Abstract:In this paper, we answer a question of Cai-Hambrook in (arXiv$\colon$ <a class="link-https" data-arxiv-id="2403.19410" href="https://arxiv.org/abs/2403.19410">2403.19410</a>). Furthermore, we compute the Fourier dimension of the multiplicative $\psi$-well approximable set $$M_2^{\times}(\psi)=\left\{(x_1,x_2)\in [0,1]^{2}\colon \|qx_1\|\|qx_2\|<\psi(q) \text{ for infinitely many } q\in \N\right\},$$ where $\psi\colon\N\to [0,\frac{1}{4})$ is a positive function satisfying $\sum_q\psi(q)\log\frac{1}{\psi(q)}<\infty.$ As a corollary, we show that the set $M_2^{\times}(q\mapsto q^{-\tau})$ is non-Salem for $\tau>1.$
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to calculate the Fourier dimension of a specific set in multiplicative Diophantine approximation and prove whether this set is a non - Salem set. Specifically, the paper focuses on the Fourier dimension of the following set:
\[
M^{\times}_2(\psi) = \left\{ (x_1, x_2) \in [0, 1]^2 : \|qx_1\| \|qx_2\| < \psi(q) \text{ for infinitely many } q \in \mathbb{N} \right\}
\]
where $\psi: \mathbb{N} \to [0, \frac{1}{4})$ is a positive function satisfying $\sum_{q} \psi(q) \log \frac{1}{\psi(q)} < \infty$.
### Main problems
1. **Calculating the Fourier dimension**: The main objective of the paper is to calculate the Fourier dimension of the set $M^{\times}_2(\psi)$.
- **Definition of Fourier dimension**: For a Borel set $E \subseteq \mathbb{R}^n$, its Fourier dimension is defined as:
\[
\dim_F E = \sup \left\{ s \in [0, n] : \exists \mu \in M(E) \text{ such that } |\hat{\mu}(\xi)| \ll_s (1 + |\xi|)^{-s/2} \right\}
\]
where $M(E)$ represents all Borel probability measures that take full measure on $E$.
2. **Proving non - Salem set**: As a corollary, the paper proves that when $\psi(q) = q^{-\tau}$ and $\tau > 1$, the set $M^{\times}_2(\psi)$ is a non - Salem set.
- **Definition of Salem set**: If the Fourier dimension of a set is equal to its Hausdorff dimension, then the set is called a Salem set. Otherwise, it is called a non - Salem set.
### Solution
The paper calculates the Fourier dimension by introducing a new parameter $\tau(\psi, Q)$, where:
\[
\tau(\psi, Q) = \inf \left\{ s \in [0, 1] : \sum_{q \in Q} q^{-s} (\psi(q))^s < \infty \right\}
\]
Finally, the conclusion is:
\[
\dim_F M^{\times}_2(\psi, Q) = 2 \tau(\psi, Q)
\]
In particular, when $\psi(q) = q^{-\tau}$ and $\tau > 1$, the paper proves that:
\[
\dim_H M^{\times}_2(\psi) = \frac{\tau + 3}{\tau + 1}, \quad \dim_F M^{\times}_2(\psi) = \frac{4}{\tau + 2}
\]
Since $\frac{4}{\tau + 2} < \frac{\tau + 3}{\tau + 1}$, so $M^{\times}_2(\psi)$ is a non - Salem set.
### Summary
This paper solves a problem proposed by Cai - Hambrook by calculating the Fourier dimension of the multiplicative Diophantine approximation set, and further proves that this set is a non - Salem set. This result deepens the understanding of the theory of multiplicative Diophantine approximation, especially in the cross - field of Fourier analysis and fractal geometry.