Chaos-related Technology Research, And Information Hiding

With the rapid development and extensive applications of computer technology, networks technology, multimedia technology, and Internet in particular, the security of network information is becoming increasingly a key problem that must be solved urgently. Both the chaotic cipher technology and Steganalysis are important research directions in information security field. That applying chaos theory to secure communication and information encryption has already become one of the hot research projects on the combination of nonlinear science and information science, and it is a novel branch of high-tech research fields. Steganography is an art of hiding the very presence of communication by embedding secret messages into innocuous looking cover documents, such as digital images, so its security is higher than that of encryption.The main contents of this paper include the research on the remedy theory of dynamical degradation of digital chaos and chaotic sequences application in information security. Also an improved steganalysis on regular groups of digital images with dynamic mask is proposed.The main research contents and originality of this paper can be summarized as following:1. Dynamical degradation of digital chaos obtained in computer or other digital devices is presented. A new scheme on Perturbing Digital Chaotic Systems is proposed. After being perturbed, chaotic system can generate chaotic binary sequences with uniform distribution and statistical properties invulnerable to cryptographic analysis.2. Pseudo-random sequence generated by PB-PLCM (perturb based piecewise linear chaotic map) is used in the designing of stream cipher , the fine statistical properties of the sequence make it easy to obtain a high security lever.3. A scheme on scrambling digital image based on PB-PLCM is proposed. The address of pixels to be scrambled is selected by pseudo random number sequence generated by PB-PLCM. The simulation results show that the scheme in this thesis is effective for scrambling digital images.4. This paper also summaries several present steganalysis algorithms and presents a dynamic regular groups steganalysis (DRS) algorithm. This novel algorithm dynamically selects an appropriate mask for each image to eliminate the initial error and estimates the length of embedding message by constructing equations with the statistics of regular groups in a dubious image. Experimental results show the algorithm is more accurate than the conventional RS method, and has a lower missing error rate and false alarm rate.