A novel hybrid grey wolf optimizer with variable neighbourhood search algorithm for an efficient distributed cloud storage system

Abstract

Cloud computing has been viewed as the future of the IT industry. Data storage has been recognized as one of the main concerns of information technology. The bene ts of network-based applications have led to the transition from server-attached storage to distributed storage. Security is the main aspect of secure data transmission over unreliable network. Cryptography uses the encryption technique to send confidential messages through an insecure channel. Steganography techniques that modify the cover image in the spatial domain are known as spatial domain methods which involve encoding at the Least-Signi cant-Bit (LSBs) level. In image steganography image is a cover object for the hiding purpose. The work proposed a distributed data storage system in cloud where blowfish is used to encrypt the data stored in the cloud and the key is transmitted across the cloud network using image steganography. A Particle Swarm Optimization (PSO) algorithm is proposed for a distributed data storage system in cloud using steganography. A Variable Neighborhood Search algorithm is proposed. A hybrid Grey Wolf Optimization with variable neighbourhood search algorithm is proposed. The LSB-based approach is a popular type of steganographic algorithms in the spatial domain. LSB Matching (LSBM) employs a minor modi cation to LSB replacement. Pixel Value Differencing (PVD) considered as good steganographic algorithm due to its high payload and good visual perception in spatial domain. Results show that the Particle Swarm Optimization (PSO) performs better for 10% embedding rate by 5.01% and by 1.94% than LSB Matching and Pixel Value Differencing respectively for MRI Image - Brain. The PSO performs better for 10% embedding rate by 5.29% and by 2.15% than LSB Matching and Pixel Value Differencing respectivelyfor Ultrasound Kidney. Similarly the PSO performs better for 40% embedding rate by 4.45% and by 1.91% than LSB Matching and Pixel Value Differencing respectively for MRI Image - Brain. The PSO performs better for 40% embedding rate b