Secured Relational Database Watermarking Using Genetic and Firefly Optimization Algorithms

The information technology plays a vital role during the rapid growth of using information systems comprising relational data. Also, the relational information is important in many areas such as medical, military, forensics and stock market, where data should be distributed to users from a centralized database. The relational databases which is used in shared environment for extracting the information; inevitably they are susceptible to security threats concerning data tampering and ownership rights. The outsourcing data results in a number of threats such as alteration of data, deletion of data, accessing by unauthorized people, producing unauthorized copies. Recently, many methods are proposed to watermark databases to protect digital rights of owners. When the watermark is enforced on data for protecting ownership right, the data quality get compromised. For protecting the data quality a robust and reversible watermarking technique for numeric and non-numeric relational data is required. Particularly, watermarking techniques based on optimization draw attention, which results in improving watermark capacity and lower distortion. In this paper, A Robust Reversible Watermarking (RRW) with Genetic Algorithm (GA) and Firefly Algorithm (FFA) is proposed to embed watermark into relational databases. GA and FFA are optimization techniques which are biologically inspired. Best attribute values are selected efficiently by the FFA and later GA is used for creating the optimum watermark string which ensures reversibility without data quality loss. Experimental results indicate that FFA and GA have reduced complexity and results in improved watermark capacity and less distortion. Keywords— Reversible Digital Watermarking, Relational Database, Robustness, Genetic Algorithm, Firefly Optimization Algorithm.