Research On Privacy-Preserving Shared Storage And Incentive Mechanism Via Blockchain

Recently,the digitalization revolution of the real world has contributed a large volume of data.Although the emergence of cloud storage systems have brought great benefits for data owners with higher flexibility and efficiency,high storage cost,single point of failure,and vulnerability to attacks have been three crucial issues in the centralized cloud storage.Compared to the centralization,decentralized storage has the advantage of better scalability,higher reliability,and lower cost,hence is a potential method to solve the aforementioned problem.Meanwhile,blockchain technology,as one of the representative decentralized storage technologies,has raised people's wide concerns in both academia and industry.Leveraging blockchain-based sharing storage system,some idle resources can be aggregated to provide low-cost and highly reliable storage services for users.It,however,also brings some crucial issues such as security,privacy disclosure,and incentive provision.To address the aforementioned issues on data security,privacy protection,and incentive provision,in this paper,we aim to propose a privacy-aware sharing storage system with incentives via blockchain.Based on the limitations in existing works,we attempt to achieve reliable and efficient data sharing storage with privacy protection of users and storage nodes.In addition,nodes are stimulated to contribute resources and prevent nodes from misbehaviors.Specifically,our main contributions are listed as follows.(1)We propose a blockchain-based privacy-aware data storage and sharing scheme.The on-chain data are publicly visible,transparent,and immutable.Any network entity is able to obtain/query the transaction information and data on the blockchain.To meet the privacy protection requirements of users,we introduce the bloom filter and blind signature techniques,and design information hiding and verification algorithms to conceal the meta-data recorded on the blockchain,with which we devise a privacy-aware data storage and sharing scheme.In addition,to address the conflict between data encryption and data deduplication,the slow hash function is introduced to improve the security of convergence encryption,i.e.,the difficulty of being attacked by the violent dictionary attacks for low-entropy data.Meanwhile,we design algorithms for secure data deduplication between storage nodes and the storage consortium,with which duplicate data can be recognized without revealing the meta-data at the storage nodes in a trusted third party free manner.Finally,we compare our scheme with existing researches in terms of data preprocessing,data storage,and data deduplication.The results show that the proposed scheme meets the defined security goals with higher efficiency.(2)We propose an incentive mechanism and a fair protocol for the storage sharing system.First,a fine-grained and win-win incentive mechanism is designed based on the magnifying effects of data deduplication,in which the data type and bandwidth are separately priced to improve the profit of nodes while the bid contract is used to decrease the user's expense.Second,to tackle the misbehaviors of malicious users,we design node trust evaluation model based on the transaction results,in which the node trust is computed in multiple dimensions.Moreover,in order to establish trust for new nodes and prevent the misbehaviors of malicious nodes,we introduce deposit scheme and preset cost,respectively.To deal with the potential denial of transactions and the low scalability of blockchain,we adopt smart contract to design fair off-chain transaction protocols.This protocol realizes the fairness of transactions via deposit,and achieves timely transaction confirmation without transaction fees by separating the frequent traffic from the blockchain main network.Finally,experimental results demonstrate the effectiveness and efficiency of our scheme.(3)We implement a privacy-preserving shared storage system based on blockchain.Specifically,we establish a prototype system for privacy-aware data storage and sharing,using go cryptography library,web3.js library,smart contract,and truffle framework etc.The system logic and algorithms are programmed with go language,and the interaction between RESTful API and front-end UI is designed by the interaction between the rpc protocol and the blockchain.According to the system requirement,we divide the system into four modular: smart contract,data storage,data deduplication,and front-end UI.The corresponding functionalities are shown in this paper.