Research On Data Traceability Technology For Data Sharing And Distribution

With the increasing demand of data sharing and analysis in various industries,the security problem in data distribution becomes increasingly important.There is a risk of unauthorized disclosure after database distribution.How to realize traceability after data leakage in data sharing has become the focus of research.Digital watermarking and blockchain is the main data provenance technology.The work of the thesis focuses on the data sharing distribution scenario.In view of the shortcomings of watermarking and blockchain traceability methods in accuracy and traceability efficiency roots,this thesis proposes the solution based on digital watermarking and blockchain that can realize the data distribution to reveal the roots of accountability and data transfer process tracking.The main work of the thesis is as follows:(1)Existing digital watermarking traceability methods are difficult to balance the data statistical availability and traceability accuracy,and the watermark embedding is limited by numerical data.To solve problem above,a database zero watermark traceability algorithm based on preserving reference relationship and grouping sorting ZWRRGS is proposed.This thesis combines the database watermarking technology with the database table splitting method that retains the reference relationship to realize the rapid positioning of the watermark after data leakage,realizing the traceability of the zero watermark and reduce the traceability time.The zero-watermark algorithm based on grouping sorting is designed to avoid the dependence of watermarking algorithm on the redundant space of numerical data.Positional relationship transformation between data packets is used to embed watermark information to ensure data accuracy unchanged.The concept of window density and centroid is introduced,and the sliding window strategy is used for obtaining more precise grouping position to improve traceability accuracy.(2)The existing blockchain data traceability mechanism has problems such as privacy disclosure and traceability efficiency bottleneck.Aiming at above issues,a data traceability method for privacy protection based on blockchain is proposed.A traceability data protection algorithm based on local differential privacy is proposed to protect privacy of traceability data transaction.To reduce traversal of invalid blocks and transactions,the bloom filter is introduced into the Merkel tree to reconstruct the Merkel index and the traceability retrieval efficiency of shared records is improved.Pre-order transaction index is introduced into traceability data model to realize efficient traceability based on transaction chain.Theoretical analysis and experimental results demonstrate the effectiveness of the proposed method.