Abstract:In this article, we define the notion of slim (normal) bases and show their existence for various fields. As an application, an algorithm will be given that computes the spectrum of a basefield transform by merely using O(n) additions.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is to define and prove the existence of "slim bases" and show their applications in computing the Algebraic Discrete Fourier Transform (ADFT). Specifically, by introducing and studying the concept of slim bases, the author aims to provide an efficient algorithm to compute the spectrum of ADFT, using only \(O(n)\) additions.
### Detailed description of the main problem
1. **Define slim bases**:
- The author defines the concept of slim bases. That is, for an extension \(K(\zeta_n)/K\), if a normal basis \(\vartheta G_n\) satisfies that for all \(s\in\mathbb{Z}\), \(\text{Tr}(\vartheta^*\zeta_s^n)\in\{0,\pm1\}\), then it is called a slim base.
- Here, \(\zeta_n = e^{2\pi i / n}\) is the \(n\)-th primitive root of unity, \(G_n\) is the Galois group of the extension \(K(\zeta_n)/K\), \(\vartheta\) is the normal basis generator (NBG), and \(\vartheta^*\) is its dual basis.
2. **Prove the existence of slim bases**:
- The author proves that when the Galois group \(G_n\cong(\mathbb{Z}/n\mathbb{Z})^\times\), the extension \(K(\zeta_n)/K\) has a slim base.
- In particular, for \(Q(\zeta_n)/Q\), if \(n\) is square - free (that is, \(\mu(n)^2 = 1\), where \(\mu\) is the Möbius function), then the element \(\vartheta:=\mu(n)\zeta_n\) generates a slim base.
3. **Application: Efficiently compute the ADFT spectrum**:
- The author proposes an algorithm. By using the properties of slim bases, it can compute the spectrum of ADFT within \(O(n)\) additions, especially when \(n = 2^t\).
- This algorithm is implemented through matrix operations and basis transformations. Specific steps include column permutations, row permutations, etc., ensuring that the computational complexity is linear.
### Summary
The core objective of the paper is to simplify and accelerate the computation of the Algebraic Discrete Fourier Transform by introducing the concept of slim bases. This not only enriches the theoretical research in algebra and number theory but also provides a more efficient computational method in practical applications.