Watermarking of Relational Databases with Emphasis on Re-watermarking Attack

The proliferation of internet-based application environments and outsourcing of relational databases have amplified the vulnerability to counterfeited ownership assertion and, thus, proving ownership rights on digital contents is immensely vital than ever. To achieve this several watermarking techniques have proposed in recent past. But an attacker can launch a re-watermarking attack by inserting a new watermark in the already marked database; thus causing conflicting ownership claims by the legitimate data owner and the attacker. Only a few watermarking techniques provide mechanisms to combat such attacks. Therefore, there is a need for a database watermarking technique which provides solutions in case of such attacks. In this paper, we identify different scenarios to identify and combat re-watermarking attacks. We resolve the re-watermarking problem by integrating the extracted watermark information from date stamps. For this purpose, our approach plugs-in bi-level security principle, thus provides robustness against re-watermarking attack.