Self-Dual Cyclic Codes with Square-Root-Like Lower Bounds on Their Minimum Distances

Hao Chen,Cunsheng Ding
2024-11-05
Abstract:Binary self-dual cyclic codes have been studied since the classical work of Sloane and Thompson published in IEEE Trans. Inf. Theory, vol. 29, 1983. Twenty five years later, an infinite family of binary self-dual cyclic codes with lengths $n_i$ and minimum distances $d_i \geq \frac{1}{2} \sqrt{n_i+2}$ was presented in a paper of IEEE Trans. Inf. Theory, vol. 55, 2009. However, no infinite family of Euclidean self-dual binary cyclic codes whose minimum distances have the square-root lower bound and no infinite family of Euclidean self-dual nonbinary cyclic codes whose minimum distances have a lower bound better than the square-root lower bound are known in the literature. In this paper, an infinite family of Euclidean self-dual cyclic codes over the fields ${\bf F}_{2^s}$ with a square-root-like lower bound is constructed. An infinite subfamily of this family consists of self-dual binary cyclic codes with the square-root lower bound. Another infinite subfamily of this family consists of self-dual cyclic codes over the fields ${\bf F}_{2^s}$ with a lower bound better than the square-root bound for $s \geq 2$. Consequently, two breakthroughs in coding theory are made in this paper. An infinite family of self-dual binary cyclic codes with a square-root-like lower bound is also presented in this paper. An infinite family of Hermitian self-dual cyclic codes over the fields ${\bf F}_{2^{2s}}$ with a square-root-like lower bound and an infinite family of Euclidean self-dual linear codes over ${\bf F}_{q}$ with $q \equiv 1 \pmod{4}$ with a square-root-like lower bound are also constructed in this paper.
Information Theory
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to construct self - dual cyclic codes with a square - root lower bound. Specifically, the article aims to construct infinite families of Euclidean and Hermitian self - dual cyclic codes with a minimum distance of a square - root or similar square - root lower bound. Before this, among the known binary self - dual cyclic codes, there is no infinite family that can reach the square - root lower bound, and among the non - binary Euclidean self - dual cyclic codes, there is no infinite family that can exceed the square - root lower bound. ### Main problems 1. **Lack of self - dual cyclic codes with a square - root lower bound**: There are no self - dual cyclic codes of an infinite family with a minimum distance of a square - root lower bound in the literature yet. 2. **Constructing self - dual codes with a square - root lower bound**: Although there are some non - cyclic self - dual codes that can reach the square - root lower bound, constructing self - dual cyclic codes with this property remains a challenge. ### Main contributions of the paper - **Generalizing the van Lint theorem**: Generalize the original van Lint construction from binary cyclic codes to cyclic codes over \(F_{2^{s}}\), and remove certain conditions in the original construction. - **Constructing new self - dual codes**: Construct an infinite family of Euclidean self - dual cyclic codes with a minimum distance of a square - root lower bound (see Theorem 7). - **Hermitian self - dual codes**: Also construct an infinite family of Hermitian self - dual cyclic codes with a minimum distance of a square - root lower bound (see Theorem 8). ### Specific content - **Square - root lower bound**: If the minimum distance \(d\) of an \([n,k,d]_{q}\) linear code satisfies \(d\geq c\sqrt{n}\), where \(c\) is a fixed positive number, it is called a square - root lower bound or a similar square - root lower bound. - **Importance of self - dual codes**: Self - dual codes have important applications in areas such as combinatorial design, invariant theory, and quantum error - correcting codes. - **Specific construction method**: By using the Plotkin construction and the generalized van Lint theorem, the author constructs self - dual cyclic codes with a square - root lower bound. ### Conclusion The paper successfully solves an important problem in the field of self - dual cyclic codes, that is, constructing an infinite family of self - dual cyclic codes with a minimum distance of a square - root lower bound. This not only fills a theoretical gap but also provides new tools and methods for the development of coding theory.