Abstract:In this paper we provide a detailed study on effective versions of the celebrated Bilu's equidistribution theorem for Galois orbits of sequences of points of small height in the $N$-dimensional algebraic torus, identifying the qualitative dependence of the convergence in terms of the regularity of the test functions considered. We develop a general Fourier analysis framework that extends previous results obtained by Petsche (2005), and by D'Andrea, Narváez-Clauss and Sombra (2017).
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to provide an effective version of Bilu's equidistribution theorem, especially for the Galois orbits of sequences of small - height points on N - dimensional algebraic tori. Specifically, through the Fourier analysis framework, the author studies the qualitative dependence of the convergence on the regularity of the test function. This research extends previous work on effective estimates and provides new insights and improvements in the multi - dimensional case.
### Background of the paper
- **Bilu's equidistribution theorem**: In 1997, Bilu proved that for a strict sequence \(\{\xi_k\}_{k \geq 1} \subset (\mathbb{Q}^\times)^N\), if \(\lim_{k \to \infty} h(\xi_k)=0\), then for any bounded and continuous function \(F: (\mathbb{C}^\times)^N \to \mathbb{C}\), we have:
\[
\lim_{k \to \infty} \int_{(\mathbb{C}^\times)^N} F \, d\mu_{S_k}=\int_{(\mathbb{C}^\times)^N} F \, d\mu_{(S^1)^N}.
\]
Here, \(\mu_{S_k}\) is the discrete probability measure on the Galois orbit \(S_k\) of \(\xi_k\), and \(\mu_{(S^1)^N}\) is the Haar probability measure on the unit polydisk.
- **Effective estimates**: Previous work mainly focused on effective estimates in the one - dimensional case, such as Petsche (2005) and D'Andrea, Narváez - Clauss, Sombra (2017), etc. These works assume that the test function has good regularity, usually some form of Lipschitz continuity.
### Goals of the paper
- **Extension to the multi - dimensional case**: The goal of this paper is to extend these results to the multi - dimensional case and provide a more generalized Fourier analysis framework.
- **Regularity and convergence**: Study how the regularity of the test function affects the convergence rate, especially when the test function is continuous but not Lipschitz continuous.
- **Effective bounds**: Provide specific bound estimates to better understand the effectiveness of the equidistribution theorem in practical applications.
### Main results
- **Theorem 2**: Gives an effective estimate under the regularity condition of the test function \(F\) in the Fourier space, in the specific form:
\[
\left|\int_{(\mathbb{C}^\times)^N} F \, d\mu_S-\int_{(\mathbb{C}^\times)^N} F \, d\mu_{(S^1)^N}\right|\leq 2\frac{G((8\pi h(\xi))^{-1})}{G((8\pi h(\xi))^{-1})}\sum_{n \in \mathbb{Z}^N}\int_{\mathbb{R}^N}|\hat{F}(n, t)|G(\|t\|_\infty) \, dt+\frac{1}{H((24 h_D(\xi))^{-1})}\sum_{n \in \mathbb{Z}^N\setminus\{0\}}|\hat{F}_0(n)|H(\|n\|_1).
\]
- **Theorem 5**: Considers the regularity of the logarithmic radial part of the test function in the physical space and the angular part in the Fourier space, in the specific form:
\[
\left|\int_{(\mathbb{C}^\times)^N} F \, d\mu_S-\int_{(\mathbb{C}^\times)^N} F \, d\mu_{(S^1)^N}\right|\leq\omega(2 h(\xi))+ \frac{1}{