The Research On Orthogonal Fingerprinting Technology

With the rapid development of networks and digital technology, Multimedia products such as audio, images and video appeared in people's lives. They can greatly meet people's entertainment and communication needs, but as the network is more accessible and digital products are easy to modify, reproduce and disseminate, publishers'interests can be seriously damaged. The technical of traditional encryption is also difficult to inhibit redistribution. In turn, digital fingerprinting is a technique for identifying users who use multimedia content for unintended purposes. It has gotten more and more attention from internal and external investigators.In this paper,we focus on design of detection algorithm for orthogonal fingerprinting system and researching of collusion-resistant fingerprinting algorithm. Matlab is our simulation environment. Following are the main contents and achievements of this paper:(1) In order to reduce the correlations of traditional detection algorithm for orthogonal fingerprinting system, the paper develops an advanced tree-structured detection algorithm, which is based on a binary tree detection algorithm. It divides fingerprinting set into many subsets and runs detection algorithm on those subsets. The experiment shows that new detection algorithm can reduce the correlations.(2) Under the inspiration of linear block codes, this paper also develops a fingerprinting scheme to solve the problem of low coding efficiency and high storage space for the existing schemes. It firstly designs a multiple linear block codes, which have n symbols. We use t symbols to identify colluders and other n-t symbols to determine whether the codes are legitimate and then connect orthogonal fingerprinting of each symbol to constitute user's fingerprinting.To prove the feasibility of this scheme, probability theory has been used.The theory analysis shows that the scheme improves coding efficiency to some extent and reduces storage space of fingerprinting to O (log (n)). The experiment with images shows that the approach is robust and has good anti-collusion performance.