Design and implementation of robust Keyed Hash functions based on Chaotic Neural Network - Archive ouverte HAL Accéder directement au contenu
Autre Publication Scientifique Année : 2018

Design and implementation of robust Keyed Hash functions based on Chaotic Neural Network

Résumé

Due to the important properties of Chaos and Neural Networks, such as non-linearity, confusion, diffusion, one-way and parallel implementation, nowadays, Keyed Chaotic Neural Network (KCNN) is used to design new cryptographic hash functions. These hash functions are used to provide message integrity, authentication and digital signature. In our thesis, we designed and implemented two structures of KCNN: the first one is based on Merkle-Dåmgard construction and the second one is based on Sponge function, used by the standards SHA-2 and SHA-3, respectively. For the first structure, we proposed two architectures, each includes three output schemes namely: Matyas-Meyer-Oseas (MMO), Modified Matyas-Meyer-Oseas (MMMO) and Miyaguchi-Preneel (MP). The first architecture is based on two-layer neural network and the second one is based on one layer neural network followed by a combination of nonlinear functions. For these two proposed architectures, the length of hash value is 256 bits. For the second structure, we also proposed two similar architectures as structure 1, with length of hash value equal to 256 and 512 bits, respectively. The theoretical analysis and the obtained experimental performances demonstrate the robustness and efficiency of our proposed structures in terms of Collision Resistance, Message Sensitivity, Key Sensitivity, Diffusion effect, Hash throughput and Immunity against many attacks compared to other proposed CNN hash functions in the literature as well as the standard SHA-2.
Fichier non déposé

Dates et versions

hal-01812203 , version 1 (11-06-2018)

Identifiants

  • HAL Id : hal-01812203 , version 1

Citer

Nabil Abdoun, Safwan El Assad, Rima Assaf, Olivier Déforges, Mohamad Khalil, et al.. Design and implementation of robust Keyed Hash functions based on Chaotic Neural Network. 2018. ⟨hal-01812203⟩
119 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More