Font Size: a A A

Research And Application Of Semi-structured Data Storage Method On Blockchain

Posted on:2024-02-14Degree:MasterType:Thesis
Country:ChinaCandidate:Y SiFull Text:PDF
GTID:2568307124460124Subject:Electronic information
Abstract/Summary:PDF Full Text Request
Blockchain is widely used in intellectual property protection and data sharing because of its decentralized,tamper-evident,and traceable features,which solve the security problems of traditional databases in data storage and management.However,most existing research on optimizing blockchain data storage focuses on standardized structured data,and there needs to be more research on semi-structured data with variable structure.The application of semi-structured data such as XML,JSON,and other formats is expanding,and it is essential to study how to chain them for secure storage.In this thesis,XML data is used for the research object and study of the storage method of semi-structured data on the blockchain.Firstly,a relational mapping of XML data is performed,and a storage method based on model mapping is proposed.Then,to further improve the retrieval efficiency of XML data on the blockchain,the block structure is improved based on the mapping relationship table,and the mapped relational table is stored on the chain.Finally,a federation chain-based paper submission system is designed and implemented based on the proposed storage method.The primary research is as follows:(1)A storage method based on model mapping is proposed for the relational mapping of XML data.The method represents XML document data as an XML document tree,designs four relational tables { file,path,innernode,data} as the mapping model of XML document data according to the type and structural relationship of each node in the tree,maps XML document data to a relational database,and uses relational database technology to store and manage XML data.(2)Improving the block structure based on the relationship table generated after mapping an X-BM structure based on the Merkle tree is proposed to improve the traditional block structure based on the relationship table generated after mapping XML documents.Firstly,the XML document IDs in the file table are used as an element to construct the balanced binary tree;secondly,all nodes are traversed bottom-up from the leaf nodes of the balanced binary tree,and the hash values of their parent nodes are obtained by hashing the neighboring nodes two by two.Meanwhile,the maximum and minimum values of the IDs of all XML documents in the merged range are kept.The location is recorded in the balanced mapping address in the binary tree,and the Bloom Filter is generated by XML document keywords,and repeat this process until the root of the X-BM tree is generated;finally,a query algorithm of XML document IDs and a query algorithm for XML document keywords are proposed based on this structure.The method’s effectiveness is demonstrated by comparing it with the traditional index structure and query methods.(3)This thesis designs and implements a consortium blockchain-based paper submission system based on the above-proposed storage method.The manuscript information is uploaded to the consortium chain platform as semi-structured data for review.Meanwhile,the review data is stored so that the data on the chain is tamperevident and traceable.Finally,by testing the system,the usability and practicality of the proposed method are demonstrated,which is vital for protecting academic results before publication.
Keywords/Search Tags:Blockchain, Semi-structured Data, XML Data, Merkle Tree, Model Mapping
PDF Full Text Request
Related items