Open AccessA secure deduplication scheme for encrypted data
Author(s) -
Vishal Passricha,
Ashish Chopra,
Pooja Sharma,
Shubhanshi Singhal
Publication year - 2019
Publication title -
international journal of informatics and communication technology/international journal of informatics and communication technology (ij-ict)
Language(s) - English
Resource type - Journals
eISSN - 2722-2616
pISSN - 2252-8776
DOI - 10.11591/ijict.v8i2.pp77-86
Subject(s) - data deduplication , computer science , encryption , cloud storage , computer security , cloud computing , data security , database , scheme (mathematics) , operating system , mathematical analysis , mathematics
Cloud storage (CS) is gaining much popularity nowadays because it offers low-cost and convenient network storage services. In this big data era, the explosive growth in digital data moves the users towards CS to store their massive data. This explosive growth of data causes a lot of storage pressure on CS systems because a large volume of this data is redundant. Data deduplication is a most-effective data reduction technique that identifies and eliminates the redundant data. Dynamic nature of data makes security and ownership of data as a very important issue. Proof-of-ownership schemes are a robust way to check the ownership claimed by any owner. However to protect the privacy of data, many users encrypt it before storing in CS. This method affects the deduplication process because encryption methods have varying characteristics. Convergent encryption (CE) scheme is widely used for secure data deduplication, but it destroys the message equality. Although, DupLESS provides strong privacy by enhancing CE, but it is also found insufficient. The problem with the CE-based scheme is that the user can decrypt the cloud data while he has lost his ownership. This paper addresses the problem of ownership revocation by proposing a secure deduplication scheme for encrypted data. The proposed scheme enhances the security against unauthorized encryption and poison attack on the predicted set of data.