How smooth are restrictions of Besov functions?

Basic Information

• Paper ID: 2509.07420
• Title: How smooth are restrictions of Besov functions?
• Author: Julien Brasseur
• Classification: math.FA (Functional Analysis), math.AP (Analysis of Partial Differential Equations), math.CA (Classical Analysis)
• Publication Date: October 15, 2025 (arXiv v3)
• Paper Link: https://arxiv.org/abs/2509.07420

Abstract

This paper refines the characterization of restriction properties for Besov spaces. The author previously proved that Besov spaces satisfy the restriction property if and only if $q \leq p$. Specifically, when $p < q$, one can always construct functions $f \in B_{p,q}^s(\mathbb{R}^N)$ such that $f(\cdot,y) \notin B_{p,q}^s(\mathbb{R}^d)$ for almost all $y \in \mathbb{R}^{N-d}$. This paper demonstrates that partial mappings actually belong to generalized smoothness Besov spaces $B_{p,q}^{(s,\Psi)}(\mathbb{R}^d)$, provided the function $\Psi$ satisfies a summability condition involving $p$ and $q$. The paper completes this picture by proving the sharpness of this characterization.

Research Background and Motivation

Core Problem

The fundamental question addressed in this research is: How much smoothness do partial mappings (restrictions) of Besov functions actually possess?

Significance of the Problem

1. Theoretical Significance: Besov spaces are fundamental tools in harmonic analysis and partial differential equation theory, and their restriction properties directly influence the development of function space theory.
2. Counterintuitive Phenomenon: Most classical function spaces (such as Sobolev spaces and Triebel-Lizorkin spaces) satisfy the restriction property, but Besov spaces fail to do so under certain parameter conditions, which is surprising.
3. Role of Parameter $q$: Typically, the parameter $q$ plays a minor role in Besov space properties, but it is crucial for restriction properties.

Limitations of Existing Research

The author previously proved in reference 3:

• When $p < q$, a "pathological" phenomenon exists: $f \in B_{p,q}^s(\mathbb{R}^N)$ but $f(\cdot,y) \notin B_{p,\infty}^s(\mathbb{R}^d)$
• When $q \leq p$, the restriction property holds
• Partial mappings belong to the generalized smoothness space $B_{p,p}^{(s,\Psi)}(\mathbb{R}^d)$

However, a key question remained unresolved: Is this characterization sharp?

Core Contributions

1. Refined Complete Characterization of Besov Space Restriction Properties: Proved that the summability condition (1.2) in Theorem 1.1 is both necessary and sufficient.
2. Established Sharpness Results: Demonstrated that when $\Psi$ does not satisfy condition (1.2), there exist functions whose partial mappings do not belong to $B_{p,\infty}^{(s,\Psi)}(\mathbb{R}^d)$.
3. Provided Precise Critical Conditions: Gave necessary and sufficient conditions for the "compensated restriction property" to hold.
4. Extended the Theoretical Framework: Generalized results from admissible functions to slowly varying functions.

Detailed Methodology

Problem Definition

Study the smoothness of partial mappings $f(\cdot,y): \mathbb{R}^d \to \mathbb{R}$ of functions $f$ in Besov space $B_{p,q}^s(\mathbb{R}^N)$, where:

• Input: $f \in B_{p,q}^s(\mathbb{R}^N)$, parameters $1 \leq d < N$, $0 < p < q \leq \infty$, $s > \sigma_p$
• Output: The optimal function space to which the partial mapping $f(\cdot,y)$ belongs
• Constraint: $\sigma_p = N(\frac{1}{p} - 1)_+$

Core Theoretical Framework

1. Generalized Smoothness Besov Spaces

Definition 2.3: For a slowly varying function $\Psi$, the generalized smoothness Besov space $B_{p,q}^{(s,\Psi)}(\mathbb{R}^N)$ consists of functions $f \in L^p(\mathbb{R}^N)$ satisfying:

$[f]_{B_{p,q}^{(s,\Psi)}(\mathbb{R}^N)} = \left(\int_0^1 \frac{\sup_{|h| \leq t} \|\Delta_h^M f\|_{L^p(\mathbb{R}^N)}^q}{\Psi(t)^q} \frac{dt}{t^{1+sq}}\right)^{1/q} < \infty$

2. Key Summability Condition

Condition (1.2): $\left(\sum_{j=0}^{\infty} \Psi(2^{-j})^{\kappa}\right)^{1/\kappa} < \infty$, where $\frac{1}{\kappa} = \frac{1}{p} - \frac{1}{q}$

Main Theorems

Theorem 1.1 (Positive Result)

If $\Psi$ satisfies condition (1.2), then for any $f \in B_{p,q}^s(\mathbb{R}^N)$: $f(\cdot,y) \in B_{p,p}^{(s,\Psi)}(\mathbb{R}^d) \text{ for a.e. } y \in \mathbb{R}^{N-d}$

Theorem 1.3 (Sharpness Result)

If $\Psi$ does not satisfy condition (1.2), then there exists $f \in B_{p,q}^s(\mathbb{R}^N)$ such that: $f(\cdot,y) \notin B_{p,\infty}^{(s,\Psi)}(\mathbb{R}^d) \text{ for a.e. } y \in \mathbb{R}^{N-d}$

Technical Innovations

1. Improved Construction Techniques

• Lemma 3.1: Provides precise discrimination criteria for convergence of positive series
• Lemma 3.2: Establishes sequence rearrangement techniques that preserve mean values while achieving specific distributional properties
• Lemma 3.3: Constructs sequences $(λ_{j,k})$ with special properties

2. Key Construction Ideas

Through constructing special atomic decompositions: $f(x) = \sum_{j,k \in \mathbb{N}} λ_{j,k} 2^{-j(s-N/p)} ψ(2^j x - m_{j,k})$

where $ψ$ is a carefully designed smooth function and $λ_{j,k}$ are coefficients satisfying Lemma 3.3.

Experimental Setup

Theoretical Verification Framework

This is a pure mathematical theory paper with no numerical experiments. Instead, it verifies theoretical results through rigorous mathematical proofs.

Proof Strategy

1. Positive Results: Using Littlewood-Paley decomposition and atomic decomposition techniques
2. Negative Results: Through counterexample construction, utilizing difference representations
3. Sharpness: Proving necessity and sufficiency of critical conditions

Main Results

Complete Characterization Theorem

Main Result: For parameters $s, p, q$ satisfying condition (1.1), the "compensated restriction property" $f \in B_{p,q}^s(\mathbb{R}^N) \Rightarrow f(\cdot,y) \in B_{p,\min(p,q)}^{(s,\Psi)}(\mathbb{R}^d) \text{ for a.e. } y \in \mathbb{R}^{N-d}$ holds if and only if condition (1.2) holds.

Technical Breakthroughs

1. Elimination of Extra Assumptions: Generalized results from admissible functions to slowly varying functions
2. Unified Framework: Unified both cases $q \leq p$ and $p < q$ under the same framework
3. Optimality: Proved the sharpness of the characterization, which cannot be further improved

Related Work

Historical Development

1. Johnsen (2000): Discovered the role of parameter $q$ in studying trace operators on critical Besov spaces $B_{p,q}^{n/p-n+1}(\mathbb{R}^n)$
2. Caetano & Haroske (2015): Observed the difference between $q \leq p$ and $p < q$ in trace embeddings on fractal sets
3. Author's Previous Work 3: Established fundamental restriction property theory

Advancement in This Paper

• Completed the final piece of the theoretical picture
• Provided sharp necessary and sufficient conditions
• Extended the applicable scope

Conclusions and Discussion

Main Conclusions

1. Complete Characterization: Condition (1.2) completely characterizes the "compensated restriction property" of Besov spaces
2. Critical Role of Parameter $q$: In the case $0 < p < 1$, parameter $q$ plays a decisive role
3. Theoretical Completeness: This characterization is optimal and cannot be further improved

Theoretical Significance

• Reveals deep structural properties of Besov spaces
• Provides new perspectives for function space theory
• Perfects the theoretical foundations of harmonic analysis

Future Directions

1. Generalization to other function spaces (such as Triebel-Lizorkin spaces)
2. Study of analogous properties on fractal domains
3. Exploration of applications in partial differential equations

In-Depth Evaluation

Strengths

1. Theoretical Completeness: Provides a complete solution to the problem
2. Technical Innovation: Ingenious construction techniques with rigorous proofs
3. Unification: Incorporates different cases into a unified framework
4. Practical Value: Provides important tools for related research

Technical Highlights

1. Application of Lemma 3.1: Clever use of series convergence discrimination methods
2. Sequence Rearrangement Techniques: Achieves special distributions while preserving averaging properties
3. Atomic Decomposition Construction: Precisely controls local and global properties of functions

Theoretical Impact

• Perfects an important aspect of Besov space theory
• Provides a paradigm for studying restriction properties of function spaces
• May inspire similar research on other spaces

Applicable Scenarios

1. Partial Differential Equation Theory: Regularity studies of boundary value problems
2. Harmonic Analysis: Embedding and restriction theory of function spaces
3. Fractal Geometry: Function space theory on fractal sets

References

This paper primarily references the following key works:

• 3 J. Brasseur: On restrictions of Besov functions (2018) - Author's previous foundational work
• 11 P. Mironescu, E. Russ, Y. Sire: Lifting in Besov spaces (2020) - Related lifting property research
• 13 H. Triebel: The Structure of Functions (2001) - Classical reference for Besov space theory

This paper concludes the study of Besov space restriction properties, providing a complete and precise theoretical characterization, representing an important advance in function space theory.