Vidéo pédagogique
Notice
Sous-titrage
Anglais
Langue :
Anglais
Crédits
Irene Marquez-Corbella (Intervention), Nicolas Sendrier (Intervention), Matthieu Finiasz (Intervention)
Conditions d'utilisation
Ces ressources de cours sont, sauf mention contraire, diffusées sous Licence Creative Commons. L’utilisateur doit mentionner le nom de l’auteur, il peut exploiter l’œuvre sauf dans un contexte commercial et il ne peut apporter de modifications à l’œuvre originale.
DOI : 10.60527/7a10-qr62
Citer cette ressource :
Irene Marquez-Corbella, Nicolas Sendrier, Matthieu Finiasz. Inria. (2015, 5 mai). 1.8. Goppa Codes , in 1: Error-Correcting Codes and Cryptography. [Vidéo]. Canal-U. https://doi.org/10.60527/7a10-qr62. (Consultée le 2 juin 2024)

1.8. Goppa Codes

Réalisation : 5 mai 2015 - Mise en ligne : 20 février 2017
  • document 1 document 2 document 3
  • niveau 1 niveau 2 niveau 3
Descriptif

In this session, we willtalk about another family of codes that have an efficientdecoding algorithm: the Goppa codes. One limitation of thegeneralized Reed-Solomon codes is the fact that the length isbounded by the size of the field over which it is defined. This implies that these codes areuseful when we use a large field size. In the sequence, we'llpresent a method to obtain a new code over small alphabets byexploiting the properties of the generalized Reed-Solomon codes. So, the idea is to construct a generalizedReed-Solomon code over a sufficiently large extensionof a field and extract only those codewords that liecompletely in the field.  Let a be a n-tuple ofelements from the field Fq^n which are all differentand let b be an n-tuple of elements from the field Fq^n which are non-zeros. Then, the alternant codesassociated to the pair a,b and parameter r is the Fqlinear restriction of the generalized Reed-Solomon codes ofdimension r associated to the pair a,b. a will be called the supportand b the column multiplier. So, the alternant code associated to the parameters a and b isa linear code with dimension greater than n - mr andminimum distance greater than r + 1  And the proof is very easy. First of all, recall thatthe dual of a generalized Reed-Solomon code is again ageneralized Reed-Solomon code.And this new generalizedReed-Solomon code has parameters n, n - r and r + 1 since the generalized Reed-Solomon code is MDS. Thus, the alternant codeassociated to the pair a and b can be defined by r paritycheck equations over Fq^m and mr parity check equations over Fq. So, the dimension of thealternant code must be at least n - mr. Moreover, theminimum distance of an alternant code is at least theminimum distance of a generalized Reed-Solomon code since thegeneralized Reed-Solomon code is a subset of the alternant code.

Intervention

Dans la même collection

Avec les mêmes intervenants et intervenantes

Sur le même thème