Canonical: https://knowlpedia.com/analysis/box-porosity/

# Box porosity

A discrete multiscale porosity condition expressed by an empty child in every occupied cube.

Fix an integer \(L\ge3\). Partition \([-1,1]^d\) at depth \(n\) into congruent
cubes of side length proportional to \(L^{-n}\). A set \(X\subseteq[-1,1]^d\)
is **box porous at scale \(L\) with depth \(n\)** if every depth-\(n\) cube
\(Q\) meeting \(X\) contains a depth-(n+1) child cube \(Q'\subset Q\) with
\(Q'\cap X=\varnothing\).

## Iterated form

If the condition holds at depths \(0,\ldots,N-1\), then at least one of the
\(L^d\) children is discarded at every occupied node. Consequently
\[
|X|\le 2^d(1-L^{-d})^N,
\]
where \(|X|\) denotes [Lebesgue measure](https://knowlpedia.com/measure-theory/lebesgue-measure/).

## Comparison with ball porosity

A set that is \(\nu\)-[porous on balls](https://knowlpedia.com/analysis/porosity-on-balls/) down to
scale \(h\) is box porous at all depths with \(L^{-n}\gtrsim h\), for a choice
\(L\asymp_d\nu^{-1}\). Box porosity is useful because the preceding counting
estimate turns geometric holes into a power-law volume bound.

## References

1. Rui Han and Wilhelm Schlag, “A higher-dimensional Bourgain–Dyatlov fractal uncertainty principle,” *Analysis & PDE* 13 (2020), 813–863. [DOI record](https://doi.org/10.2140/apde.2020.13.813).
2. Alex Cohen, “Fractal uncertainty in higher dimensions,” 2024. [arXiv record](https://arxiv.org/abs/2305.05022). Relevant: Appendix A.1.
