Abstract
Digital watermarking for relational databases emerged as a candidate solution to provide copyright protection, tamper detection, traitor tracing, maintaining integrity of relational data. Many watermarking techniques have been proposed in the literature to address these purposes. In this paper, we survey the current state-of-theart and we classify them according to their intent, the way they express the watermark, the cover type, the granularity level, and their verifiability
1 Introduction
The recent surge in the growth of the Internet results in offering of a wide range of web-based services, such as database as a service, digital repositories and libraries, e-commerce, online decision support system etc. These applications make the digital assets, such as digital images, video, audio, database content etc, easily accessible by ordinary people around the world for sharing, purchasing, distributing, or many other purposes. As a result of this, such digital products are facing serious challenges like piracy, illegal redistribution, ownership claiming, forgery, theft etc. Digital watermarking technology is an effective solution to meet such challenges. A watermark is considered to be some kind of information that is embedded into underlying data for tamper detection, localization, ownership proof, traitor tracing etc.
10 Conclusions
In this paper we survey the current state-of-the-art of different watermarking and fingerprinting techniques for relational databases. We classify all the techniques based on (i) whether the technique introduces the distortion to underlying data, (ii) the type of the cover where mark is embedded, and (iii) the type of the watermark information. Most of the distortion-based watermarking techniques mainly aim at protecting the ownership, whereas distortion-free watermarking techniques mostly are fragile and aim at maintaining integrity of the database information. Although we classify the schemes based on different watermark information, most of the numerical distortion-based schemes follow almost similar steps to identify the candidate bit positions for the watermark. Finally, we observe that the usability of the watermarked database and queries still remains an open issue for future research.
