Pseudo-Random Number Generator

RC4 algorithm is one of the most widely used stream ciphers. It is fast, simple and suitable for software and hardware. It is used in many applications, but it has a weakness in the distribution of key stream bytes, the first few key stream bytes of PRNG are biased or related to some secret key bytes and thus the analysis of key stream bytes makes it possible to attack RC4, and there is a correlation between the key stream bytes that make it weak and breakable by single and double byte bias attack. This work shows a new algorithm proposed by using initial state factorial to solve the correlation issue between public known outputs of the internal state and making this algorithm is robust against attack by using an additional state table with the same length of the state to contain the factorial of initial state elements. Also, shows the single byte bias attack on RC4 by using the newly designed algorithm. The results showed that the proposed algorithm is robust against attack and could retrieve the first 32 bytes of the plain text by using the proposed algorithm of single byte bias attack with a probability of 100%. Additionally, the developed algorithm is robust against many attacks such as distinguishing attack.

In this paper, neural network based cryptology is performed. The system consists of two stages. In the first stage, neural network-based pseudo-random numbers (NPRNGs) are generated and the results are tested for randomness using National Institute of Standard Technology (NIST) randomness tests. In the second stage, a neural network-based cryptosystem is designed using NPRNGs. In this cryptosystem, data, which is encrypted by non-linear techniques, is subject to decryption attempts by means of two identical artificial neural networks (ANNs). With the first neural network, non-linear encryption is modeled using relationbuilding functionality. The encrypted data is decrypted with the second neural network using decision-making functionality. c ⃝CTU FTS 2014 1 Neural Network World 2/14, ?-? cryptography. The concept of neural-based cryptography was first introduced by Laurie in 1990 [15]. In 1993, neural networks' weights were used as secret keys by Tanrıverdi [28]. Neural cryptography applications were researched by Pointcheval in 1994 [20]. Following this, a practical study of neural cryptography appeared in 1996 [24]. Cin also used neural networks for authentication [5]. A symmetric probabilistic encryption scheme based on the chaotic attractors of neural networks was studied by Guo et al. [11]. Chaotic neural encryption and decryption was studied by Sue et al. [26]. The Hopfield network and its applications were studied by Chi-Kwong and Cheng [4]. The implementation of an MLP network using its one-way property in the hash function was investigated by Yee and Silva [33]. Karras and Zorkadis presented an ANN-based technique to strengthen traditional generators such as IDEA and ANS X.9. Neural network-based pseudo-random stream generators have also been evaluated [14]. Godhavari et al. applied neural synchronization and shared weights values as a symmetric secret key in the DES algorithm [10]. Ruttor performed the neural synchronization of two tree parity networks with mutual outputs by means of the hebbian learning rule; weight and bias values were then used as a secret key [22]. Arvandi et al. presented an ANN-based symmetric chipper method [3]. Sagıroglu andÖzkaya studied ANN security for electronic communications with an algorithm developed by DELPHI, which uses ANNs as a secret key [23]. Gnanam and Munukur presented a method that aimed to remove the need for encoding to follow a general rule through use of ANNs. According to Gnanam and Munukur, there is a way to decode the encrypted data. However, the plain texts are encoded [17].İlker and Kenan presented an ANN-based chaotic generator to overcome the weakness of chaotic cryptosystems [7]. Sivagurunathan presented a neural network that classifies chipper texts encrypted with Playfair, Vigenere and Hill chipper [25]. Mohammad and Babak proposed an ANN-based S-box design scheme [18]. Othman and Jammas developed stream chipper cryptosystem software and hardware in FPGA, a pseudo-random number generator designed using ANNs [19]. Yayık and Kutlu proposed a neural network model for improving the randomness of a classical random number generator in 2013 [31]. Yayık and Kutlu proposed a neural network model for hash function that can be used in electronic signature, in 2013 [30].Yayık and Kutlu proposed a neural network model for asymmetric cryptosystem model that weights, bias and topology of NN are shared between receiver and sender as secret key , in 2013 [32]. In this study, a neural network-based pseudo-random number generator and a neural networkbased non-linear cryptosystem are developed. This study is implemented in five chapters. The first chapter gives information about information security, cryptology and existing literature. In the second chapter, the artificial neural network (ANN), the advantages of neural networks for cryptology, pseudo-random number generators, neural-based pseudo-random number generators and randomness tests are all explained. In the third chapter, the implementation of neural cryptology is explained step by step. In the fourth chapter, the results and conclusions are given. Finally, in the fifth chapter, discussions are presented.

Blockchain-based random number generation represents a provably fair mechanism for online games of chance if implemented correctly. Depending on the blockchain technology, seeds used for random number generation may not have sufficient entropy as most inputs are deterministic or inherently favor a participating party. This paper explores possible implementations of pseudorandom number generators (PRNG) using Turtle Network (v1.2.6) which implements the Waves open blockchain protocol (v1.2.6).

Smart cards rely on pseudorandom number generators to provide uniqueness and freshness in their cryptographic services i.e. encryption and digital signatures. Their implementations are kept proprietary by smart card manufacturers in order to remain competitive. In this paper we look at how these generators are implemented in general purpose computers. How architecture of such generators can be modified to suit the smart card environment. Six variations of this modified model were implemented in Java Card along with the analysis of their performance and randomness. To analyse the randomness of the implemented algorithms, the NIST statistical test suite is used. Finally, an overall analysis is provided, that is useful for smart card designers to make informed decisions when implementing pseudorandom number generators.

Portfolio Analysis of Financial Market Risks by Random Set Tools
https://www.academia.edu/187396

The eventological theory of decision-making, the theory of eventfull decision-making is a theory of decision-making based on eventological principles and using results of mathematical eventology; a theoretical basis of the practical eventology. The beginnings of this theory which have arisen from eventfull representation of the reasonable subject and his decisions in the form of eventological distributions (E-distributions) of sets of events and which are based on the eventological H-theorem are offered. The illustrative example of the eventological decision-making by the reasonable subject on his own eventfull behaviour in the financial or share market is considered.

If you are interested in my work, and you have the opportunity to provide financial support for the publication of my book "Theory of experience and of chance as a theory of being," I ask you to use my PayPal account: paypal.me/EventologyTheory

The purpose of this paper is to present a novel Hybrid Random Number Generator (HRNG). Here "hybrid" refers to the fact that to construct this generator it is necessary to use 1) physical components — texts — and a physical process, and 2) a mathematical procedure. This HRNG makes it possible to generate genuine random numbers which may be used both for computer simulation of probabilistic systems and in the field of cryptography. The results of a comparative study of the binary strings generated by this HRNG and of those generated by two highly used implementations of a congruential algorithm designed to generate pseudorandom numbers are given here. One of the latter is the implementation incorporated into the Java 2 platform (version 1.6), and the other is the implementation incorporated into the runtime library of Microsoft's Visual C++ 2008 compiler.

In this article, we propose a new regenerator of pseudo-random sequences R.A NMJ, based on evolutionist algorithms, in that case the genetic algorithms, in order to regenerate pseudorandom sequences cryptographically reliable. R.A NMJ is an algorithm that simulates the evolution, via dissipative functions, of a population of individuals created from a password. Each individual evolves since its creation in a dynamic manner, thus generating a chaotic behaviour. The algorithm R.A NMJ will create at the time of process I the initial population from a password, which will evolve at the time of process II. Process III, that is iterative will regenerate at each iteration a binary sequence where all individuals of the population will contribute; a sequence is combined with the clear message, thus generating the cipher. The addressee, equipped with a similar generator deciphers the received message while applying the same process. For the evaluation of the statistical characteristics of regenerated sequences by R.A NMJ, we have used the software NIST Statistical Test Suite, for many passwords and which lead to nice results.

Pseudo Random Number Generators (PRNG) are the base for stochastic simulations. The usage of good generators is essential for valid simulation results. OPNET Modeler a well-known tool for simulation of communication networks provides a Pseudo Random Number Generator. The extension of OPNET Modeler with external generators and additional statistical evaluation methods that has been performed for this paper increases the flexibility and options in the simulation studies performed.