Gauss lattices and complex continued fractions

Nicolas Chevallier
DOI: https://doi.org/10.48550/arXiv.2101.05480
2021-10-04
Abstract:Our aim is to find a complex continued fraction algorithm finding all the best Diophantine approximations to a complex number. Using the sequence of minimal vectors in a two dimensional lattice over Gaussian integers, we obtain an algorithm defined on a submanifold of the space of unimodular two dimensional Gauss lattices. This submanifold is transverse to the diagonal flow. Thanks to the correspondence between minimal vectors and best Diophantine approximations, the algorithm finds all the best approximations to a complex number. A byproduct of the algorithm is the best constant for the complex version of Dirichlet Theorem about approximations of complex numbers by quotients of Gaussian integers.
Number Theory
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to construct a complex continued fraction algorithm in order to find the best Diophantine approximations of complex numbers. Specifically, the author uses the sequence of minimal vectors of lattices on the two - dimensional Gaussian integer ring to define an algorithm acting on a specific manifold. This manifold is a sub - manifold in the SL(2, C)/SL(2, Z[i]) space that is transverse to the diagonal flow. Through the correspondence between the minimal vectors and the best Diophantine approximations, the proposed algorithm can find the best approximations of all complex numbers. ### Main Problem Summary 1. **Finding the Best Diophantine Approximations** - The author's goal is to construct a complex continued fraction algorithm for finding all the best Diophantine approximations of complex numbers. 2. **The Space Defined by the Algorithm** - The algorithm is defined on a specific sub - manifold in the SL(2, C)/SL(2, Z[i]) space and is transverse to the diagonal flow. 3. **The Relationship between Minimal Vectors and the Best Approximations** - There is a direct correspondence between the minimal vectors and the best Diophantine approximations, so the best approximations can be found by studying the minimal vectors. 4. **The Best Constant in Dirichlet's Theorem** - As a by - product of the algorithm, the author also found the best constant in the complex - number version of Dirichlet's theorem. ### Specific Problem Description - **Construction of the Complex Continued Fraction Algorithm**: How to construct a complex continued fraction algorithm through the sequence of minimal vectors of lattices. - **Selection of the Manifold**: Select an appropriate sub - manifold so that the algorithm can run effectively on this manifold. - **Study of Minimal Vectors**: Study the properties of minimal vectors and their relationship with the best Diophantine approximations. - **Optimization of Dirichlet's Theorem**: Determine the best constant in the complex - number version of Dirichlet's theorem and prove its optimality. By solving these problems, the author not only provides a new method for dealing with the problem of the best Diophantine approximations of complex numbers, but also provides new insights for the development of the complex continued fraction theory.