Differentially Private Data Sharing in a Cloud Federation with Blockchain

Abstract

Cloud federation is an emergent Cloud-computing paradigm that allows services from different Cloud systems to be aggregated in a single pool. To support secure data-sharing in a Cloud federation, anonymisation services that obfuscate sensitive datasets under differential privacy have been recently proposed. However, by outsourcing data protection to the Cloud, data owners lose control over their data, raising privacy concerns. This is even more compelling in multi-query scenarios where maintaining privacy amounts to controlling the allocation of so-called privacy budget. In this paper we propose a blockchain-based approach that enables data owners to control the anonymisation process, and enhances the security of the services. Our approach relies on blockchain to validate the usage of privacy budget and adaptively change its allocation via smart contracts, depending on the privacy requirements provided by data owners. Prototype implementation with the Hyperledger permissioned blockchain validates our approach with respect to privacy guarantee and practicality. Cloud federation builds up interconnectivity and cooperation among already deployed clouds, enabling organisations to achieve various business goals, such as controlled sharing of data, services and optimisation of computing resources usage 1-3. To support secure sharing of federated data, anonymisation services have been proposed as a building component of Federation-as-a-Service (FaaS), 3-4 a recent Cloud federation solution. This component implements differential privacy in order to obfuscate the result of statistical queries towards sensitive datasets, 5 enabling its privacy-preserving sharing. Offering this service in the context of a Cloud federation has benefits-access to multiple data sources and different service providers-but raises significant challenges for privacy management: sensitive datasets

DNSTSM: DNS Cache Resources Trusted Sharing Model Based on Consortium Blockchain

“…Yang et al [20] proposed a Differentially Private Data Sharing scheme based on the blockchain. The solution builds different cloud servers into a cloud federation for interoperability and collaboration, and enables privacy owners to control the anonymization process and enhance service security through a privacy technology.…”

A Self-Controllable and Balanced Data Sharing Model

“…There are two different solutions: one is proposed by Gkatzelis et al [43], which bundle the buyer's demands to decrease the price C has to pay. The other is proposed by LI et al [35], [45], a state-memory access approach, in which the system checks whether a history released result can approximate the current query.…”

“…Ouaddah [32] proposes a blockchain-based research data sharing framework that incentives O by digital tokens. Yang et al [45] presents a blockchain-based approach that enables O to control the anonymization process, and enhances the security of the services, by validating the usage of privacy budget and adaptively changing its allocation via smart contracts according to the privacy requirements of O.…”

Privacy-Preserving Mechanism in Smart Home Using Blockchain

“…Therefore, various researchers [12,[32][33][34][35] have turned-out the attention towards distributed Frameworks and proposed popular blockchain based solutions for various IoT use cases. Furthermore, because in the design of blockchain-based IoT systems, privacy is not pre-enforced and private data can be leaked using certain attacking approaches, researchers proposed various privacy preservation strategies such as differential privacy for different applications of blockchain based on cloud computing and machine learning [49][50][51][52][53][54][55][56][57][58][59][60][61].…”

“…To address these flaws, researchers now look to employ strategies around the integration of privacy protection with blockchain-based cloud as a possible solution. In [60], authors undertake the integration to distribute autonomous privacy budget when mining in blockchain. The integration resulted in increased work-load while executing queries, with the authors claiming that the method both provides answers to queries more effectively as well as protect user privacy.…”