Pith. sign in

REVIEW 4 major objections 5 minor 24 references

Direct spectral problems for Paley-Wiener canonical systems

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read For real Dirac systems, the direct spectral problem is solved by step-function approximation: the spectral measures of the approximating Hamiltonians converge to the true spectral measure on Paley-Wiener test functions.

desk verdict Useful explicit algorithm for step-function Hamiltonians, but the convergence theorem for non-step Hamiltonians has a non-rigorous proof as written. read the letter →

arxiv 2505.00669 v1 pith:EUQ5JVPW submitted 2025-05-01 math.SP

Ashley R. Zhang This is my paper · ORCID
classification math.SP MSC 34A5534L0547B35
keywords spectralmeasurecanonicalHamiltoniansystemsPaley-WienerspacesdeBrangesrealDiracstep-functionapproximationsToeplitzdeterminantsVerblunskycoefficients
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper solves the direct spectral problem for a class of Paley-Wiener canonical systems: given the Hamiltonian, find its spectral measure. The method is to approximate a general Hamiltonian by step functions, solve the direct problem for each step-function approximation using orthogonal polynomials and Toeplitz determinants, and then shrink the step size to zero. The central theorem proves that for Hamiltonians coming from real Dirac systems, the spectral measures of the step-function approximations converge to the spectral measure of the original system in the sense of $L^2$ norms on Paley-Wiener test functions. This gives a finite-dimensional algorithm for the direct problem and shows that the truncated-Toeplitz approach to inverse spectral problems can be run in reverse.

What carries the argument

The argument moves through the Hermite-Biehler functions $E(t,z)=u(t,z)-iv(t,z)$ associated with the canonical system; the spectral measure satisfies the norm identity $\|\varphi\|_{L^2(\mu)}^2 = \int_{\mathbb{R}} |\varphi(x)|^2\,dx/|E(a,x)|^2$ for $\varphi$ in the corresponding de Branges space. For real Dirac systems, $E$ obeys the scattering equation $\partial_t E(t,x)=f(t)e^{2itx}\overline{E(t,x)}$, which yields uniform bounds and, when $f$ is replaced by its step average $f_T$, a uniform convergence estimate $E_T(a,x)\to E(a,x)$ as $T\to0$. The step-function case is handled algebraically: the Szegő recurrence for orthogonal polynomials on the unit circle and finite Toeplitz determinants recover the moments and Verblunsky coefficients of the periodic spectral measure from the step heights $h_{11,n}=|\varphi_n(1)|^2$.

What would settle it

For the real Dirac system with $h_{11}(t)=e^t$ (Example 6.5), compute numerically the norms $\|\varphi\|_{L^2(\mu_T)}$ for a fixed $\varphi\in PW_1$ as $T\to0$ and compare them with the integral against the claimed limiting density $\sqrt{4x^2-1}/(2|x|)\,dx$; a systematic discrepancy would show the convergence in Theorem 5.1 fails.

Watch

Extended reading notes

Core claim

The paper's central result, Theorem 5.1, states that if $H$ is the det-normalized diagonal Hamiltonian of a real Dirac system, with $h_{11}(t)=\exp(\int_0^t f(s)ds)$ for a real-valued $f\in L^1_{\mathrm{loc}}$, and $H_T$ are the step-function approximations obtained by averaging $h_{11}$ over intervals of length $T$, then for every $a>0$ and every $\varphi\in PW_a$, the $L^2(\mu_T)$ norm of $\varphi$ converges to its $L^2(\mu)$ norm as $T\to0$. This means the spectral measures $\mu_T$ converge to the true spectral measure $\mu$ on the Paley-Wiener test functions that characterize PW-systems. The proof establishes uniform convergence of the scattering functions $E_T(a,x)$ to $E(a,x)$, which transfers to the reciprocals $1/|E_T(a,x)|^2$ because $|E(a,x)|$ stays bounded away from zero; Hölder's inequality then gives the norm convergence.

Load-bearing premise

The load-bearing premise is that the Hamiltonian has the special exponential form produced by a real Dirac system, and the paper does not establish the same convergence for general Paley-Wiener Hamiltonians.

Editorial extensions

If this is right

  • For any real Dirac system, the spectral measure can be computed to arbitrary accuracy by averaging the Hamiltonian over small intervals and then running the finite-dimensional Toeplitz/orthogonal-polynomial algorithm for periodic measures.
  • The direct and inverse spectral problems for Paley-Wiener canonical systems become two directions of the same step-function approximation procedure, so a measure and its Hamiltonian can be recovered from each other consistently.
  • The convergence of the approximate spectral measures holds on every Paley-Wiener space PW_a, the exact class of test functions relevant to the sampling property of PW-measures.
  • For step-function Hamiltonians with uniform step size, the direct spectral problem reduces to finding the unique even measure on the circle with prescribed values |φ_n(1)|^2, and both the Verblunsky coefficients and the moments can be recovered algebraically.
  • The examples confirm the method on known cases and produce a closed-form limiting measure for the exponential Hamiltonian h11(t)=e^t.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • We conjecture that Theorem 5.1 extends to any Paley-Wiener Hamiltonian whose Hermite-Biehler functions satisfy a scattering-type equation with an L1-loc coefficient, not only the exact exponential form of real Dirac systems; the paper's proof only needs the uniform bound and the L1 approximation.
  • The closed-form limit for h11(t)=e^t suggests a more general correspondence between exponential growth rates of the Hamiltonian and the location of spectral support; testing h11(t)=e^{ct} for c different from 1 would probe this.
  • Because the paper treats only diagonal Hamiltonians and even spectral measures, a natural extension is to apply the same periodization to non-diagonal step-function Hamiltonians, where the Verblunsky coefficients are replaced by matrix-valued recurrences.
  • The connection to orthogonal polynomials may allow the direct spectral problem to be computed by fast Toeplitz algorithms, making the step-function approximation practical for numerical spectral computations.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. This note studies direct spectral problems for Paley–Wiener canonical Hamiltonian systems on the half-line. For diagonal Hamiltonians that are step functions with uniform step size, the spectral measure is an even periodic Paley–Wiener measure, and the paper gives two algebraic algorithms to recover it from the sequence of step values h_11^n: one via Verblunsky coefficients (Section 3) and one via moments and Toeplitz determinants (Section 4). For non-step Hamiltonians, the paper defines step-function approximations H_T by averaging h_11 on intervals and claims in Theorem 5.1 that, for Hamiltonians arising from real Dirac systems, the associated spectral measures μ_T converge to μ in the sense that ∫|φ|^2 dμ_T → ∫|φ|^2 dμ for every φ in a Paley–Wiener space. The final section contains numerical illustrations of the algorithms and explicit limiting measures.

Significance. The step-function part is a clean application of OPUC theory that yields explicit, finite-dimensional formulas and complements the inverse spectral algorithm of Makarov–Poltoratski. If Theorem 5.1 is established, the paper would provide a practical periodization method for direct spectral problems in a nontrivial subclass of PW-systems, and the examples demonstrate that the method produces meaningful approximations. However, the main convergence theorem currently rests on an incorrect differential identity and an invalid energy estimate, so the contribution is not yet rigorously established.

major comments (4)
  1. [Section 5, Eq. (17)] The identity ∂_t E(t,x)=f(t)e^{2itx}E(t,x) is false on the real axis because E^#(t,x)=overline{E(t,x)}, not E(t,x). With the scattering function defined as E(t,z)=e^{itz}E(t,z), the correct real-axis equation is ∂_t E(t,x)=f(t)e^{2itx}overline{E(t,x)}. This error invalidates the differential equation for D=E−E_T that drives the proof of Theorem 5.1.
  2. [Section 5, proof of Theorem 5.1] The energy estimate treats the complex-valued D as real. The displayed identity (1/2)∂_t[D]^2=f e^{2itx}|D|^2−[f_T−f]e^{2itx}E_T D is not the derivative of |D|^2, and the subsequent inequality |∂_t D^2|≤... does not follow. The argument can likely be repaired by multiplying by \bar{D} and using ∂_t|D|^2=2 Re(\bar{D}∂_t D), but that argument is absent.
  3. [Section 5, paragraph after (14)] The paper states that f_T is a discrete measure and asserts ∫_0^a |f_T(s)|ds → ∫_0^a |f(s)|ds and ∫_0^a |f_T−f|dt → 0 as T→0. Since f_T is defined implicitly through the logarithmic jumps of the averages of h_11, these convergence statements are not automatic and no proof is provided. The final estimate in Theorem 5.1 requires such an L^1-convergence statement.
  4. [Abstract and Section 1] The abstract claims convergence of the spectral measures for 'a non-step-function Hamiltonian' without the restriction to real Dirac systems. Theorem 5.1 is proved only for real Dirac Hamiltonians, and the introduction itself states that the class of PW-Hamiltonians is not characterized. The abstract and the opening paragraphs should be amended to reflect the actual scope.
minor comments (5)
  1. [Section 2.1] The definition of H^2(C+) is missing the dx in the integral; it should read sup_{y>0} ∫_R |f(x+iy)|^2 dx.
  2. [Section 4.2 and 4.3] The symbol ⊮ is used without definition; it appears to be an OCR artifact for a row vector of ones. Please define it or replace it with standard notation.
  3. [Section 5, sentence after (14)] The sentence contains a typo: 'as T → ∞' should be 'as T → 0' in the sentence about ∫_0^t |f_T(s)|ds.
  4. [Example 6.5] The formula for w_T(θ) should include absolute values around sin(Tθ), and the support interval should reflect the absolute value of the arcsine. Also, 'dTµ(x)' should be 'dμ_T(x)'.
  5. [Section 3.2, Eq. (4)] The derivation of (4) uses φ_n^*(1)=φ_n(1), which holds for even measures; this justification could be stated explicitly for the reader.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the step-Hamiltonian inversion is an algebraic converse of a known correspondence, and Theorem 5.1 derives convergence from the scattering equation rather than assuming it.

full rationale

The central inversion is not circular. For a step Hamiltonian, the constants h_{11}^{n} are inputs and the spectral measure is the output; the paper recovers the Verblunsky coefficients from the algebraic ratio (φ_{n+1}(1)/φ_n(1))² = (1−α_n)/(1+α_n) and then invokes Verblunsky's theorem, an external uniqueness result, to produce the measure. The moments algorithm in Section 4 solves for c_n from the same h-values via Heine formulas and Toeplitz determinants; none of these steps defines the Hamiltonian in terms of the output measure. Theorem 5.1 is likewise one-way: it estimates E−E_T using the scattering equation (17), obtains uniform convergence of |E_T|² to |E|² from an integrated bound on |f_T−f|, and concludes μ_T→μ on Paley-Wiener test functions. The proof does not presuppose the convergence it aims to establish, and no fitted parameter is renamed as a prediction. The self-citations ([18], [24]) are contextual or accompany external references ([14], [20]) for a standard reduction; they are not load-bearing for Theorem 2.9 or Theorem 5.1. A reviewer's objection about the complex-valued energy estimate is a question of rigor, not circularity, and the introduction explicitly restricts the convergence theorem to real Dirac systems rather than all PW-Hamiltonians.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The paper introduces no free parameters or invented entities. Its reasoning relies on standard orthogonal polynomial theory, Toeplitz determinant identities, and the de Branges space framework, plus the specialized class of real Dirac Hamiltonians. The key imported result is the equivalence between step-function Hamiltonians and |φ_n(1)|^2 values from [13].

assumptions (6)
  • domain assumption The Hamiltonian H is real symmetric positive semidefinite, locally integrable, det H ≠ 0 a.e., and det-normalized.
    Standard assumptions for canonical systems; the det H ≠ 0 condition ensures no jump intervals and isometric de Branges inclusions (Section 1, equation (1)).
  • domain assumption The Hamiltonian has the real Dirac form h_11(t)=exp(∫_0^t f(s) ds) with f∈L^1_loc.
    Theorem 5.1 is proved only for real Dirac systems; the differential equation (17) for the scattering function relies on this structure.
  • domain assumption For a step-function Hamiltonian with step size 1/2, h_11 on the n-th step equals |φ_n(1)|^2.
    Theorem 2.7, taken from [13], connects the Hamiltonian to values of orthonormal polynomials; it is the bridge for the direct algorithm.
  • standard math Verblunsky's theorem: every sequence in the unit disk defines a unique probability measure on T.
    Used to recover the measure from Verblunsky coefficients in Section 3.
  • standard math Carathéodory-Toeplitz: positivity of Toeplitz determinants characterizes finite positive measures with infinite support.
    Used to guarantee the moment sequence defines a measure in Section 4.
  • standard math Heine formulas relate monic orthogonal polynomials to Toeplitz determinants.
    Used to derive the moment recovery formula (12) in Section 4.2.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Direct spectral problems for Paley-Wiener canonical systems." pith.science (2026). https://pith.science/paper/EUQ5JVPW

@misc{pith2026250500669,
  author       = {Pith},
  title        = {Pith review of: Direct spectral problems for Paley-Wiener canonical systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EUQ5JVPW}},
  note         = {Machine review of arXiv:2505.00669}
}
abstract

This note focuses on the direct spectral problem for canonical Hamiltonian systems on the half-line $\mathbb{R}_+$. Truncated Toeplitz operators have been effectively used to solve the inverse spectral problem when the spectral measure is a locally finite periodic measure (see \cite{MP}). Here, we reverse the inverse problem algorithm to solve the direct spectral problem for step-function Hamiltonians. For a non-step-function Hamiltonian, we consider its step-function approximations and their corresponding spectral measures, and show that these spectral measures converge to the spectral measure of the original Hamiltonian.

Figures

Figures reproduced from arXiv: 2505.00669 by the authors.

Figure 1
Figure 1. First 20 terms of the cosine series of µ T and µ. Upper left: T = 1 2 , upper right: T = 1 4 , lower left: T = 1 8 , lower right: T = 1 16 . Now we look at some new examples. Example 6.4. Let H be the diagonal det-normalized Hamiltonian with h11(t) = 1 + t. For each step-function approximation h T 11, we can compute the moments of the spectral measure dµT (x) = wT (x)dx, where wT (x) is a cosine series. In the follo… view at source ↗
Figure 2
Figure 2. First 20 terms of the cosine series of µ T . Blue: T = 1 2 . Orange: T = 1 4 . Yellow: T = 1 8 . Purple: T = 1 16 . 16 [PITH_FULL_IMAGE:figures/full_fig_p016_2.png] view at source ↗
Figure 3
Figure 3. wT (x) where d T µ(x) = wT (x)dx. Blue: T = 1 2 . Orange: T = 1 4 . Purple: T = 1 8 . Black: T = 1 16 . Since wT (x) is explicitly known for each T, we can determine the limiting measure µ as well. The limit of µT is given by dµ(x) = √ 4x 2 − 1 2∣x∣ dx, (18) supported on (−∞,− 1 2 ] ∪ [ 1 2 ,∞). 17 [PITH_FULL_IMAGE:figures/full_fig_p017_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

24 extracted references · 23 canonical work pages

  1. [13]

    Makarov and A

    N. Makarov and A. Poltoratski. ”Etudes for the inverse spectral problem”, Journal of the London Mathematical Society 108.3(2023), 916-977

    work page 2023

  2. [1]

    Bessonov

    R. Bessonov. ”Sampling measures, Muckenhoupt Hamiltonians, and triangular factorization”, Inter- national Mathematics Research Notices, (2018) no. 12, 3744–3768

    work page 2018

  3. [2]

    Bessonov and R

    R. Bessonov and R. Romanov. ”An inverse problem for weighted Paley-Wiener spaces”, Inverse Problems, 32 (11), (2016)

    work page 2016

  4. [3]

    G. Borg. ”Eine Umkehrung der Sturm-Liouvilleschen Eigenwertaufgabe”, Acta Mathematics, 78 (1946), 1-96

    work page 1946

  5. [4]

    G. Borg. ”Uniqueness theorems in the spectral theory ofy′′ +(λ−q(x))y = 0”, Proc. 11-th Scandinavian Congress of Mathematicians, Johan Grundt Tanums Forlag, Oslo, (1952), 276-287

    work page 1952

  6. [5]

    J.-F. Burnol. ”Two complete and minimal systems associated with the zeros of the Riemann zeta function”, Journal de Th´ eorie des Nombres de Bordeaux,16.1 (2004), 65-94

    work page 2004

  7. [6]

    de Branges

    L. de Branges. ”Hilbert spaces of entire functions”, Prentice-Hall, Englewood Cliffs, NJ, 1968

    work page 1968

  8. [7]

    S. A. Denisov. ”Continuous analogs of polynomials orthogonal on the unit circle and Krein systems”, International Mathematics Research Surveys, Volume 2006 (2006), 426-447

    work page 2006

Show all 24 references
  1. [8]

    Dym and H

    H. Dym and H. P. McKean. ”Gaussian processes, function theory and the inverse spectral problem”, Academic Press, New York, 1976. Courier Corporation, Jan. 2008

    1976

  2. [9]

    I. M. Gelfand and B. M. Levitan. ”On the determination of a differential equation from its spectral function” (Russian), Izvestiya Akad. Nauk SSSR, Ser. Mat. 15 (1951), 309-360; English translation in American Mathematical Society Translation, Series 2, Volume 1 (1955), 253-304

    1951

  3. [10]

    Golinskii

    L. Golinskii. ”Geronimus polynomials and weak convergence on a circular arc”,Methods of applications of analysis, 6.4 (1999), 412-436

    1999

  4. [11]

    Lagarias

    J.C. Lagarias. ”Zero Spacing Distributions for Differenced L-Functions”,Acta Arithmetica 120 (2006), 159-184

    2006

  5. [12]

    Lagarias

    J.C. Lagarias. ”The Schr¨ odinger Operator with Morse Potential on the Right Half Line”,Communi- cations in Number Theory and Physics 3.2(2009), 323-361

    2009

  6. [14]

    Makarov, A

    N. Makarov, A. Poltoratski, and M. Sodin. ”Lectures on linear complex analysis”, unpublished lecture notes

  7. [15]

    Marchenko

    V. Marchenko. ”Some questions in the theory of one-dimensional linear differential operators of the second order. I”, Trudy Mosk. Mat. Obsch. 1 (1952), 327–420

    1952

  8. [16]

    Marchenko

    V. Marchenko. ”Sturm-Liouville operators and applications”, Birkh¨ auser Basel, 1986

    1986

  9. [17]

    Poltoratski

    A. Poltoratski. ”Pointwise convergence of the non-linear Fourier transform”, Annals of Mathematics 199 (2024) 741-793

    2024

  10. [18]

    Poltoratski and A

    A. Poltoratski and A. R. Zhang ”Periodic approximations in inverse spectral problems for canonical Hamiltonian systems”, Journal of Functional Analysis 284.11(2023)

    2023

  11. [19]

    C. Remling. ”Spectral Theory of Canonical Systems”, De Gruyter Studies in Mathematics, De Gruyter, 2018

    2018

  12. [20]

    R. Romanov. ”Canonical Systems and de Branges Spaces”, 2014. arXiv:1408.6022

    2014 arXiv

  13. [21]

    B. Simon. ”Orthogonal polynomials on the unit circle, part I”, Vol. 54. Colloquium Publications. American Mathematical Society, 2004

    2004

  14. [22]

    W. F. Trench. ”An Algorithm for the Inversion of Finite Toeplitz Matrices”, Journal of the Society for Industrial and Applied Mathematics 12.3 (1964), 515-522

    1964

  15. [23]

    Valk´ o and B

    B. Valk´ o and B. Vir´ ag. ”Schr¨ odinger operators and de Branges spaces”, Geometric and Functional Analysis 32(2022), 1160-1231

    2022

  16. [24]

    A. R. Zhang. ”Convergence from the discrete to the continuous non-linear Fourier transform”, Trans. Amer. Math. Soc. 378 (2025), 1483-1501. V anderbilt University, Department of Mathematics, 1326 Stevenson Center, Nashville, TN 37240, USA Email address : ashley.zhang@vanderbilt.edu 18

    2025

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.