Quantum Chaotic Cryptography : A New Approach

Loading...
Thumbnail Image
Date
2015-07
Authors
Akhshani, Afshin
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
Since 1990s chaotic dynamical systems have been widely used to design new strategies to encrypt information in analog and digital areas. Recently, many digital chaos-based cryptosystems are proposed and a number of them have been cryptanalyzed. The fact that an optimum designs in the context of chaotic cryptography not only demands a solid background in cryptography, but also a thorough knowledge of nonlinear dynamics and chaos. However, in the design of digital chaotic cryptosystems, there are two important issues that have not been seriously considered by most designers of digital chaotic ciphers. First, avoiding the reconstruction of the dynamics of the underlying chaotic systems and the second, non-periodicity of chaotic orbits. The two main contributions of this work are its responses to these two theoretical and methodological problems. In this way, one possible solution for avoiding the reconstruction of chaotic orbits is increasing the complexity of orbits. In this work, two new dynamical systems, quantum and synchronized coupled chaotic maps are proposed and applied in two main application fields, which are pseudo random number generator and image encryption fields. Both theoretical and experimental analysis of proposed cryptosystems are reported and discussed. In addition, in this work in order to quantify the degree of non-periodicity of chaotic orbits, the scale index approach as a mathematical tool is proposed. This method allows the selection of the adequate configurations (optimal parameter selection) of a dynamical system to implement strategies of confusion and diffusion of information. The security analysis of the chaos-based cryptosystems is one of the most important aspects in chaotic cryptography. In this sense, measuring the randomness and spatial complexity of cipher images are two important tasks that should be considered. To this end, first the accuracy of different entropy (as a measure of randomness) measures such as Shannon entropy, Min-entropy and Renyi entropy is examined. However, the results obtained are not reliable. In this work, new alternative approaches, the Mean-block entropy and entropy growth curve, are proposed for quantifying the presence randomness in the cipher images. Moreover, in this work two fractal measures, lacunarity and succolarity, are proposed to quantify spatial complexity in the chaos-based cipher images since the fractal dimension alone could not sufficiently explain the complex patterns of the encrypted images. The results obtained indicate that the mean-block entropy is more accurate than Shannon, Min and Renyi entropies. Also, the asymptotic behavior of entropy growth curve can be considered as a signature of randomness. Moreover, the results of lacunarity and succolarity analyses clearly show that the encryption process could lead to the destruction of self-similarity and connectivity among the pixels, respectively. As a conclusion, it can be strongly recommended to use the proposed mathematical tool and scale-dependent complexity measures including mean-block entropy, entropy growth curve and fractal measures, to guide researchers through designing and implementing of new cryptosystems.
Description
Keywords
Chaotic cryptography
Citation