1998. doi:10.1088/0266-5611/14/6/009

    work page doi:10.1088/0266-5611/14/6/009 1998

  25. [25]

    A. Kirsch. Remarks on the Born approximation and the fac torization method. Appl. Anal., 96(1):70– 84, 2017. doi:10.1080/00036811.2016.1188286

    work page doi:10.1080/00036811.2016.1188286 2017

  26. [26]

    A. Kirsch. An introduction to the mathematical theory of inverse problem s, volume 120 of Applied Mathematical Sciences. Springer, third edition, 2021. doi:10.1007/978-3-030-63343-1

    work page doi:10.1007/978-3-030-63343-1 2021

  27. [27]

    Kirsch and N

    A. Kirsch and N. Grinberg. The factorization method for inverse problems , volume 36 of Oxford Lecture Series in Mathematics and its Applications . Oxford University Press, Oxford, 2008

    2008

  28. [28]

    S. Meng. Data-driven basis for reconstructing the cont rast in inverse scattering: Picard cri- terion, regularity, regularization, and stability. SIAM J. Appl. Math. , 83(5):2003–2026, 2023. doi:10.1137/23M1545409. 22

    work page doi:10.1137/23m1545409 2003

  29. [29]

    Vainikko

    G. Vainikko. Fast solvers of the Lippmann-Schwinger eq uation. In Direct and inverse problems of mathematical physics (Newark, DE, 1997) , volume 5 of Int. Soc. Anal. Appl. Comput. , pages 423–440. Kluwer Acad. Publ., Dordrecht, 2000. doi:10.1007/978-1-4757-3214-6\_25

    work page doi:10.1007/978-1-4757-3214-6 1997

  30. [30]

    G. N. Watson. A Treatise on the Theory of Bessel Functions . Cambridge University Press, Cam- bridge; The Macmillan Company, New York, 1944

    1944

  31. [31]

    Weidmann

    J. Weidmann. Linear operators in Hilbert spaces , volume 68 of Graduate Texts in Mathematics . Springer-Verlag, New York-Berlin, 1980

    1980

  32. [32]

    Y. Zhou, L. Audibert, S. Meng, and B. Zhang. Exploring lo w-rank structure for an inverse scattering problem with far-field data. SIAM J. Appl. Math. , 86(1):179–205, 2026. doi:10.1137/24M1663922. 23 Figure 5.8: Example 5.3. Real part (left) and imaginary part (right) of reconstruct ed contrast function for wave number κ = 30 from full far field data with 2...

    work page doi:10.1137/24m1663922 2026