Searcharxiv⌕ Search

arXiv subjects

Ángel Luis Muñoz Castañeda

Publications and source records attributed to Ángel Luis Muñoz Castañeda.

12 recordsLinked to original sources

On the Goppa morphism

We develop a moduli-theoretic framework, over an arbitrary base field, for the classical Goppa construction of algebraic-geometric codes. We consider the moduli stack $\mathcal{LS}_{g,n,d}$ of line bundles of degree $d$ on $n$-pointed smooth projective curves of genus $g$, trivialized along the marked points, as the space of Goppa structures and show that evaluation of global sections defines a natural Goppa morphism $ \mathrm{Goppa}_{g,n,d}\colon \mathcal{LS}_{g,n,d}\longrightarrow \mathrm{Gr}(k,n)$, $k=1-g+d$, together with an extended morphism to $\mathrm{Gr}(k,n)\times \mathcal{M}_{g,n}$. Under explicit numerical hypotheses, we prove that the extended Goppa morphism is an immersion. The pullbacks of the Plücker coordinates are identified with sections of the theta line bundle on the relative Picard scheme, and the locus of non-degenerate codes for which a subset $I$ fails to be an information set is precisely the corresponding theta locus. We identify the fibers of the Goppa morphism over non-degenerate codes $C$ with moduli spaces of non-degenerate $n$-pointed genus-$g$ curves of degree $d$ in the projective space $\mathbb{P}C$ canonically determined by $C$. This yields a Hilbert-incidence description of the Goppa construction and, in a natural range, implies that $\mathrm{Goppa}_{g,n,d}$ is schematic and quasi-projective and that $\mathcal{LS}_{g,n,d}$ is a quasi-projective scheme. In genus zero, we further obtain, for $2\leq d\leq n-3$, a canonical $\mathbb{G}_m^{n-1}$-family of immersions $\mathcal{M}_{0,n}\hookrightarrow \mathrm{Gr}(d+1,n)$, which connects the Goppa morphism with the classical geometry of $\mathcal{M}_{0,n}$.

math.AG↗

Differential Goppa Codes

Rosenbloom and Tsfasman, in their foundational work on the $m$-metric, introduced algebraic-geometric codes defined by multiple points on a smooth projective curve $X$. This construction involves a divisor $G$ and another divisor $D=\sum n p_i$, where $p_i$ are distinct rational points with $p_i \notin \text{supp}(G)$ and $n\in\mathbb{N}$. Although these codes are significant, their formal development for arbitrary genus remains incomplete in the literature, as most studies have concentrated on the genus $0$ case. We present a rigorous treatment of this class of codes. Starting with a smooth projective curve $X$, an invertible sheaf $L$, and an effective divisor $D=\sum n_i p_i$ where the $n_i$ are not necessarily equal, as well as tuples of uniformizers $t_D$ at the points of $D$ and trivializations $γ_D$ for the localizations $L_{p_i}$, the associated differential Goppa code is defined. This code arises from the theory of $n$-jets of invertible sheaves on curves, which enables the description of codewords using Hasse-Schmidt derivatives of sections of $L$. The variation of the code under changes in the data $(t_D, γ_D)$ is examined, and the group acting on these parameters is described. The behavior of the minimum Hamming distance under such variations is analyzed, with explicit examples provided for curves of genus $0$ and $1$. A duality theorem is established, involving principal parts of meromorphic differential forms. It is demonstrated that Goppa codes constitute a proper subclass of differential Goppa codes, and that every linear code admits a differential Goppa code structure on $\mathbb P^1$ using only two rational points.

math.AG↗

Decoding convolutional codes over finite rings. A linear dynamical systems approach

Observable convolutional codes defined over Zpr with the Predictable Degree Property admit minimal input/state/output representations that preserve structural properties under scalar restriction. We make use of this fact to present Rosenthal's decoding algorithm for these convolutional codes. When combined with the Greferath-Vellbinger algorithm and a modified version of the Torrecillas-Lobillo-Navarro algorithm, the decoding problem of convolutional codes over Zpr reduces to selecting two decoding algorithms for linear block codes over a field. Finally, we analyze both the theoretical and practical error-correction capabilities of the combined algorithm as well as its time complexity.

cs.IT↗

On the subalgebra of invariant elements: finiteness and immersions

Let $R$ be an algebra over a ring $\Bbbk$, $T$ an $R$-algebra, $M$ a finitely generated projective $R$-module, and $N$ a $T$-module. Let $G$ be a linearly reductive group scheme over $\Bbbk$ equipped with a representation $ρ:\underline{G}_{R}\rightarrow \underline{\textrm{Aut}}_{\textrm{Mod}(R)}(M)$. For the graded $T$-algebra $A$, defined as $ A := \left( S_{T}^{\bullet} (M^{\vee} \otimes_{R}N )\right)^{G}, $ we determine the conditions under which the graded $T$-algebra $A$ is finitely generated, finitely presented, or flat. Furthermore, we establish the conditions under which a closed embedding of $\textrm{Proj} \ A$ into a projective space exists. Since we do not impose any Noetherian hypotheses, our results generalize those in the literature, providing new powerful tools regarding moduli problems.

math.AG↗

Singular principal bundles on reducible nodal curves

Studying degenerations of moduli spaces of semistable principal bundles on smooth curves leads to the problem of constructing and studying moduli spaces on singular curves. In this note, we will see that the moduli spaces of $δ$-semistable pseudo bundles on a nodal curve constructed by the first author become, for large values of $δ$, the moduli spaces for semistable singular principal bundles. The latter are reasonable candidates for degenerations and a potential basis of further developments as on irreducible nodal curves. In particular, we find a notion of semistability for principal bundles on reducible nodal curves. The understanding of the asymptotic behavior of $δ$-semistability rests on a lemma from geometric invariant theory. The results will allow the construction of a universal moduli space of semistable singular principal bundles relative to the moduli space $\overline{\mathcal M}_g$ of stable curves of genus $g$.

math.AG↗

Generalized parabolic structures over smooth curves with many components and principal bundles over reducible nodal curves

Let $Y_{1},\dots,Y_{l}$ be smooth irreducible projective curves and let $Y$ be its disjoint union. Given a semisimple reductive algebraic group $G$ and a faithful representation $ρ:G\hookrightarrow \textrm{SL}(V)$ we construct a projective moduli space of $(\underlineκ,δ)$-(semi)stable singular principal $G$-bundles with generalized parabolic structure of type $\underline{e}$. In case $Y$ is the normalization of a connected and reducible projective nodal curve $X$, there is a closed subscheme coarsely representing the subfunctor corresponding to descending bundles. We prove that the descent operation induces a birational, surjective and proper morphism onto the schematic closure of the space of $δ$-stable singular principal $G$-bundles whose associated torsion free sheaf is of local type $\underline{e}$.

math.AG↗

On the moduli space of singular principal G-bundles over stable curves

In this paper we proof the existence of a linearization for singular principal G-bundles not depending on the base curve. This allow us to construct the relative compact moduli space of δ-(semi)stable singular principal G-bundles over families of reduced projective and connected nodal curves, and to reduce the construction of the universal moduli space over $\overline{M}_{g}$ to the construction of the universal moduli space of swamps.

math.AG↗

Linear representations of convolutional codes over rings

In this paper we extend the relation between convolutional codes and linear systems over finite fields to certain commutative rings through first order representations . We introduce the definition of rings with representations as those for which these representations always exist, and we show that finite products of finite fields belong to this class. We develop the input/state/output representations for convolutional codes over these rings, and we show how to use them to construct observable convolutional codes as in the classical case.

math.OC↗

On the K-theory of feedback actions on linear systems

A categorical approach to linear control systems is introduced. Feedback actions on linear systems arises as a symmetric monoidal category. Stable feedback isomorphisms generalize enlargement of pairs of matrices. Subcategory of locally Brunovsky linear systems is studied and the stable feedback isomorphisms of locally Brunovsky linear systems are characterized by the K_0 group of the (monoidal) category. Hence a link from linear dynamical systems to algebraic K-theory is stablished.

math.AC↗