Tangency counting for well-spaced circles

Basic Information

• Paper ID: 2504.14118
• Title: Tangency counting for well-spaced circles
• Authors: Dominique Maldague (University of Cambridge & UCLA), Alexander Ortiz (Rice University)
• Classification: math.CA (Classical Analysis and ODEs)
• Publication Date: October 14, 2025
• Paper Link: https://arxiv.org/abs/2504.14118

Abstract

In the late 1990s, Tom Wolff introduced the circle tangency counting problem in his expository article on the Kakeya conjecture. For sets of well-spaced circles, this paper breaks through the $N^{3/2}$ barrier, proving that a set consisting of $N$ well-spaced circles has at most $N^{25/18+\varepsilon}$ tangency points. The circle tangency problem can be related to incidence problems between points and rays in $\mathbb{R}^3$. To this end, the authors introduce a stopping-time argument to extract maximal information about well-spaced points from a refined decoupling theorem for light cones in $\mathbb{R}^3$, thereby obtaining sharp bounds on the number of $\mu$-rich tangency rectangles.

Research Background and Motivation

Problem Background

1. Circle Tangency Counting Problem: This problem was introduced by Tom Wolff in 1999, asking for an estimate of the number of pairs of internally tangent circles in a set of $N$ circles in the plane, where no three circles are tangent at the same point.
2. Connection to the Unit Distance Problem: This problem is a cousin of the famous Erdős unit distance problem in $\mathbb{R}^3$. By mapping circles $(z,r)$ to points in $\mathbb{R}^3$, the tangency condition transforms into the quadratic constraint $|z-z'|^2 = |r-r'|^2$.
3. Historical Bounds:
   • Traditional methods (cellular partitioning) yield $O_\varepsilon(N^{3/2+\varepsilon})$ bounds
   • Ellenberg-Solymosi-Zahl improved this to $O(N^{3/2})$
   • For the unit distance problem, Zahl obtained $O_\varepsilon(N^{295/197+\varepsilon})$

Research Motivation

1. Breaking the $3/2$ Barrier: The $N^{3/2}$ bound has been difficult to improve for a long time; this paper is the first to break this barrier in the case of well-spaced circles.
2. Theoretical Importance: This problem is closely related to important geometric measure theory problems such as the Kakeya conjecture and the dimension of Besicovitch-Robinson-Kaufman sets.
3. Methodological Innovation: Introduction of new Fourier analysis techniques, particularly refined decoupling theory and stopping-time algorithms.

Core Contributions

1. Breakthrough Bound: For well-spaced circles, proves an upper bound of $N^{25/18+\varepsilon}$, breaking through the $N^{3/2}$ barrier for the first time.
2. New Technical Framework:
   • Introduction of stopping-time algorithm to extract frequency information
   • Development of refined decoupling theorem for light cones
   • Establishment of duality between tangency rectangles and lightplanks
3. Sharpness Results: Construction of random examples proving that the main results are sharp in a certain sense.
4. Optimal Bounds for Continuous Version: For the $\mu$-rich tangency rectangle problem, obtains the sharp estimate $\mu^{4/3}|R_{\mu,\tau}| \leq A_\varepsilon|X|^{4/3+\varepsilon}$.

Detailed Methodology

Problem Formulation

Problem 1 (Discrete Tangency Counting): Given a set $\mathcal{C}$ of $N$ circles in the plane such that no three circles are tangent at the same point, estimate the cardinality of the set of internally tangent circle pairs: $T_{\text{pair}}(\mathcal{C}) = \{(C,C') \in \mathcal{C}^2 : C,C' \text{ are internally tangent}\}$

Problem 2 (Tangency Point Counting): Estimate the cardinality of the set of tangency points: $T(\mathcal{C}) = \{z \in \mathbb{R}^2 : \text{at least two circles are tangent at } z\}$

Well-spaced Condition

Definition 1.1: If $Q \subset \mathbb{R}^n$ is a unit cube, then a subset $X \subset Q$ is well-spaced if and only if $X$ is $\rho$-separated (for some $\rho \in (0,1)$) and $X$ is maximal with respect to this property. In particular, $c_n\rho^{-n} \leq |X| \leq C_n\rho^{-n}$.

Core Technical Framework

1. Lifting Procedure

Maps circle center-radius pairs $(z,r)$ to points in $\mathbb{R}^3$, where the tangency condition $|z-z'|^2 = |r-r'|^2$ is equivalent to two points lying on the same ray.

2. Lightplank Duality

• Tangency Rectangles: $\delta \times \sqrt{\delta}$ rectangles $\Omega$
• Lightplanks: Corresponding $\sim 1 \times \sqrt{\delta} \times \delta$ light cone slabs
• $\mu$-rich Property: The $C\delta$-neighborhood of a lightplank contains at least $\mu$ points

3. Stopping-time Algorithm

For a function $f = \sum_{P \in P_\mu} \varphi_P$ (where $\varphi_P$ is a smooth approximation of the lightplank $P$), define multi-scale averages:

$A_k f := \sum_\theta f_\theta * \tilde{\omega}_{0,\theta} * \tilde{\omega}_{1,\theta} * \cdots * \tilde{\omega}_{k,\theta}$

Stopping-time Rule: Define $k(x)$ as the smallest $k$ such that $A_{k-1}f(x) > R^\delta A_k f(x)$

4. Refined Decoupling Theorem

Theorem 3.1 (Refined Decoupling for Light Cones): For each $\varepsilon > 0$, there exists $C_\varepsilon$ such that for Schwartz functions $g_\theta$ with Fourier support in $\theta$:

$\int_{U_\beta \cap B_R} \left|\sum_\theta g_\theta\right|^6 \leq C_\varepsilon R^\varepsilon \beta^2 \sum_\theta \int_{\mathbb{R}^3} |g_\theta|^2$

where $U_\beta := \{x \in \mathbb{R}^3 : \beta \leq \sum_\theta |g_\theta(x)|^2 * w_{M,\theta} \leq 2\beta\}$.

Main Proof Strategy

1. High-Low Frequency Analysis: Decompose the function $f$ by frequency, applying different estimation techniques to different frequency ranges.
2. Stopping-time Analysis: Use the stopping-time algorithm to determine the dominant frequency components near each point $x \in X$.
3. Local Constancy Property: Utilize the property that $H_k f$ is approximately constant on $\rho_{k-1}$-balls.
4. Refined Decoupling Application: Apply the refined decoupling theorem within the determined frequency ranges.

Experimental Setup

Theoretical Verification

This is primarily a theoretical work, with results verified through rigorous mathematical proofs rather than numerical experiments.

Sharpness Construction

Theorem 4.1: Constructs random well-spaced sets to prove the sharpness of the main results. For $R^\varepsilon \leq \rho \leq R^{1/2}$, with probability at least 0.9, there exists a random set $X \subset [0,R]^3$ satisfying:

• $|X| \sim R^{3+\varepsilon}\rho^{-3}$
• Each $\rho$-cube contains at most $R^\varepsilon$ points
• Each lightplank contains $\sim R^{3/2+\varepsilon}\rho^{-3}$ points

Experimental Results

Main Results

Theorem 1.1: For a well-spaced set $X \subset [0,1]^2 \times [1,2]$ and corresponding circle set $\mathcal{C}_X$: $|T_{\text{pair}}(\mathcal{C}_X)| \leq A_\varepsilon |X|^{25/18+\varepsilon}$

Theorem 1.2 (Continuous Version): For well-spaced $X$ and $\mu \geq 1$, $\tau \in [|X|^{-1/3}, 1)$: $\mu^{4/3}|R_{\mu,\tau}| \leq A_\varepsilon |X|^{4/3+\varepsilon}$

Comparison with Existing Results

• Traditional Bound: $O(N^{3/2})$
• This Paper's Bound: $O(N^{25/18+\varepsilon}) = O(N^{1.388...+\varepsilon})$
• Improvement: Exponent reduced from $1.5$ to approximately $1.388$

Sharpness Verification

The main results are shown to be essentially optimal for well-spaced configurations through probabilistic constructions.

Related Work

Historical Development

1. Wolff (1999): Introduced the problem, gave $O_\varepsilon(N^{3/2+\varepsilon})$ bounds
2. Ellenberg-Solymosi-Zahl (2016): Improved to $O(N^{3/2})$
3. Zahl (2019): Obtained $O_\varepsilon(N^{295/197+\varepsilon})$ for the unit distance problem

Related Techniques

1. Cellular Partitioning: Traditional combinatorial geometry methods
2. Polynomial Partitioning: Algebraic geometry methods
3. Decoupling Theory: Orthogonality measurements in Fourier analysis

Application Areas

1. Kakeya Conjecture: Related to geometric measure theory of direction sets
2. Restriction Theory: Fundamental problems in harmonic analysis
3. Geometric Incidence Theory: Study of incidences among geometric objects

Conclusions and Discussion

Main Conclusions

1. First breakthrough of the $N^{3/2}$ barrier for circle tangency counting in the well-spaced setting
2. Establishment of a new technical framework combining stopping-time algorithms and refined decoupling theory
3. Obtainment of sharp bounds for the continuous version of the problem

Limitations

1. Well-spaced Restriction: Results apply only to well-spaced circle sets
2. Technical Limitations: Cases with $\tau \in (0, |X|^{-1/3})$ require new ideas
3. General Case: The $N^{3/2}$ bound remains unbroken for arbitrary circle sets

Future Directions

1. Extending Scope: Generalize results to more general circle sets
2. Improving Bounds: Further reduce the exponent, approaching the conjectured $N^{4/3}$
3. Related Problems: Apply techniques to other geometric incidence problems

In-depth Evaluation

Strengths

1. Major Breakthrough: First breakthrough of the long-standing $N^{3/2}$ barrier in the field
2. Technical Innovation:
   • Introduction of stopping-time algorithm demonstrates application of probabilistic ideas in harmonic analysis
   • Development of refined decoupling theory advances restriction theory
   • Sophisticated application of high-low frequency analysis
3. Theoretical Depth: Proof techniques are complex and insightful
4. Sharpness: Construction of examples proves results are essentially optimal

Weaknesses

1. Limited Applicability: Applies only to well-spaced configurations
2. Technical Complexity: Extremely complex proofs may limit generalizability of methods
3. Practical Applications: Primarily theoretical results; practical applications remain to be explored

Impact

1. Theoretical Contribution: Provides new technical tools for geometric incidence theory
2. Methodological Value: The combination of stopping-time algorithms and refined decoupling may apply to other problems
3. Inspirational Significance: Provides insights for breaking other long-standing bounds

Applicable Scenarios

1. Theoretical Research: Geometric measure theory, harmonic analysis, combinatorial geometry
2. Related Problems: Unit distance problem, Kakeya conjecture, restriction problems
3. Technical Generalization: Other problems requiring refined frequency analysis

References

This paper cites important literature in the field, including:

• Wolff's foundational work 21, 22
• Improvements by Ellenberg-Solymosi-Zahl 7
• Zahl's breakthrough on the unit distance problem 23
• Bourgain-Demeter's decoupling theory 2
• Related geometric incidence theory literature 4, 10, 11, 14

This paper represents significant progress in geometric incidence theory, breaking through a long-standing barrier through innovative techniques and laying the foundation for further development in the field.