The importance of digital watermarking for digital assets such as relational databases to preserve their copyrights is becoming more and more important as time goes by. In the past few years, a large number of techniques have been proposed for hiding copyright marks specifically on relational databases. In this paper, we present an effective watermarking technique for relational data that is robust against various attacks. While previous techniques have been mainly concerned with introducing errors into the actual data, our approach inserts new tuples that are not real and we call them "fake" tuples, to the relation as watermarks. We will show that our approach leads to an effective technique that is robust against different forms of malicious attacks as well as benign updates to the data.