Fourier quasicrystals with unit masses

Alexander Olevskii; Alexander Ulanovskii

Fourier quasicrystals with unit masses

T0 review · 0 major / 5 minor · reviewed 2026-08-27 · deepseek-v4-flash

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

Pith's one-line read Every unit-mass Fourier quasicrystal is the zero set of a finite exponential polynomial with real frequencies.

desk verdict Complete converse characterization of unit-mass Fourier quasicrystals; short, honest proof with only minor technical gaps, and the result is clearly new. read the letter →

arxiv 2009.12810 v1 pith:WF5C6I4A submitted 2020-09-27 math.CA math.FA

Alexander Olevskii , Alexander Ulanovskii This is my paper · ORCID

classification math.CAmath.FA MSC 42A3830D1552C23

keywords Fourierquasicrystalscrystallinemeasuresunitmassesexponentialpolynomialszerosetsatomictempereddistributionsuniformlydiscrete

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 claims a complete classification of one-dimensional Fourier quasicrystals with unit masses: measures of the form $\mu=\sum_{\lambda\in\Lambda}\delta_\lambda$ whose distributional Fourier transform is again an atomic measure. The theorem states that if such a measure is a Fourier quasicrystal, then $\Lambda$ is exactly the set of real simple zeros of an exponential polynomial $p(x)=\sum_{k=1}^n b_k e^{i\gamma_k x}$ with $\gamma_k\in\mathbb{R}$. This matters because it rules out genuinely wild supports: despite the many aperiodic constructions, every unit-mass example is controlled by one finite trigonometric sum. The paper also notes the same conclusion holds for integer masses $c_\lambda\in\mathbb{N}$, with real zeros no longer required to be simple.

What carries the argument

The argument runs through a logarithmic-derivative identity. For $\mathrm{Im}\,w>0$ the paper proves $$\sum_{\$\lambda$\in\Lambda}\left(\frac{1}{w-\$\lambda$}+\frac{1}{\$\lambda$}\right)=\$\alpha$-2\pi i\sum_{s\in S,\,s<0}a_s $e^{{-2\pi i s w}}$,$$ where $S$ is the spectrum and $a_s$ the masses of $\hat\mu$; a mirrored identity holds in the lower half-plane. This identity is obtained by convolving the measure with smoothed kernels and using Corollary 1, which says the support is relatively uniformly discrete. Integrating it gives an entire product representation for $\psi$; after removing the factor $e^{\alpha w}$, Phragmén–Lindelöf estimates make the resulting function $p$ of exponential type. The key finishing step multiplies $p$ by $(\sin\pi\varepsilon w/\pi\varepsilon w)^2$ and uses the Paley–Wiener theorem: the inverse Fourier transform of $p_\varepsilon$ has compact support, and evaluating it at a point $u$ isolates the coefficient $b_u$ in the expansion (11), forcing all coefficients $b_u$ with $u<-\sigma$ to vanish. Thus the infinite semigroup expansion collapses to a finite sum.

What would settle it

Two concrete refutations would settle it: exhibit a unit-mass Fourier quasicrystal whose negative spectrum generates a semigroup that is not locally finite (for instance, containing two incommensurable frequencies), or compute a nonzero coefficient $b_u$ with $u<-\sigma$ in expansion (11). Either observation would refute the theorem.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1. If $\mu=\sum_{\lambda\in\Lambda}\delta_\lambda$ is a Fourier quasicrystal, then there exist $n\in\mathbb{N}$, $b_k\in\mathbb{C}$, and $\gamma_k\in\mathbb{R}$ such that the exponential polynomial $p(x)=\sum_{k=1}^n b_k e^{i\gamma_k x}$ has real simple zeros and $\Lambda=\{x\in\mathbb{R}:p(x)=0\}$. In the opposite direction, every exponential polynomial with real simple zeros is the support of a unit-mass Fourier quasicrystal, so the two classes coincide. The paper's Remark 1 extends the forward direction to measures with integer masses $c_\lambda\in\mathbb{N}$: the support is again the zero set of an exponential polynomial with imaginary frequencies, with real zeros not necessarily simple.

Load-bearing premise

The proof assumes that the infinite series (11), summed over the semigroup $U$ generated by the negative spectrum, converges absolutely on the upper half-plane and passes to the real axis in the distributional sense, so that the compact-support Fourier integral genuinely isolates the individual coefficient $b_u$; if this boundary continuation fails, the conclusion that all coefficients below $-\sigma$ vanish does not follow.

Editorial extensions

If this is right

Unit-mass Fourier quasicrystal supports are exactly the real zero sets of finite exponential polynomials with real frequencies; there are no other examples.
An aperiodic unit-mass Fourier quasicrystal must still have a support set with the rigid structure of trigonometric zero sets, so its counting function and gap structure are constrained by finite exponential-type data.
The integer-mass case, with coefficients in $\mathbb{N}$, has the same support rigidity, with real zeros allowed to have multiplicities.
The frequencies $\gamma_k$ of the exponential polynomial lie in the additive semigroup generated by the negative spectrum of the measure, because the expansion (11) runs over that semigroup before collapsing to a finite sum.

Reading between the lines

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

One question the paper leaves open is whether positivity or equal masses can be relaxed: the proof uses the unit-mass form to isolate simple zeros, so a classification for arbitrary complex masses would require additional arguments.
A testable consequence is that every concrete unit-mass Fourier quasicrystal should satisfy the coefficient-vanishing test: computing the expansion (11) from its spectrum should yield finitely many nonzero coefficients, and finding one with infinitely many would contradict the theorem.
The one-dimensional proof depends on canonical products and boundary-value continuation, so the classification may not transfer automatically to higher dimensions; a different mechanism would be needed there.

Share X Bluesky LinkedIn Reddit HN

Formalized claims in Lean

Claim #1: The central claim is Theorem 1. If $\mu=\sum_{\lambda\in\Lambda}\delta_\lambda$ is a Fourier quasicrystal, then there exist $n\in\mathbb{N}$, $b_k\in\mathbb{C}$, and $\gamma_k\in\mathbb{R}$ such that the exponential polynomial $p(x)=\sum_{k=1}^n b_k e^{i\gamma_k x}$ has real simple zeros and $\Lambda=\{x\in\mathbb{R}:p(x)=0\}$. In the opposite direction, every exponential polynomial with real simp

/-- @claim 1 The central claim is Theorem 1. If $\mu=\sum_{\lambda\in\Lambda}\delta_\lambda$ is a Fourier quasicrystal, then there exist $n\in\mathbb{N}$, $b_k\in\mathbb{C}$, and $\gamma_k\in\mathbb{R}$ such that the exponential polynomial $p(x)=\sum_{k=1}^n b_k e^{i\gamma_k x}$ has real simple zeros and $\Lambda=\{x\in\mathbb{R}:p(x)=0\}$. In the opposite direction, every exponential polynomial with real simp -/
def central_claim : Prop :=

Signed reviews

No signed human review yet.

Request a human review

A listed scientist reviews the paper for a fee and the review publishes here regardless of verdict. See the reviewers or get listed.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

This paper proves the converse of the earlier Olevskii–Ulanovskii construction: every Fourier quasicrystal with unit masses is the real zero set of an exponential polynomial with imaginary frequencies. That is a real theorem, not a variation on known material, and it gives the clean structural description people have been circling since Kurasov–Sarnak and Meyer. What it does well: the argument is direct and mostly self-contained, the two main analytic ingredients (the logarithmic-derivative identity in Lemma 1 and the Paley–Wiener step to force finite support of the exponent expansion) are the right tools, and Proposition 1 with its corollary on relative uniform discreteness is a useful standalone result. The paper is also honest: it says explicitly when a step is left to the reader.

The genuinely soft spots are minor. The "one may check" after (11) is the largest one, and the stress-tester is right that it does not sink the paper: local finiteness of U follows from local finiteness of S minus zero accumulation, and the majorant for the series is exp of a finite sum because |a_s| = O(|s|^m). The boundary issue can be avoided by integrating along a horizontal line in the upper half-plane and using Cauchy's theorem termwise. There is a small sign/convention inconsistency in the claimed support interval for the Paley–Wiener step, but it is harmless because it only shifts the cut-off point, and the conclusion that b_u = 0 for all sufficiently negative u stands. So the proof holds.

I have no objection to the citation pattern. The paper cites the construction it is completing, and that is not circularity—it is the actual converse. The authors do not overclaim; the integer-mass remark is clearly labeled as an extension, not a theorem.

Who gets value from this: anyone working on crystalline measures, quasicrystals, or exponential polynomial zero sets. The paper deserves a serious referee; it is short, important within the subfield, and the result is likely to be used. I would bring it to reading group and cite it myself.

Referee Report

0 major / 5 minor

Summary. The paper proves that if the unit-mass atomic measure μ = Σ_{λ∈Λ} δ_λ is a Fourier quasicrystal, then Λ is exactly the real zero set of an exponential polynomial p with real simple zeros and imaginary frequencies. The proof combines a growth bound (Proposition 1) showing that Λ is relatively uniformly discrete, a logarithmic-derivative identity (Lemma 1) for the canonical product of Λ in the upper and lower half-planes, and a Phragmén–Lindelöf/Paley–Wiener argument that forces the expansion of p into exponentials to be finite. The converse direction is cited to the authors' companion paper [OU20].

Significance. The result is a clean structural classification of unit-mass Fourier quasicrystals and provides the converse to the existence construction in [OU20]. The proof is self-contained and uses no fitted parameters or ad hoc assumptions; the main work is a careful complex-analysis argument. If the proof is correct, the theorem fully characterizes which uniformly discrete sets can carry a unit-mass FQ, and the paper is likely to be a standard reference in the area. The main theorem is sharp and falsifiable, and the auxiliary Proposition 1 is of independent interest.

minor comments (5)

[Section 3, Eq. (11)] The sentence 'One may check that U is a locally finite set and that the series in (11) converges absolutely' should be replaced with a short proof: S_- is bounded away from 0 by local finiteness, which makes every bounded interval contain only finitely many finite sums from U, and for Im w > 0 the sum of absolute values is bounded by |C| exp(Σ_{s∈S_-} |a_s/s| e^{2π s y}), which is finite by (2).
[Section 3, final paragraph] Please specify that the integral defining ˇp_ε(u) is taken along a horizontal line Im w = η > 0, where the series (11) converges uniformly in x, and justify the interchange of summation and integration as well as the shift of each term to the real axis via Cauchy's theorem; no distributional boundary limit is needed.
[Lemma 1] In the displayed properties of \hat f_ε, the statement 'fε(t)=0, t≥0' is a typo; it should read '\hat fε(t)=0, t≥0'.
[Lemma 1] The proof assumes the existence of the greatest negative element s1 of S; if S has no negative elements, the same argument with the sum over s<0 empty (and α0=0) yields (5), so a sentence covering this case would make the lemma fully general.
[Section 3, after Corollary 1] The implication from relative uniform discreteness to Σ_{λ≠0}|λ|^{-1-ε}<∞ should be stated for some (or all) ε∈(0,1), so that the genus-one canonical product (7) converges by Σ|λ|^{-2}<∞.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the proof of Theorem 1 is self-contained and does not depend on the authors' prior construction.

full rationale

The derivation chain of Theorem 1 is self-contained and does not assume the conclusion. Proposition 1 and Corollary 1 are proved from the definition of a Fourier quasicrystal, temperedness, and the growth condition (2) using a standard Schwartz-function mollifier argument. Lemma 1 derives the identities (5) and (6) from representation (4) and the convergence provided by Corollary 1, letting a mollifier tend to zero; no target statement is used. The entire function ψ is defined directly from the support Λ by the canonical product (7), so its logarithmic derivative equals the left-hand side of (5), and the upper half-plane representation (9) is the integrated form of Lemma 1. The expansion (11) is obtained by expanding the exponentials exp((a_s/s)e^{-2πisw}) into absolutely convergent power series; the sentence 'One may check that U is a locally finite set and that the series in (11) converges absolutely' is a technical verification, not a circular invocation of the result. The vanishing of coefficients b_u is a genuine Paley-Wiener argument: for each u < -σ, one chooses ε with u < -σ-ε and no other U-point nearby, and then 0 = ˇp_ε(u) = b_u ∫ (sin π ε x / π ε x)^2 dx, forcing b_u = 0. This is a direct test of the coefficient, not a renamed fit or a predicated conclusion. The only cited prior work, [OU20], is used in the introduction and Remark 1 for the converse direction, not as a premise in the proof of Theorem 1. There are no fitted parameters, no normalization that forces the result by construction, and no imported uniqueness theorem. Any possible technical concerns about the 'One may check' assertion are correctness risks, not circularity.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claim rests on classical theorems of complex and harmonic analysis. No free parameters are fitted. No new entities are postulated. The only potentially non-explicit support is the distributional boundary value step, which is standard.

assumptions (4)

standard math Paley-Wiener theorem for entire functions of exponential type
Used in Section 3 to conclude that the inverse Fourier transform of p_epsilon vanishes outside (-sigma-epsilon, epsilon). This is a classical theorem.
standard math Phragmen-Lindelof principle variants
Used in Section 3 to deduce that p is of exponential type and bounded in strips, relying on bounds in half-planes. Standard complex analysis.
standard math Weierstrass product theorem with Borel's theorem on order-one entire functions
Used to construct psi(w) from the zero set Lambda and to assert psi is entire of order one. Standard.
standard math Distributional boundary values of analytic functions
Implicitly used to apply the series (11) on the real axis when computing the Fourier transform of p_epsilon. This is a standard but unstated tool.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Fourier quasicrystals with unit masses." pith.science (2026). https://pith.science/paper/WF5C6I4A

@misc{pith2026200912810,
  author       = {Pith},
  title        = {Pith review of: Fourier quasicrystals with unit masses},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WF5C6I4A}},
  note         = {Machine review of arXiv:2009.12810}
}

abstract

Every set $\Lambda\subset R$ such that the sum of $\delta$-measures sitting at the points of $\Lambda$ is a Fourier quasicrystal, is the zero set of an exponential polynomial with imaginary frequencies.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

8 extracted references · 8 canonical work pages

[1]

Kolountzakis, M., Fourier pairs of discrete support with little structure. J. Fourier Anal. Appl. 22, no. 1, (2016), 1--5

work page 2016
[2]

Stable polynomials and crystalline measures

Kurasov, P., Sarnak, P. Stable polynomials and crystalline measures. J. Math. Phys. 61, no. 8. 083501 (2020); https://doi.org/10.1063/5.0012286

work page doi:10.1063/5.0012286 2020
[3]

Lectures on Entire Fuctions

Levin, B.Ya. Lectures on Entire Fuctions. American Mathematical Soc. 1996

work page 1996
[4]

Quasicrystals with discrete support and spectrum

Lev, N, Olevskii, A. Quasicrystals with discrete support and spectrum. Revista Mat. Iberoam. 32 , no. 4, (2016), 1341--1352

work page 2016
[5]

Meyer, Y., Measures with locally finite support and spectrum. Proc. Natl. Acad. Sci. USA 113, no. 12, (2016), 3152--3158

work page 2016
[6]

Measures with locally finite support and spectrum

Meyer, Y. Measures with locally finite support and spectrum. Rev. Mat. Iberoam., 33 , no. 3, (2017), 1025--1036

work page 2017
[7]

Curved model sets and crystalline measures

Meyer, Y. Curved model sets and crystalline measures. To be published in: `Applied and Numerical Harmonic Analysis'. Springer

work page
[8]

A Simple Crystalline Measure

Olevskii, A., Ulanovskii A. A Simple Crystalline Measure. arXiv:2006.12037, (2020)

work page Pith review arXiv 2020

Pith tools

Reviewed August 27, 2026 · model on record in the stance chip above.

[1] [1]

Kolountzakis, M., Fourier pairs of discrete support with little structure. J. Fourier Anal. Appl. 22, no. 1, (2016), 1--5

work page 2016

[2] [2]

Stable polynomials and crystalline measures

Kurasov, P., Sarnak, P. Stable polynomials and crystalline measures. J. Math. Phys. 61, no. 8. 083501 (2020); https://doi.org/10.1063/5.0012286

work page doi:10.1063/5.0012286 2020

[3] [3]

Lectures on Entire Fuctions

Levin, B.Ya. Lectures on Entire Fuctions. American Mathematical Soc. 1996

work page 1996

[4] [4]

Quasicrystals with discrete support and spectrum

Lev, N, Olevskii, A. Quasicrystals with discrete support and spectrum. Revista Mat. Iberoam. 32 , no. 4, (2016), 1341--1352

work page 2016

[5] [5]

Meyer, Y., Measures with locally finite support and spectrum. Proc. Natl. Acad. Sci. USA 113, no. 12, (2016), 3152--3158

work page 2016

[6] [6]

Measures with locally finite support and spectrum

Meyer, Y. Measures with locally finite support and spectrum. Rev. Mat. Iberoam., 33 , no. 3, (2017), 1025--1036

work page 2017

[7] [7]

Curved model sets and crystalline measures

Meyer, Y. Curved model sets and crystalline measures. To be published in: `Applied and Numerical Harmonic Analysis'. Springer

work page

[8] [8]

A Simple Crystalline Measure

Olevskii, A., Ulanovskii A. A Simple Crystalline Measure. arXiv:2006.12037, (2020)

work page Pith review arXiv 2020