In this thesis, an FPGA based hash processor is designed and implemented using a hardware description language / VHDL. Hash functions are among the most important cryptographic primitives and used in the several fields of communication integrity and signature authentication. These functions are used to obtain a fixed-size fingerprint or hash value of an arbitrary long message. The hash functions SHA-1 and SHA2-256 are examined in order to find the common instructions to implement them using same hardware blocks on the FPGA. As a result of this study, a hash processor supporting SHA-1 and SHA2-256 hashing and having a standard UART serial interface is proposed. The proposed hash processor has 14 instructions. Among these instructions, 6 of them are special instructions developed for SHA-1 and SHA-256 hash functions. The address length of the instructions is six bits. The data length is 32 bits. The proposed instruction set can be extended for other hash algorithms and they can be implemented over the same architecture. The hardware is described in VHDL and verified on Xilinx FPGAs. The advantages and open issues of implementing hash functions using a processor structure are also discussed.

No description available.

In the present work we study a linearization framework for assessing the security of hash functions and analyze the proposal of hash function BLAKE. The thesis demonstrates a limitation of a method presented in the linearization framework for which the method could not be applied to the full extent. Further in the thesis, it is explained how to find a message difference for second preimage attack with the help of linear codes. To that end, a matrix representing the linearized compression function of BLAKE is constructed. My thesis as a PDF file and source codes of computations that I created in Mathematica software are on an enclosed CD.

La démocratisation des ordinateurs, des téléphones portables et surtout de l'Internet a considérablement révolutionné le monde de la communication. Les besoins en matière de cryptographie sont donc plus nombreux et la nécessité de vérifier la sûreté des algorithmes de chiffrement est vitale. Cette thèse s'intéresse à l'étude d'une nouvelle cryptanalyse, appelée cryptanalyse logique, qui repose sur l'utilisation de la logique propositionnelle - à travers le problème de satisfaisabilité - pour exprimer et résoudre des problèmes cryptographiques. Plus particulièrement, les travaux présentés ici portent sur une certaine catégorie de chiffrements utilisés dans les protocoles d'authentification et d'intégrité de données qu'on appelle fonctions de hachage cryptographiques. Un premier point concerne l'aspect modélisation d'un problème cryptographique en un problème de satisfaisabilité et sa simplification logique. Sont ensuite présentées plusieurs façons pour utiliser cette modélisation fine, dont un raisonnement probabiliste sur les données du problème qui permet, entres autres, d'améliorer les deux principaux points d'une attaque par cryptanalyse logique, à savoir la modélisation et la résolution. Un second point traite des attaques menées en pratique. Dans ce cadre, la recherche de pré-Image pour les fonctions de hachage les plus couramment utilisées mènent à repousser les limites de la résistance de ces fonctions à la cryptanalyse logique. À cela s'ajoute plusieurs attaques pour la recherche de collisions dans le cadre de la logique propositionnelle. / Democratization of increasingly high-Performance digital technologies and especially the Internet has considerably changed the world of communication. Consequently, needs in cryptography are more and more numerous and the necessity of verifying the security of cipher algorithms is essential.This thesis deals with a new cryptanalysis, called logical cryptanalysis, which is based on the use of logical formalism to express and solve cryptographic problems. More precisely, works presented here focuses on a particular category of ciphers, called cryptographic hash functions, used in authentication and data integrity protocols.Logical cryptanalysis is a specific algebraic cryptanalysis where the expression of the cryptographic problem is done through the satisfiabilty problem, fluently called sat problem. It consists in a combinatorial problem of decision which is central in complexity theory. In the past years, works led by the scientific community have allowed to develop efficient solvers for industrial and academical problems.Works presented in this thesis are the fruit of an exploration between satisfiability and cryptanalysis, and have enabled to display new results and innovative methods to weaken cryptographic functions.The first contribution is the modeling of a cryptographic problem as a sat problem. For this, we present some rules that lead to describe easily basic operations involved in cipher algorithms. Then, a section is dedicated to logical reasoning in order to simplify the produced sat formulas and show how satisfiability can help to enrich a knowledge on a studied problem. Furthermore, we also present many points of view to use our smooth modeling to apply a probabilistic reasoning on all the data associated with the generated sat formulas. This has then allowed to improve both the modeling and the solving of the problem and underlined a weakness about the use of round constants.Second, a section is devoted to practical attacks. Within this framework, we tackled preimages of the most popular cryptographic hash functions. Moreover, the collision problem is also approached in different ways, and particularly, the one-Bloc collision attack of Stevens on MD5 was translated within a logical context. It's interesting to remark that in both cases, logical cryptanalysis takes a new look on the considered problems.

Dans cette thèse, nous nous intéressons à la cryptanalyse de certaines primitives de cryptographie symétrique qui utilisent les concepts de construction du schéma de chiffrement AES. Nous commençons par une analyse de l'AES lui-même dans trois modèles de sécurité différents: le modèle standard, le modèle à clefs reliées et le modèle ouvert. Dans le modèle standard, où l'adversaire cherche à récupérer la clef secrète, nous décrivons les meilleures attaques différentielles existantes sur cet algorithme de chiffrement, en améliorant les attaques différentielles précédemment publiées. Ensuite, nous procédons à une analyse structurelle de l'AES dans le modèle à clefs reliées. Nous montrons des résultats d'impossibilité, indiquant que l'on ne peut pas prouver la sécurité de la structure de l'AES contre les attaques différentielles dans ce modèle. Enfin, dans le modèle ouvert, nous proposons le premier distingueur pour neuf tours d'AES-128, ce qui résout un problème ouvert depuis plusieurs années dans la communauté symétrique. Dans une deuxième partie, nous analysons en détail l'application de l'attaque par rebond sur les primitives basées sur l'AES. Nous montrons qu'il est possible de considérer un tour de plus dans la première des deux phases de cette stratégie, ce qui améliore les meilleurs résultats connus sur les permutations à base d'AES. Ceci résout le problème ouvert consistant à augmenter le nombre total de tours attaqués grâce à cette technique. Nous montrons également qu'il est possible de relâcher certaines contraintes pour augmenter la probabilité de succès de la deuxième étape. Ceci conduit à une diminution des complexités de toutes les attaques publiées. Nous appliquons ces améliorations à la fonction de hachage Grostl, obtenant les meilleures attaques sur la permutation interne. Finalement, nous nous intéressons à la fonction de hachage ECHO pour montrer qu'il est possible d'appliquer plusieurs fois l'attaque par rebond et ainsi attaquer plus de tours de la permutation interne.

The Keccak Sponge Function is the winner of the National Institute of Standards and Technology (NIST) competition to develop the Secure Hash Algorithm-3 Standard (SHA-3). Prior work has developed reference implementations of the algorithm and described the structures necessary to harden the algorithm against power analysis attacks which can weaken the cryptographic properties of the hash algorithm. This work demonstrates the architectural changes to the reference implementation necessary to achieve the theoretical side channel-resistant structures, compare their efficiency and performance characteristics after synthesis and place-and-route when implementing them on Field Programmable Gate Arrays (FPGAs), publish the resulting implementations under the Massachusetts Institute of Technology (MIT) open source license, and show that the resulting implementations demonstrably harden the sponge function against power analysis attacks.

This thesis concerns with hash functions and their usage in authentication. It presents basics of hash functions theory and construction elements. In particular the thesis focuses on LMHash, MD4, MD5 and SHA family hash functions, which are compared from the security point of view. The thesis describes in general the most frequently used hash function attacks, points out the weaknesses of current construction and mentions the future perspective of hash functions. Furthermore the thesis outlines the area authentication and describes usage of hash functions in the area. Practical part of the thesis contains an implements of a general authentication framework implemented in programming language C#. The result is client and server applications, in which two selected authentication methods were successfully tested. The result implementation is flexible with respect to the possible future use of other authentication methods.

Master's thesis analyses modern hash functions. The requirements for these features and briefly outlined some of the types of attacks are given in the first part. The second part focuses on the specication Skein hash function, which is among the candidates for the new SHA-3 standard, and a description of the JCOP platform, which is a function implemented. In the last part of the work there are discussed implementation problematic parts and evaluation of the selected solution.

This work studies systems and methods for anomaly detection in computer networks. At first, basic categories of network security systems and number of methods used for anomaly detection are briefly described. The core of the work is an optimization of the method based on detection of changes in distributions of packet features originally proposed by Lakhina et al. This method is described in detail and two optimizations of it are proposed -- first is focused to speed and memory efficiency, second improves its detection capabilities. Next, a software created to test these optimizations is briefly described and results of experiments on real data with artificially generated and also real anomalies are presented.

Title: Security and usability of standard hash functions, in particular MD-5, SHA-1 and SHA-2 Author: Galaczová Barbora Department: Department of Algebra Supervisor: Doc. RNDr. Tůma Jiří, DrSc., Department of Algebra Consultant: Ing. Budiš Petr, Ph.D. Abstract: In the present work we try to digestedly describe standard hash functions, in particular MD-5, SHA-1 and SHA-2. We describe resume of existing attacks on these hash functions. We closely focused on MD-5 collision attacks, because the other hash function collision attacks result from these. Next we describe posibilities of practical usage of hash function collisions, in particular into the qualified certificates area and possible threats. At the end to the present work we describe new hash functions, which could replace current hash functions. This work also contains software to calculate MD-5 hash and search it`s collisions. The software is based on method invented by Czech cryptoanalytist Vlastimil Klíma. Keywords: hash function, collision, qualified certificate, security.