Oleg N. German,Nikolay G. Moshchevitin
Abstract:In this paper we give a simple proof of an inequality for intermediate Diophantine exponents obtained recently by W. M. Schmidt and L. Summerer.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to simplify and provide a proof of the recently proposed intermediate Diophantine exponent inequality by W.M. Schmidt and L. Summerer. In their previous work, Schmidt and Summerer studied the asymptotic behavior of specific geometric bodies and proved important inequalities by complex methods. These inequalities improved Jarník's well - known inequality regarding the relationship between the uniform and ordinary Diophantine exponents. However, their proof methods are very complex, involving Mahler's complex theory and a cumbersome analysis of special piecewise - linear functions.
The main contribution of this paper is to provide a simpler and more intuitive proof method, avoiding the use of complex theories and techniques, making this result easier to understand and verify. Specifically, the author utilizes a simple geometric observation (see Lemma 1) and simplifies the proof process by cleverly choosing lattices and paths. This method not only simplifies the original proof but also provides a new perspective for understanding certain key problems in Diophantine approximation.
### Specific Problem Description
1. **Background and Motivation**:
- The research objects are the $d$-dimensional unimodular lattice $\Lambda$ and the unit hypercube $B^d_\infty$.
- The transformed geometric body $G_tB^d_\infty$ is defined by the mapping $G_t:\mathbb{R}^d\rightarrow\mathbb{R}^d$.
- Schmidt and Summerer studied the asymptotic behavior of these geometric bodies with respect to the successive minima of a given lattice and related it to the classical Diophantine approximation problem.
2. **Main Results**:
- Proved important inequalities regarding the first and the $p$-th successive minima (see formulas (5) and (6)).
- These inequalities improved Jarník's relationship between the uniform and ordinary Diophantine exponents.
3. **Innovations**:
- Proposed a simple geometric lemma (Lemma 1) to simplify the proof process.
- Avoided using Mahler's complex theory, making the proof more concise and clear.
### Key Formulas
- **Mapping $G_t$**:
\[
G_t((z_1,\ldots,z_d)^T)=(t^{d - 1}z_1,t^{-1}z_2,\ldots,t^{-1}z_d)^T
\]
- **Schmidt - Summerer Index**:
\[
\psi_p(\Lambda,T)=\liminf_{s\rightarrow+\infty}\frac{\ln(\lambda_p(B(s)))}{s},\quad\psi_p(\Lambda,T)=\limsup_{s\rightarrow+\infty}\frac{\ln(\lambda_p(B(s)))}{s}
\]
- **Inequality (5)**:
\[
(1+\psi_p)\left(\frac{1}{n}-\psi_p\right)\leq(1+\psi_1)\left(\frac{1}{n}-\psi_p\right)
\]
- **Inequality (6)**:
\[
(1+\psi_d)\left(\frac{1}{n}-\psi_p\right)\leq(1+\psi_p)\left(\frac{1}{n}-\psi_p\right)
\]
Through these simplifications and innovations, the paper not only provides a more concise proof method but also provides new ideas and tools for further research.