Research On Secure And Efficient Data Traceability Based On Double Blockchain Model

The rapid development of social economy promotes the growth rate of data.In order to ensure traceable data sources,traceable whereabouts and traceable responsibility,data traceability technology is particularly important.However,centralized traceability mode is not easy to supervised and can not guarantee the credibility and security of data.Blockchain as a tamper-resistant,decentralized technology ensures high data transparency and easier to trace,but the existing blockchain traceability system query efficiency is low,the lack of regulation,there is no guarantee that the chain on the authenticity of the data,and the inevitable in the process of traceability are privacy sensitive data,due to the transparent blockchain,The shortcomings of privacy protection in the process of traceability.In view of the above problems,the main research contents of this dissertation are as follows:(1)The traceability model based on double blockchain(DBTM)is proposed to solve the problems of single point of failure,information island,low transaction throughput and lack of supervision in the traditional Traceability Model.In this model,multiple private chains are anchored on the alliance blockchain,and data information is stored on the alliance blockchain,and hash summaries of private chain transactions are stored on the alliance blockchain to ensure the authenticity and reliability of private blockchain information.The data in the model is divided into publicly shareable data and privacy-protected data,and the storage content is refined to protect privacy security.Through comparative analysis of the existing blockchain traceability model,the model has better performance in throughput,data security,query efficiency and other aspects.(2)To solve the problem of low transaction traceability speed in existing blockchain,an efficient data query method based on DBTM is proposed.Firstly,a new structure BMerkle tree is proposed,which can quickly search the existence of transactions and search the specific address and content of transactions within the block.Then,by adding multilevel indexes outside the private chain,an optimized hop table structure,hot＿skiplist is proposed,which can quickly locate blocks and dynamically adjust the index hierarchy according to the access frequency of blocks,thus reducing the consumption of system space and improving the query efficiency.Experimental results show that BMerkle tree and hot＿skiplist index structure can effectively improve the transaction query efficiency.(3)In view of the problems that privacy traceability data is easy to leak and difficult to share in the existing blockchain traceability,a DBTM based secure traceability data sharing method was proposed.The improved Paillier public key encryption algorithm is used to encrypt the key privacy data and store it in the chain to ensure that the private data on the chain will not be leaked.In the process of tracing query,the homomorphism of Paillier algorithm is used to deeply process the data without knowing the specific data,so as to mine and analyze the accurate information of the data.In order to prevent data leakage caused by channel monitoring,secret sharing mechanism is used to share deeply processed data securely.Experiments verify that the improved Paillier public key encryption algorithm can effectively reduce the time consumption in the initialization stage and encryption stage,and the homomorphic encryption and secret sharing of private data are tested,which proves that the method has strong privacy protection and sharing performance.