Abstract
Since transmitting images, video, audio and other valuable media by digit form on the internet is very easy to be misappropriated and copied illegally, protecting the digital copyright has become an important issue of the information security area. The DRM (Digital Right Management) system is a general solution of protecting copyright of e-books via traditional encryption mechanism. However, it has the risk of being cracked. Instead, information hiding technology provides a light weight solution for clarifying the digital copyright. This technology can embed information into cover-media imperceptibly. EPUB (short for electronic publication) is a free, open and also the most popular e-book format based on XML (eXtensible Markup Language). Although there is some previous research on information hiding algorithm of XML, there is not any suitable algorithm for EPUB e-books. When these algorithms work in EPUB files, its hiding capacity is roughly proportional to the chapter number. We further found that the hiding capacity of these algorithms is extremely insufficient while working in few-chapter e-books. Moreover, to prevent unexpected error, most of XML-based services or applications own a strict specification. Some of these algorithms do not care for the specification and may make XML documents work abnormally. Therefore, EPUB e-books are in urgent need of appropriate information hiding algorithm. In this thesis, we propose an information hiding algorithm based on XML via changing the order of all XML attributes and unordered sibling-elements. Through our algorithm, publishers can embed unique serial number into most of EPUB files for clarifying the digital copyright. Compared with previous algorithms, our algorithm has high capacity even if it works in few-chapter EPUB e-books. The availability of our algorithm is very well. Also, the imperceptibility is good, and the robustness is at a certain level.