Upper bounds for Fourier decay rates of fractal measures

Xiumin Du

Upper bounds for Fourier decay rates of fractal measures

T0 review · 0 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read This paper proves that fractal measures of dimension α in [d−1,d) have parabolic average Fourier decay exactly (d−1)α/d, and it improves all prior upper bounds on spheres by an intermediate-dimension construction.

desk verdict Du improves the known upper bounds for average Fourier decay rates and pins down the exact parabolic rate on [d−1,d); the paper deserves a serious referee. read the letter →

arxiv 1908.05753 v1 pith:WXCPCDLH submitted 2019-08-15 math.CA

Xiumin Du This is my paper · ORCID

classification math.CA MSC 42B1028A80

keywords Fourierdecayratesfractalmeasuressphericalaveragesparabolicaverageintermediatedimensiondistancesetsrestrictionestimates

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

Fractal measures have Fourier transforms that typically decay as the frequency grows; this paper asks how fast the average of that transform over a sphere or a paraboloid must decay. For a measure of dimension $\alpha$ in $\mathbb{R}^d$, the answer is unknown in most of the range $d/2<\alpha

What carries the argument

The load-bearing mechanism is the intermediate-dimension parameter $\kappa$, chosen from five functions $\kappa_1,\dots,\kappa_5$ of $m,\alpha,d$. A cap $\Omega$ is formed by an $m$-dimensional ball of radius $R^{-1/2}$ fibered over lattice points spaced $R^{\kappa-1}$ on the hypersurface, and $\Lambda$ is the matching set in physical space, with the normal coordinate quantized so that $R(x\cdot\xi)$ is nearly an integer multiple of $2\pi$. The two lemmas do the real work: on $\Lambda$ the phase is almost constant modulo $2\pi$, giving $|Ef(Rx)|\sim\sigma(\Omega)$, and the dimension constant $c_\alpha(\mu)$ is $\sim R^{\alpha-d}$ exactly when $\kappa$ equals the prescribed formula. Balancing these two estimates reduces the upper bound to the exponent computation $\beta\le\alpha-1+2\kappa$; the piecewise definitions in Theorems 1.1 and 1.2 are exactly the choices of $m$ and $\kappa$ that minimize this exponent for each interval of $\alpha$.

What would settle it

For the exact parabolic claim, take $d=3$, $\alpha=2$, where the paper predicts $\beta_3(\alpha,P^2)=4/3$; a direct evaluation of $\|\hat\mu(R\,\cdot)\|^2_{L^2(P^2)}$ for the measure constructed in Section 3 along $R\to\infty$ would either stay comparable to $R^{-4/3}$ or grow relative to it, settling the formula. For the spherical upper bound, a direct number-theoretic count of the lattice-point set $\Gamma$ for $d-m=2$ would test the external estimate on which the construction's volume lower bound rests.

Watch

Extended reading notes

Core claim

The paper establishes that for the truncated paraboloid $P^{d-1}\subset\mathbb{R}^d$ and any $\alpha$-dimensional measure with $d-1\le\alpha<d$, $d\ge3$, the average Fourier decay exponent is exactly $\beta_d(\alpha,P^{d-1})=(d-1)\alpha/d$. The upper-bound half is new: for each $\alpha$ and a chosen integer $m$ with $0<m<d/2$, one constructs a measure $\mu=\chi_\Lambda\,dx$ supported on a thin lattice-like set and a test function $f$ supported on a matching cap $\Omega$ on the paraboloid; the phase of the extension operator is within $1/100$ of $2\pi\mathbb{Z}$ on $\Lambda$, so $|Ef(Rx)|\sim|\Omega|$ there. Choosing the lattice spacing parameter $\kappa$ so that $c_\alpha(\mu)\sim R^{\alpha-d}$ makes the failure of (2.5) as strong as possible, forcing $\beta_d(\alpha,P^{d-1})\le\alpha-1+2\tilde\kappa(\alpha,d)$. Part (a) gives $\tilde\kappa=(d-\alpha)/(2d)$, which together with the lower bounds yields the exact formula. The same construction on the sphere gives improved piecewise upper bounds $\beta_d(\alpha,S^{d-1})\le\alpha-1+2\kappa(\alpha,d)$, though these do not reach the lower bounds.

Load-bearing premise

The load-bearing premise is the cited but unproved lattice-point estimate that $\#\{\omega\in S^{d-m-1}:R^\kappa\omega\in2\pi\mathbb{Z}^{d-m}\}\gtrsim R^{\kappa(d-m-2)}$ along a sequence $R\to\infty$, which the spherical volume estimate (2.9) needs and which the exact parabolic result does not use.

Editorial extensions

If this is right

For every $d\ge3$ and $d-1\le\alpha<d$, the parabolic average Fourier decay rate is now known exactly, so any competing upper-bound construction in that range must agree with $\beta_d(\alpha,P^{d-1})=(d-1)\alpha/d$.
In the full range $\alpha\in((d-1)/2,d)$ the new upper bounds are strictly better than all previous ones, for spheres when $d\ge4$ and for paraboloids when $d\ge3$, because the chosen $\kappa$ is always smaller than the previous $\kappa_1(0)$ or $\kappa_3(0)$.
Since $\tilde\kappa(\alpha,d)<\kappa(\alpha,d)$, the paraboloid admits examples with slower decay than the sphere at the same $\alpha$; the curvature of the hypersurface measurably changes the optimal rate.
Through the standard scheme connecting Fourier decay to distance sets, the new parabolic bounds cap the dimension threshold that this scheme can reach at $d/2+1/(d+2)$ for even $d$ and $d/2+1/(d+3)$ for odd $d\ge7$, with the small-dimensional values listed in Remark 1.6, so the scheme alone cannot settle the distance-set conjecture.

Reading between the lines

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

The same intermediate-dimension construction is local and depends only on nonvanishing curvature, so the exact parabolic rate at $\alpha\in[d-1,d)$ may transfer to any bounded curved hypersurface through the standard reduction; the paper only states it for the paraboloid.
The piecewise structure of $\tilde\kappa$ suggests the true rate for $\alpha\in((d-1)/2,d-1)$ may itself have several regimes; one could test sharpness numerically by running the Section 3 construction for the finitely many intervals in Theorem 1.2 at, say, $d=5$ or $d=6$.
A variational reading of the construction—choose $m$ to minimize the resulting exponent subject to $c_\alpha(\mu)\sim R^{\alpha-d}$—gives a general ansatz for upper bounds on other hypersurfaces; this is an editorial extrapolation, not a theorem in the paper.

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.

Desk Editor's Note

read the letter

Short version: this is a genuine advance, not a repackaging. Du takes the intermediate-dimension trick used for Schrödinger maximal estimates and applies it to the Fourier-average problem. The new upper bounds in Theorems 1.1 and 1.2 are strictly better than the previous best in the full stated ranges, and Corollary 1.4 gives the exact parabolic decay rate β_d(α,P^{d−1})=(d−1)α/d for α∈[d−1,d). That exact range was open, so the paper resolves a mini-Mattila problem in those dimensions.

The proofs are explicit and honest. The test measures are concrete products of boxes on rectangular lattices, and the c_α computations are done by hand in Lemmas 2.2 and 3.2. The case analysis is tedious but coherent; I checked the exponent balance in the main regimes and it works. The lower bound side of Corollary 1.4 is imported from Du–Zhang and Du–Guth–Ou–Wang–Wilson–Zhang, which are published independent theorems; citing them here is not circular because the upper bound is what is new.

The main soft spot is the spherical example in Section 2. It relies on a lattice point bound #Γ ≳ R^{κ(d−m−2)} for integer points on a sphere, which is cited to Fricker's book but not proved. That is standard, and I would not block a paper over it. Still, it means Theorem 1.1 inherits a mild external input. The parabolic result does not: the lattice there is rectangular, so the volume estimates are elementary. If you care about a completely self-contained proof, the spherical part is the only place with a gap.

I also like Remark 1.6: it quantifies what Mattila's scheme can and cannot do for Falconer's conjecture. The numbers he gets for the best possible threshold via that route are worth having in the literature, even if they are not a breakthrough by themselves.

Who is this for? Anyone working on Fourier decay of fractal measures, distance sets, or restriction estimates. The paper is clearly written and the numerology is laid out in small dimensions, which helps a reader check the statements. It deserves full peer review and, in my view, acceptance. The one thing I would ask a referee to scrutinize is the lattice point bound and the exactness of the volume lower bound in Section 2, plus the case analysis in part (b)–(f) of Theorem 1.2, which is easy to get wrong but appears correct.

Referee Report

0 major / 5 minor

Summary. The paper studies the average Fourier decay rate β_d(α,S), defined via the squared L2(S)-norm of μ̂(R·) bounded by c_α(μ)‖μ‖R^{-β} for all α-dimensional measures μ. It constructs explicit product-type 'intermediate dimension' measures and uses them to prove new upper bounds: Theorem 1.1 for the unit sphere and Theorem 1.2 for the truncated paraboloid. The main consequence is Corollary 1.4, which determines the exact value β_d(α,P^{d−1})=(d−1)α/d for d≥3 and d−1≤α<d, by matching the new upper bound from Theorem 1.2(a) with the lower bounds of [4,6]. The proofs are explicit calculations of volumes, c_α(μ) constants, and phase-stationarity; the parabolic construction is self-contained, while the spherical example uses a cited lattice-point bound.

Significance. If correct, the paper is a clear advance: it resolves the parabolic average Fourier decay rate on the interval d−1≤α<d for every dimension d≥3, and it improves the previously known upper bounds for both paraboloids and spheres throughout the intermediate range. The test-measure construction is simple and explicit, and the c_α(μ) computations are checkable line by line. The lower bounds used in Corollary 1.4 come from independent published theorems, so there is no circularity. The 'intermediate dimension' idea is likely to be useful for further problems in Fourier restriction and maximal estimates.

minor comments (5)

[Section 2, Eq. (2.9)] The bound #Γ ≳ R^{κ(d−m−2)} is quoted as 'well known' with reference [9], but no precise theorem or page is given; since Theorem 1.1 depends on it, please add a precise statement with a reference or a short proof sketch, and note explicitly that this input is only needed for the spherical theorems and not for Corollary 1.4.
[Section 3.2] In the proof of Lemma 3.2 the notations c_α(μ,r) and C_α(μ,r) are used interchangeably; please unify the notation.
[Section 3.2, Case I] In the summary for m≤α≤d−m, the term c_α(μ,R^{2κ−1}) arising from (3.44) is omitted from the displayed maximum; the domination by the other terms is true under the stated restrictions, but the 'combining' step should include one sentence justifying this omission.
[Section 1, after (1.2)] The arguments in κ4(m;d,α) and κ5(m;d,α) should be written in the same order as in (1.3), namely κ4(m;α,d) and κ5(m;α,d).
[Section 3, after (3.24)] The notation B^d_r and I_r is introduced but used only locally; the proof would be easier to follow if the notation were either used consistently throughout or removed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the upper bounds are explicit constructions and the exact parabolic result combines those constructions with independent cited lower bounds.

full rationale

The paper's central claims are upper bounds for spherical and parabolic Fourier decay rates. Each upper bound is proven by exhibiting an explicit alpha-dimensional measure mu (defined by chi_Lambda dx in (2.11) and (3.28)) and testing the dual estimate (2.5). The c_alpha(mu) computations in Lemmas 2.2 and 3.2 are carried out directly by summing the measure of balls over the relevant scales; the parameter kappa is then chosen so that c_alpha(mu) ~ R^{alpha-d}, and the resulting bound beta_d <= alpha-1+2*kappa is a consequence of the explicit volume estimates (2.9), (2.12), (3.26), and (3.29). No equation in this chain is loaded with the conclusion: the test measures are not defined in terms of beta_d, and the kappa formulas are not fitted to the claimed beta. Corollary 1.4 matches Theorem 1.2(a) with the lower bound cited from [4,6]; that lower bound is a stated prior theorem with its own proof, not an output of the present construction. The only external input with a merely cited proof is the lattice-point bound #Gamma >~ R^{kappa(d-m-2)} used in Section 2 for the spherical example; it is independent of the parabolic construction and is presented as a known fact (with reference [9]) rather than as a derived prediction. Thus no circular step, self-definitional reduction, or renamed fit is present.

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

The results rest on standard background results: a lattice point bound on spheres (cited, not proved), the duality reduction, published lower bounds from the literature (including the author's own papers), and the definition of α-dimensional measures. No invented entities are introduced, and no parameters are fitted to data; the construction parameters κ are determined by the requirement c_α(μ)∼R^{α−d}.

assumptions (4)

standard math Lattice point bound on spheres: for d−m≥2, #Γ ≳ R^{κ(d−m−2)} along a sequence R→∞, where Γ={ω∈S^{d−m−1}:R^κω∈2πZ^{d−m}}.
Used to lower-bound σ(Ω) in (2.9) for the spherical example; cited to [9] and not proved in this paper.
standard math Duality equivalence between the defining bound (1.1) and the L1(dμ) estimate (2.5).
First paragraph of Section 2; a standard functional-analytic reduction.
standard math Known lower bounds, in particular β_d(α,P^{d−1}) ≥ (d−1)α/d for α≥d/2, from the literature.
Used in Corollary 1.4 and in the introduction to state the prior state of the art; includes the author's own papers [4,6].
domain assumption Definition of α-dimensional measures and the normalization by c_α(μ) in (1.1).
The paper works entirely within Mattila's framework; the constructed measures must satisfy c_α(μ)∼R^{α−d}, verified in Lemmas 2.2 and 3.2.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Upper bounds for Fourier decay rates of fractal measures." pith.science (2026). https://pith.science/paper/WXCPCDLH

@misc{pith2026190805753,
  author       = {Pith},
  title        = {Pith review of: Upper bounds for Fourier decay rates of fractal measures},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WXCPCDLH}},
  note         = {Machine review of arXiv:1908.05753}
}

read the original abstract

For spherical and parabolic averages of the Fourier transform of fractal measures, we obtain new upper bounds on rates of decay by an "intermediate dimension" trick.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

15 extracted references · 14 canonical work pages

[1]

Barcel´ o, J.M

J.A. Barcel´ o, J.M. Bennett, A. Carbery and A. Ruiz and M. C. Vilela, Some special solutions of the Schr¨ odinger equation, Indiana Univ. Math. J. 56 (2007), no. 4, 1581-1593

work page 2007
[2]

Bourgain, A note on the Schr¨ odinger maximal function, J

J. Bourgain, A note on the Schr¨ odinger maximal function, J. Anal. Math. 130 (2016), 393-396

work page 2016
[3]

Pointwise convergence of the fractional Schr\"odinger equation in $\mathbb R^2$

C.H. Cho and H. Ko, Note on maximal estimates of generalized Schr¨ odinger equa tion, arXiv:1809.03246

work page Pith review arXiv
[4]

X. Du, L. Guth, Y. Ou, H. W ang, B. Wilson and R. Zhang, Weighted restriction estimates and application to Falconer distance set problem , Amer. J. Math. (to appear)

work page
[5]

X. Du, J. Kim, H. W ang, R. Zhang, Lower bounds for estimates of the Schr¨ odinger maximal function, Math. Res. Lett. (to appear)

work page
[6]

Du and R

X. Du and R. Zhang, Sharp L2 estimates of the Schr¨ odinger maximal function in higher dimensions, Ann. of Math. 189 (2019), no. 3, 837-861

work page 2019
[7]

K. J. Falconer, On the Hausdorﬀ dimensions of distance sets , Mathematika 32 (1985), no. 2, 206-212 (1986)

work page 1985
[8]

L. Guth, A. Iosevich, Y. Ou and H. W ang, On Falconer’s distance set problem in the plane , arXiv:1808.09346

work page Pith review arXiv

Show all 15 references

[9]

Fricker, Einf¨ uhrung in die Gitterpunktlehre, Birkh¨ auser (1982)

F. Fricker, Einf¨ uhrung in die Gitterpunktlehre, Birkh¨ auser (1982)

1982
[10]

Liu, An L2-identity and pinned distance problem , Geom

B. Liu, An L2-identity and pinned distance problem , Geom. Funct. Anal. 29 (2019), no. 1, 283-294

2019
[11]

Luc` a and K

R. Luc` a and K. Rogers, Average decay for the Fourier transform of measures with app lica- tions, J. Eur. Math. Soc. 21 (2019), 465-506

2019
[12]

Mattila, Spherical averages of Fourier transforms of measures with ﬁ nite energy; dimen- sions of intersections and distance sets , Mathematika 34 (1987), no

P. Mattila, Spherical averages of Fourier transforms of measures with ﬁ nite energy; dimen- sions of intersections and distance sets , Mathematika 34 (1987), no. 2, 207-228

1987
[13]

Mattila, Hausdorﬀ dimension, projections, and the Fourier transfor m., Publ

P. Mattila, Hausdorﬀ dimension, projections, and the Fourier transfor m., Publ. Mat. 48 (2004), no. 1, 3-48

2004
[14]

Mattila, Fourier Analysis and Hausdorﬀ Dimension , Cambridge Univ

P. Mattila, Fourier Analysis and Hausdorﬀ Dimension , Cambridge Univ. Press. https://doi.org/10.1017/CBO9781316227619

doi
[15]

W olﬀ, Decay of circular means of Fourier transforms of measures , Int

T. W olﬀ, Decay of circular means of Fourier transforms of measures , Int. Math. Res. Not. 10 (1999), 547-567. FOURIER DECAY RATES 19 Department of Mathematics, University of Maryland, College Park, MD E-mail address : xdu@math.umd.edu

1999

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Barcel´ o, J.M

J.A. Barcel´ o, J.M. Bennett, A. Carbery and A. Ruiz and M. C. Vilela, Some special solutions of the Schr¨ odinger equation, Indiana Univ. Math. J. 56 (2007), no. 4, 1581-1593

work page 2007

[2] [2]

Bourgain, A note on the Schr¨ odinger maximal function, J

J. Bourgain, A note on the Schr¨ odinger maximal function, J. Anal. Math. 130 (2016), 393-396

work page 2016

[3] [3]

Pointwise convergence of the fractional Schr\"odinger equation in $\mathbb R^2$

C.H. Cho and H. Ko, Note on maximal estimates of generalized Schr¨ odinger equa tion, arXiv:1809.03246

work page Pith review arXiv

[4] [4]

X. Du, L. Guth, Y. Ou, H. W ang, B. Wilson and R. Zhang, Weighted restriction estimates and application to Falconer distance set problem , Amer. J. Math. (to appear)

work page

[5] [5]

X. Du, J. Kim, H. W ang, R. Zhang, Lower bounds for estimates of the Schr¨ odinger maximal function, Math. Res. Lett. (to appear)

work page

[6] [6]

Du and R

X. Du and R. Zhang, Sharp L2 estimates of the Schr¨ odinger maximal function in higher dimensions, Ann. of Math. 189 (2019), no. 3, 837-861

work page 2019

[7] [7]

K. J. Falconer, On the Hausdorﬀ dimensions of distance sets , Mathematika 32 (1985), no. 2, 206-212 (1986)

work page 1985

[8] [8]

L. Guth, A. Iosevich, Y. Ou and H. W ang, On Falconer’s distance set problem in the plane , arXiv:1808.09346

work page Pith review arXiv