Research On Secure Data-sharing And Computable Encryption Schemes Over Outsourced Data In A Cloud Environment

With the development and popularization of cloud computing,services provided by the cloud platform such as storage and computing have been favored by more and more individuals and enterprises.Despite the rapid development of cloud services,the majority of users are reluctant to outsource sensitive data to the public cloud due to privacy issues.Various kinks of encryption methods have emerged for the purpose of protecting data privacy.However,encryption increases the difficulty and cost of data processing.What's worse,it will be very difficult to perform advanced functions provided by cloud services over encrypted data.On the one hand,most of the existing data sharing encryption researches ignore the communication overhead and workload of the data owners and users during data sharing.On the other hand,the cloud computing services over encrypted sharing data have not yet been effectively addressed.In particular,the implementation of the influence maximization calculation which aims at selecting a seed set that maximize its overall influence to all network through propagation over encrypted data in social networks still faces great challenges.This paper summarizes and analyzes the difficulties of existing encryption schemes in addressing the secure sharing and computable problem over outsourced data.In addition,the existing influence maximization works can not solve the privacy issues when performing the influence maximization function in a cloud environment.In order to solve these problems,we proposed three solutions,including:(1)We proposed a secure outsourced data sharing encryption scheme(SDS2),considering the communication overhead and workload of the data owners and users during data sharing.We proposed a secure data sharing protocol that allows the data owner to assign authorized users to share its outsourced data at any time.Our scheme can effectively ensure the data privacy of the data owner and reduce the data owner's computing and communication overhead during data sharing.Furthermore,our scheme effectively reduces the computing cost of sharing users through the half-decryption mechanism.(2)In order to let the users perform simple computation over encrypted shared data and reduce the extra communication and computing cost of the users,we proposed a secure data-sharing and computable encryption scheme(SDCS).A secure data-sharing and computable encryption protocol allows data owners to designate specific users to share outsourced data.Authorized users can cooperate with the CSP to perform computations over shared data to obtain some useful shared data.This solution ensures that data owner's data privacy and reduces the additional computing and communication overhead of sharing users.(3)To satisfy the influence maximization computing demand of users and address the issues of letting the cloud perform influence maximization function over encrypted outsourced data,we proposed an influence maximization encryption scheme.Firstly,we built a SWHE-based propagation estimation module that could securely compute influence propagation given arbitrary input seed set.We proposed a secure influence maximization query protocol using the influence propagation module as the building block.We solved two major technical challenges in our work: “node activation judge” and the limited number of consecutive multiplications of SWHE.Our scheme also supports user-designated source and target group selection.Our proposed scheme can let sharing users securely and efficiently perform influence maximization function over shared data by collaborating with the CSP in a cloud environment.