Research On Blockchain-based Credible Data Publishing Technology And Applications

With the continuous development of digital society,data is growing at an exponential rate.Data publishing has become a kind of very important data infrastructure service.However,traditional data publishing schemes face new challenges in data security and data deduplication.On one hand,data publishing typically relies on a trusted third party(TTP)and data stored in centralized servers.The TTP has problems like single point of failure and internal/external attacks.At the same time,centralized servers are prone to hardware and software failures or system administrators intentionally or unintentionally operating the data so that data can be vulnerable to be tampered with and corrupted,the data security is threatened.On the other hand,traditional data publishing schemes do not support the mechanism of near duplicate data detection and easily store the near duplicate data multiple times,which will increase the data storage burden.Therefore,how to publish data between two or more entities in a credible and secure way is a problem to be solved urgently.For the problem of data security and data deduplication in data publishing,blockchain serves as a decentralized distributed ledger technology,which can provide a new technical approach for data publishing by utilizing its decentralized,tamper resistant and traceability.This paper mainly addresses data security and data deduplication in data publishing.Based on blockchain technology and convergent encryption(CE)technology,we propose a blockchain-based credible data publishing scheme(BCDP).The main results have been accepted by the international journal of Journal of Internet Technology(SCI source journal,Impact Factor: 1.930).Our main contributions can be summarized as follows:(1)We propose a blockchain-based credible data publishing scheme.The scheme has the following advantages: Firstly,the use of blockchain network to store metadata information can prevent tampering and deletion,and permanently save data publishing records.Convergence encryption technology can be used to protect the confidentiality of published data,and the key for individual data encryption is compromised without affecting the security of other data.Then,in order to solve the problem of data deduplication for data publishing,we construct a two-layer near duplicate data detection algorithm.In this algorithm,the data is divided into blocks by using asymmetric extremum content-defined chunking(AE-CDC)technology,then a counting bloom filter(CBF)is used to generate data tags.And the last,near duplicate data is detected according to the similarity of data tags.Therefore,our proposed scheme can save storage overhead and avoid near duplicate data being stored multiple times.(2)According to the blockchain-based credible data publishing scheme,we construct and exploit a blockchain-based academic platform service system.Based on the PyCrypto library and the EthereumJS TestRPC client,a private blockchain network based on Ethereum is built in the Linux environment.And a cloud service network is simulated by taking advantage of the MongoDB.Then,we implement the front-end and back-end of the system,providing Web interfaces for users,which achieves functions are as follows: Firstly,the system can realize permanent storage of metadata and integrity verification of full-text data.Secondly,we construct an academic scoring mechanism to achieve reward and punishment for academic publishing by utilizing smart contract.Finally,it provides the functions of publishing,querying and downloading data.The establishment and design method of the system has been applied for a patent.