Abstract:This work investigates the relation between the Fourier transform of a planar convex body and its geometric properties. The second matter of study is the quadratic discrepancy of planar convex bodies, and we present sharp results on its asymptotic behaviour. In particular, averages over intervals of rotations are considered.
What problem does this paper attempt to address?
### What problems does this paper attempt to solve?
This paper mainly explores the relationship between the Fourier transform of planar convex bodies and their geometric properties, and studies the quadratic discrepancy of planar convex bodies on rotation intervals. Specifically, the author focuses on the following two aspects:
1. **Relationship between Fourier transform and geometric properties**:
- Study how the Fourier transform of planar convex bodies reflects their geometric features. By introducing some geometric tools, such as chord length, the longest and shortest directional diameters, etc., the author establishes the connection between these geometric quantities and the attenuation of the Fourier transform.
- The specific results are reflected in Lemma 1.3, which gives the upper and lower bound estimates of the Fourier transform, and these estimates depend on the geometric parameters of the convex body.
2. **Study of quadratic discrepancy**:
- Study the quadratic discrepancy of planar convex bodies on different rotation intervals, especially the quadratic discrepancy of non - full rotation intervals. The quadratic discrepancy measures the uniformity of the distribution of point sets with respect to a given geometric shape.
- The author considers the cases of large - angle intervals and small - angle intervals respectively, and obtains different asymptotic behaviors. For a sufficiently large rotation interval, the asymptotic order of the quadratic discrepancy is \( N^{1/2} \); while for a sufficiently small rotation interval, the asymptotic order is \( N^{2/5} \).
In addition, the author also studies the polynomial orders between \( N^{1/2} \) and \( N^{2/5} \), and constructs appropriate planar convex bodies to verify these results.
### Key formulas
- Fourier transform estimate in Lemma 1.3:
\[
\kappa_4 \rho^{-2} \gamma_C(\theta, \rho^{-1})^2 \leq \int_0^1 \left| \widehat{1_{[\delta]C}}(\rho u(\theta)) \right|^2 d\delta \leq 2 \rho^{-2} \gamma_C(\theta, \rho^{-1})^2
\]
where \(\gamma_C(\theta, \lambda)\) is the maximum length of the chord of the convex body \(C\) at a distance of \(\lambda\) in the direction \(u(\theta)\).
- Semi - chord integral geometric result in Theorem 1.8:
\[
\lim_{\lambda \to 0} \frac{1}{2\lambda} \int_I |K_C^+(\theta, \lambda)|^2 d\theta = P_C(I)
\]
where \(P_C(I)\) is the partial perimeter of the convex body \(C\) corresponding to the angle interval \(I\) on its boundary.
- Asymptotic behavior of quadratic discrepancy:
- For large - angle intervals \( \psi_C < |I| \leq 2\pi \):
\[
\inf_{\#P = N} D_2(P, C, I) \asymp N^{1/2}
\]
- For small - angle intervals \( 0 < |I| < \psi_C < \pi \):
\[
\inf_{\#P = N} D_2(P, C, I) \asymp N^{2/5}
\]
These results not only deepen the understanding of the Fourier transform of planar convex bodies, but also provide a new perspective for studying the irregularity of point set distributions.