[Paper Review] A group-theoretic viewpoint on Erdos-Falconer problems and the Mattila integral

Allan Greenleaf, Alex Iosevich+2 more|arXiv (Cornell University)|Jun 15, 2013

Geometric and Algebraic Topology7 references7 citations

TL;DR

This paper introduces a group-theoretic framework to study Erdős-Falconer-type problems on point configurations, specifically the distribution of congruent and similar $k$-dimensional simplices in $$\mathbb{R}^d$$. By analyzing measures induced by orthogonal and scaling-group actions, it establishes improved dimension thresholds—$\operatorname{dim}_\mathcal{H}(E) > \frac{dk+1}{k+1}$ for congruent simplices and $\operatorname{dim}_\mathcal{H}(E) > \frac{dk}{k+1}$ for similar simplices—under which the corresponding configuration sets have positive Lebesgue measure in their natural parameter spaces.

ABSTRACT

We obtain nontrivial exponents for Erd\H os-Falconer type problems. Let $T_k(E)$ denote the set of distinct congruent $k$-dimensional simplexes determined by $(k+1)$-tuples of points from $E$. We prove that there exists $s_0(d)s_0(d)$, then the ${k+1 \choose 2}$-dimensional Lebesgue measure of $T_k(E)$ is positive. Results were previously obtained for triangles in the plane \cite{GI12} and in higher dimensions \cite{GGIP12}. In this paper, we improve upon those exponents, using a group-theoretic method that sheds new light on the classical approach to these problems. The key to our approach is a group action perspective which leads to natural and effective formulae related to the classical Mattila integral.

Motivation & Objective

• To improve the known lower bounds on the Hausdorff dimension of a set $E \subset \mathbb{R}^d$ that guarantee the set of distinct congruent $k$-simplices $T_k(E)$ has positive $\binom{k+1}{2}$-dimensional Lebesgue measure.
• To extend this analysis to the set of distinct similar $k$-simplices $S_k(E)$, establishing improved dimension thresholds for positive $\binom{k+1}{2}-1$-dimensional measure.
• To unify and generalize the classical Mattila integral approach to the Falconer distance problem via group actions, particularly using orthogonal and scaling transformations.
• To demonstrate that the classical $\frac{d+1}{2}$ threshold for the Falconer distance problem can be recovered without stationary phase methods, using only geometric and harmonic analysis tools.
• To provide a new perspective on point configuration problems through the lens of group actions, revealing deeper structural connections in geometric measure theory.

Proposed method

• Define a measure $\nu_g$ on $\mathbb{R}^d$ via the action of the orthogonal group $\mathrm{O}(d)$, using a convolution-like construction $\int f(u - gv)\,d\mu(u)\,d\mu(v)$, which models the difference set $E - gE$.
• Introduce a pushforward measure $\nu$ on the space of distance tuples $\mathbf{t} = (|x^i - x^j|)_{i<j}$, representing the configuration space of $k$-simplices.
• Establish a sufficient condition for $\nu$ to have an $L^2$ density (hence positive Lebesgue measure) via the integrability of $\int_{\mathrm{O}(d)} \int_{\mathbb{R}^d} \nu_g^{k+1}(x)\,dx\,dg < \infty$.
• Extend the method to similarity classes by incorporating scaling parameters $a \in \mathbb{R}^+$, defining measures $\nu_{a,g}$ and requiring integrability of $\int_I \int_{\mathrm{O}(d)} \int_{\mathbb{R}^d} \nu_{a,g}^{k+1}(x)\,dx\,dg\,\frac{da}{a} < \infty$.
• Use the group action to re-derive the classical Mattila integral as a consequence of Plancherel’s theorem and invariance under $\mathrm{O}(d)$, avoiding stationary phase techniques.
• Apply Frostman’s lemma and $s$-energy estimates to bound the $L^2$-norm of the Fourier transform of $\mu$, linking the integrability condition to the Hausdorff dimension of $E$.

Experimental results

Research questions

• RQ1 What is the optimal lower bound on the Hausdorff dimension of a compact set $E \subset \mathbb{R}^d$ that ensures the set of distinct congruent $k$-simplices $T_k(E)$ has positive $\binom{k+1}{2}$-dimensional Lebesgue measure?
• RQ2 How does the group-theoretic framework improve upon prior results in the Erdős-Falconer problem for higher-dimensional simplices?
• RQ3 Can the classical Mattila integral, central to the Falconer distance problem, be derived without stationary phase methods using group actions and harmonic analysis?
• RQ4 What is the sharp threshold for the Hausdorff dimension of $E$ that guarantees the set of distinct similar $k$-simplices $S_k(E)$ has positive $\binom{k+1}{2}-1$-dimensional measure?
• RQ5 To what extent does the group action perspective reveal new structural insights into the geometry of point configurations in Euclidean space?

Key findings

• The paper establishes a new dimension threshold $\dim_{\mathcal{H}}(E) > \frac{dk+1}{k+1}$ for the set $T_k(E)$ of distinct congruent $k$-simplices to have positive $\binom{k+1}{2}$-dimensional Lebesgue measure.
• For the case $k=d=2$, the threshold improves to $\dim_{\mathcal{H}}(E) > \frac{8}{5}$, which is sharper than previous results.
• The paper derives a similar improved threshold $\dim_{\mathcal{H}}(E) > \frac{dk}{k+1}$ for the set $S_k(E)$ of distinct similar $k$-simplices to have positive $\binom{k+1}{2}-1$-dimensional Lebesgue measure.
• The authors show that the classical Mattila integral, which controls the Falconer distance problem, can be re-derived using group-theoretic and Plancherel-based arguments, bypassing the need for stationary phase methods.
• The method reveals that the $\frac{d+1}{2}$ threshold for the Falconer problem is sharp under the new framework, and the integrability condition is satisfied when $\dim_{\mathcal{H}}(E) > \frac{d+1}{2}$.
• The paper poses an open problem: whether the number of non-congruent triangles with vertices on three lattice spheres of radius $\approx q$ is $O(q^{3-\delta})$ for some $\delta > 0$, which would imply a nontrivial improvement in the dimension threshold for $k=3$, $d=3$.