Fingerprinting Long Forgiving Messages

Abstract

In his 1983 paper, Neal Wagner' defines a perfect fingerprint to be an identifying fingerprint added to an object in such a way t h a t any alteration to it that makes the fingerprint unrecognizable will also make the object unusable. A perfect fingerprinting scheme for binary data would seem difficult t o devise, since it would be possible t o discover the fingerprints by comparing different fingerprinted copies of the same piece of data. In this paper we discuss a fingerprinting scheme which, although i t does not surmount this problem entirely, at least specifies the number of copies an opponent must obtain in order to erase the fingerprints. The fingerprints involved will be rather lengthy, so we will restrict ourselves to what we will call long forgiving messages. A forgiving message is one which is still readily understandable and not jarring when up t o 0.1% of i t has been altered. Examples are voice and television. People can speak comfortably amid the noise of a cafeteria and can enjoy watching a television show with several pixels per frame altered. The idea in each case is that the support of the noise (the set outside which the additive noise must vanish) must have small

Efficient Dynamic Traitor Tracing

“…This algorithm is used as a building block by later algorithms. The optimal algorithm that uses p + 1 versions is presented in section 4, thus establishing (1). In section 5 we present two algorithms that use more than the minimal number of versions required and prove bounds (2) and (3).…”

“…The last color is given to the union of the remaining q + 1 sets (including J). (2) If the pirate ever broadcasts a color given to only one (nonsingleton) set A i,a : Given such a set A j,1 , we add it to J, set A j = A j,0 , and go to step (1). In order to find such a set A j,1 , we distinguish between two subcases.…”

“…A basic tool used by many traitor tracing schemes is cryptographic fingerprinting (e.g., [13,1,2,7]), which allows the schemes to generate different versions of the same content and send them to subsets of users. This can be done, for example, by watermarking each version, with no noticeable degradation in quality.…”

Anonymous Buyer-Seller Watermarking Protocol With Additive Homomorphism

“…On the other hand, the literature can also be categorized as as symmetric schemes, asymmetric schemes, and anonymous schemes. In symmetric schemes (Blakley et al, 1985;Boneh and Shaw, 1995;Cox et al, 1997), both the seller and the buyer know the watermark and the watermarked content. As a consequence, it is possible for a malicious seller to frame an innocent buyer, or for an accused buyer to repudiate the guilt.…”

Robustness and security of digital watermarks

“…An early and intriguing paper is that of Wagner [43], who proposed the use of randomly selected additive watermarks and did a preliminary study of the resistance of such marks to collusive attacks. Blakely, et al [3] looked at a combinatorial model for collusion resistance, and proposed a scheme that offers k-way collusion resistance within the model but requires a number of watermark bits exponential in the number of colluders. Chor, et al [7] worked on a related problem involving tracing pirates in a broadcast distribution system using a multiple-key protocol.…”