Catálogo de publicaciones - libros

The current deployment of Digital Right Management (DRM) schemes to distribute protected contents and rights is leading the way to massive use of sophisticated embedded cryptographic applications. Embedded microprocessors have been equipped with bulky and power-consuming co-processors designed to suit particular data sizes. However, flexible cryptographic platforms are more desirable than devices dedicated to a particular cryptographic algorithm as the increasing cost of fabrication chips favors large volume production. This paper proposes a novel approach to embedded cryptography whereby we propose a vector-based general purpose machine capable of implementing a range of cryptographic algorithms. We show that vector processing ideas can be used to perform cryptography in an efficient manner which we believe is appropriate for high performance, flexible and power efficient embedded systems.

Peer-to-peer (P2P) networks are widely used to share copyrighted contents and software illegally around the world. Development and applications of P2P technologies have faced strong legal and legislative challenges. A P2P network has to have built-in copyright protection to enable P2P technologies to advance on its own without court’s and legislature’s interference. In this paper, we propose a novel and practical P2P network with a strong privacy protection and reliable tracking mechanism to track the original uploader of any material in the P2P network. When a pirated material is found, its uploader is tracked down, and can be punished by revoking the access to the network, removing his or her uploaded materials, etc. The whole protection system is completely transparent to end users. The proposed scheme would effectively deter users from uploading illegally any copyrighted materials to and dramatically reduce copyrighted materials shared through our P2P network.

The growth of the Internet has created an electronic marketplace for digital goods, such as mp3, software, etc., and peer-to-peer(P2P) networks for distribution and sharing of files have enjoyed enormous popularity in recent. However, there is an essential problem related with trustiness and reliability in P2P, where untrustworthy parties may participate in financial transactions. One general solution to this problem is distributing the encrypted version of digital content so that a legitimate user can get the decryption key from key issuing server. In this paper, we propose a new approach to key issuing for designing a secure content distribution system. Each user can recovery the original digital content from the encrypted version of it by purchasing decryption key from key issuing server. However, instead of generating a random key, the key used for content encryption and decryption can be directly derived from the usage rules or transaction terms in our system. This approach allows to separate content encrypter and key issuer, and then it also enables to remove the need of secure key transfer between content provider and key issuing server in secure content distribution system.

The present study implemented a system to facilitate broadcast switching between a central broadcasting station and its local broadcasting stations using audio watermarking technology. Because extraction errors in synchronization system for broadcast switching cause broadcast accidents, audio watermark technology must be highly reliable. This research developed an audio watermarking algorithm efficiently applicable to broadcast synchronization system and implemented a broadcast switching system through the real-time application of the algorithm. SNR of broadcast signal with watermark inserted by the developed audio watermarking algorithm was 66.1dB, which is hardly distinguishable from broadcast signal before watermark insertion. In a robustness test, when MP3 compression of 96kbps and 128kbps was used, the extraction rate was 100%. In addition, the extraction rate was 100% for the addition of noise below -50db The implemented system was proved to be reliable as a broadcasting system, showing extraction rate of 100% and error rate of 0% for broadcast signal with watermark inserted in 240 hours’ actual broadcasting situation.

We present the problem of enforcing a Digital Rights Management (DRM) system that needs to consider location-dependent licensing policies and operates on top of existing conditional access standards. A major application for location-dependent DRM is Pay-TV broadcasting as rightsholders require different business models in different regions. A global provider’s enduser equipment needs to validate the user location in some way in order to enforce DRM in this scenario. We will depict several solutions to the problem and compare their security qualities. The main result is that trusted computing hardware may not be the most appropriate solution given reasonable conditions.

We present a Digital Rights Management (DRM) system that supports what you see is what you pay. In our system, multimedia is compressed with a scalable codec and encrypted preserving the scalable granularity and multi-access capability. This paper focuses on the DRM modules enabling efficient key generation and management. We employ a light license server which stores only the master keys of content publishers, which are used to regenerate decryption keys for clients during license acquisition. All the remaining information needed in key generation is efficiently packaged in a DRM header of protected content. The DRM header is sent to a license server during license acquisition to allow the license server to generate a single key for a requested access, which is sent to the client in a license along with the acquired rights. The key is used by the client to generate all the remaining keys of subordinate accesses.