Privacy-preserving watermarking transformation technique in a distributed environment

Abstract

Watermarking relational data is an important topic in the field of copyright protection, given the increasing prevalence of piracy and illegal duplication of databases. While there exist solutions for watermarking relational data, the problems that arise due to multiple ownerships have not been acknowledged. When data in consideration are independently owned, it may not be possible to integrate the data together and then embed a watermark because of privacy and data confidentiality issues. This paper focuses on developing a privacy-preserving watermarking transformation technique to counter the aforementioned problems. The proposed technique works with most existing watermarking methods and necessitates the data owners to collaboratively perform the watermarking process. The transformation technique is applicable for both numerical and categorical data. The computation overhead of the proposed technique is evaluated through comprehensive experiments with various factors that affect the underlying watermarking methods.