SUMMARYOn the basis of grey system theory, an image information hiding algorithm is proposed in this paper. First, to choose the blocks of rich texture, a method based on two-dimensional grey relational analysis is designed; making use of the improved GM(1,1) prediction model, the secret image is compressed losslessly, and then scrambled by means of time-division multiplexing method in communication theory. Block discrete cosine transform is applied to the chosen blocks, and coefficients of mid frequencies are chosen as the embedded data, whereas the embedding intensity is determined on the basis of human visual system, and the embedding capacity of sub-blocks can be controlled by its two-dimensional grey relational grade adaptively. The experiment results show that the algorithm cannot only achieve satisfactory invisibility and robustness, but also improve the payload of the cover image, and meanwhile solve the problem that the small capacity of traditional algorithms in transform domain efficiently. Copyright © 2013 John Wiley & Sons, Ltd. , which takes the uncertain, little sample, poor information system as the research object, presented by Chinese scholar Julong Deng as an emerging discipline. Its contents mainly include grey prediction model, grey relational analysis, grey decision-making, and so forth. Information hiding technology is one of its important applied fields [2,3].Information hiding model was derived from 'prisoners problem', put forward by Simmons in 1984. To deliver information secretly, traditional cryptography [4, 5] mainly use replacing or scrambling technologies to disrupt the original features of the secret information. Just because cipher texts resemble a stream of meaningless codes, they easily attract attacker to try to either recover them or simply destroy them, which seriously affect the security of information communication. To circumvent this problem, information hiding technology [6-8] mainly focus on how to hide the coded information into another host media, which makes it hard for attackers to find out the secret information. And, it is also impossible to detect the carrier from vast amounts of data in communication channel. These characteristics make it widely accepted in information security fields.[9] provides one efficient means of concealing the authentication messages exchanged

No abstract

Recently, secure query has become a significant issue for researchers. It proposes the concept of secure multi-party query in Internet of things or sensor networks making use of “underground parties” to guarantee network security. Based on a novel network model assumption, it adopts homomorphic privacy and secure multi-party computation techniques to realize the above mentioned secure query, and then it puts forward the horizontal and vertical query algorithm in the top-k query through cooperation of underground parties. Thereafter, by the simulation tests it compares the time efficiency of the proposed algorithm with other typical sort algorithms under different top-k scenarios. At last, it analyzes the relationship between query time and number of underground party nodes.