LITERATURE SURVEY 1) Multiple re-watermarking scenarios AUTHORS: A. Mascher-Kampfer, H.
St€ogner, and A. Uhl The use of classical robust watermarkingtechniques for multiple re-watermarking is discussed. In particular we focus ona comparison of the usefulness of blind and non-blind algorithms for this typeof applications. A surprisingly high number of watermarks may be embedded usingboth approaches, provided that additional data is recorded in the non-blindcase. 2) Data leakage detection AUTHORS: P.Papadimitriou and H. Garcia-Molina We study the following problem: A datadistributor has given sensitive data to a set of supposedly trusted agents(third parties).
Some of the data are leaked and found in an unauthorized place(e.g., on the web or somebody’s laptop). The distributor must assess thelikelihood that the leaked data came from one or more agents, as opposed tohaving been independently gathered by other means. We propose data allocationstrategies (across the agents) that improve the probability of identifyingleakages. These methods do not rely on alterations of the released data (e.
g.,watermarks). In some cases, we can also inject “realistic but fake” datarecords to further improve our chances of detecting leakage and identifying theguilty party. 3) Secure spread spectrum watermarking formultimedia AUTHORS: I. J. Cox, J. Kilian, F. T.
Leighton, and T.Shamoon This paper presents a secure(tamper-resistant) algorithm for watermarking images, and a methodology fordigital watermarking that may be generalized to audio, video, and multimediadata. We advocate that a watermark should be constructed as an independent andidentically distributed (i.i.d.) Gaussian random vector that is imperceptiblyinserted in a spread-spectrum-like fashion into the perceptually mostsignificant spectral components of the data.
We argue that insertion of awatermark under this regime makes the watermark robust to signal processingoperations (such as lossy compression, filtering, digital-analog andanalog-digital conversion, requantization, etc.), and common geometrictransformations (such as cropping, scaling, translation, and rotation) providedthat the original image is available and that it can be successfully registeredagainst the transformed watermarked image. In these cases, the watermarkdetector unambiguously identifies the owner.
Further, the use of Gaussiannoise, ensures strong resilience to multiple-document, or collusional, attacks.Experimental results are provided to support these claims, along with anexposition of pending open problems 4) Asymmetric fingerprinting for largercollusions, AUTHORS: B. Pfitzmann and M.
Waidner Fingerprinting schemes deter people fromillegal copying of digital data by enabling the merchant of the data toidentify the original buyer of a copy that was redistributed illegally. Allknown fingerprinting schemes are symmetric in the following sense: Both thebuyer and the merchant know the fingerprinted copy. Thus, when the merchantfinds this copy somewhere, there is no proof that it was the buyer who put itthere, and not the merchant.
We introduce asymmetric fingerprinting.where only the buyer knows the fingerprinted copy, and the merchant, uponfinding it somewhere, can find out and prove to third parties whose copy itwas. We present a detailed definition of this concept and constructions. Thefirst construction is based on a quite general symmetric fingerprinting schemeand general cryptographic primitives; it is provably secure if all theseunderlying schemes are. We also present more specific and more efficientconstructions. 5)A digital signature scheme secure against adaptive chosen-message attacks AUTHORS: S.
Goldwasser, S. Micali, and R. L.
Rivest We present a digital signature schemebased on the computational difficulty of integer factorization. The schemepossesses the novel property of being robust against an adaptive chosen-messageattack: an adversary who receives signatures for messages of his choice (whereeach message may be chosen in a way that depends on the signatures ofpreviously chosen messages) cannot later forge the signature of even a singleadditional message. This may be somewhat surprising, since in the folklore theproperties of having forgery being equivalent to factoring and beinginvulnerable to an adaptive chosen-message attack were considered to becontradictory. More generally, we show how to construct a signature scheme withsuch properties based on the existence of a “claw-free” pair ofpermutations–a potentially weaker assumption than the intractibility ofinteger factorization. The new scheme is potentially practical: signing andverifying signatures are reasonably fast, and signatures are compact.