Dimensions and dimension spectra of non-autonomous iterated function systems

Jun Jie Miao, Tianrui WangJournal of Mathematical Analysis and Applications 2025 · arXiv

TL;DR

This paper introduces and studies "intermediate dimension spectra" for a class of fractal sets called non-autonomous conformal sets. These new dimensions provide a unifying framework that bridges the gap between traditional Hausdorff and box-counting dimensions, offering a more nuanced understanding of fractal geometry.

What Problem Does This Paper Address?

Fractal geometry often relies on dimension measures like the Hausdorff dimension and the box-counting dimension to quantify the complexity of irregular shapes. While these dimensions are identical for some simple fractal sets (like self-similar sets satisfying the open set condition), they can differ significantly for more complex sets, such as self-affine or Moran sets. This discrepancy arises because Hausdorff dimension allows for coverings of widely varying scales, while box-counting dimension uses coverings of uniform scale. This paper addresses this by developing a more general theory of "intermediate dimensions" that smoothly interpolates between Hausdorff and box-counting dimensions. Specifically, it focuses on non-autonomous iterated function systems (NAIFS), which are a generalization of standard iterated function systems, and their attractors, known as non-autonomous conformal sets. The challenge is to extend the understanding of dimension theory to these more complex, non-autonomous fractal structures.

Methods

The core of this paper lies in defining and calculating intermediate dimension spectra for non-autonomous conformal sets. The authors introduce "upper and lower topological pressures" (denoted as $P(t, \theta)$ and $\underline{P}(t, \theta)$) which are functions of a parameter $t$ and a "level" $\theta$ (ranging from 0 to 1). These pressure functions are then used to define "jump points" ($s^0$ and $s^e$) which correspond to the upper and lower intermediate dimensions. The paper also defines simpler "pressure functions" ($P(t)$ and $\underline{P}(t)$) that lead to the box and packing dimensions ($s^*$). A significant portion of the work involves proving theorems that relate these pressure functions and their jump points to the various dimensions of the non-autonomous conformal sets, often under specific conditions like the open set condition (OSC) or the strong separation condition (SSC). The proofs often involve intricate limit arguments and the use of properties of iterated function systems, including bounded distortion and the concept of "cut sets."

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • Unifying Dimensions: The introduction of intermediate dimension spectra provides a continuous spectrum of dimensions that elegantly bridges the gap between Hausdorff and box-counting dimensions. This offers a more refined way to characterize the complexity of fractal sets.
  • Power of Pressure Functions: The paper demonstrates the crucial role of topological pressure functions in determining fractal dimensions. By analyzing the behavior of these functions, particularly their jump points, the authors can precisely calculate various dimension values.
  • Generalization to Non-Autonomy: The work extends existing dimension theory from autonomous iterated function systems to their non-autonomous counterparts, which are more general and can exhibit richer geometric behaviors.
  • Conditions Matter: The paper highlights how specific conditions on the iterated function systems (like the open set condition, strong separation condition, or cone condition) significantly impact the resulting fractal dimensions and the applicability of theoretical results.

Key Results

  • Intermediate Dimension Formulas: The paper provides explicit formulas for the upper and lower intermediate dimensions of non-autonomous finite conformal sets satisfying the open set condition and a specific condition on contraction ratios (Theorem 2.1).
  • Box and Packing Dimension Formulas: It establishes that the box and packing dimensions of these sets are equal to a critical value $s^*$ derived from the pressure function (Theorem 2.2).
  • Hausdorff Dimension Results: Under certain conditions (like the open set condition and the condition $L^d(\partial J) = 0$), the Hausdorff dimension is shown to be equal to $s^*$ (Theorem 2.3 and Corollary 2.4).
  • Infinite Systems: The paper extends these results to non-autonomous infinite conformal iterated function systems, providing dimension estimates under specific assumptions (Theorem 2.5).
  • Relaxed Conditions: The authors show that some results can be generalized by relaxing certain conditions, such as replacing condition (1.11) with (1.13) for Hausdorff dimensions.

Limitations and Future Work

  • Complexity of Infinite Systems: While the paper extends results to infinite non-autonomous conformal iterated function systems, these cases are noted as being more "awkward to study." Further work might explore these more general infinite systems.
  • Specific Conditions: Many of the precise dimension formulas are derived under specific conditions (e.g., open set condition, cone condition, specific limits on contraction ratios). Investigating how these dimensions behave when these conditions are not met would be a natural next step.
  • General Non-Autonomous Systems: The paper focuses on "conformal" non-autonomous systems. Extending the framework to more general non-autonomous iterated function systems (e.g., affine) could be a valuable direction.
  • Numerical Exploration: While the paper provides theoretical results, exploring these dimensions numerically for specific examples could offer deeper intuition and validation.

Summary updated

Abstract

Non-autonomous iterated function systems are a generalization of iterated function systems. If the contractions in the system are conformal mappings, it is called a non-autonomous conformal iterated function system, and its attractor is called a non-autonomous conformal set. In this paper, we study intermediate dimension spectra of non-autonomous conformal sets which provide a unifying framework for Hausdorff and box-counting dimensions. First, we obtain the intermediate dimension spectra formula of non-autonomous conformal sets by using upper and lower topological pressures. As a consequence, we obtain simplified forms of their Hausdorff, packing and box dimensions. Finally, we explore the Hausdorff dimensions of the non-autonomous infinite conformal iterated function systems which consists of countably many conformal mappings at each level, and we provide the Hausdorff dimension formula under certain conditions.
BibTeX
@article{miao2025dimensions,
  author = {Jun Jie Miao and Tianrui Wang},
  title = {{Dimensions and dimension spectra of non-autonomous iterated function systems}},
  journal = {Journal of Mathematical Analysis and Applications},
  year = {2025},
}
Figure