Secure Verifiability of Outsourced Data with Client Verification

Abstract

The information security requirements are increasing due to huge expansion of data in various fields. For the Clients with the limited resources, dealing with huge data is always been a significant problem. Cloud computing offers a variety of interconnected resources to its users. Heavy computations are always been a problem for resource constrained devices. This problem is reduced to a greater limit as cloud computing provide ease to the users by providing the facility of outsourcing their different complex computational work along with utilizing the huge data storage of the Cloud servers. At the same time, the existence of untrusted service provider creates the problem of lack of effective controllability of outsourced data by the legitimate users, as clients may not be certain enough to have a complete smooth control over outsourced data. Hence the need to have secure outsourced data gets more strengthened. In this thesis, we have suggested a secure mechanism for outsourcing the data, its update, data / client verification along with data encryption. The cloud server is asked to have complex computations and return the desired data. In case of any change, corruption or dummy data is returned by the cloud server, the querying client identifies the error, and can change or update the existing data securely. This whole process is monitored by the client through generation of proofs. Moreover, in contrast to existing works in literature, the proposed scheme takes into consideration of the ownership verification, prior to entertaining any of the queries. The major intend is to have a higher level of efficiency and reduced computational costs for clients. The performance evaluation shows its effectiveness for resource constrained devices.