Research On Multi User Fuzzy Encryption Search Scheme For Blockchain

Blockchain decentralization enables users to operate anonymously and solves the problem of security risks associated with local data management.Users deposit data documents into the blockchain,which not only reduces their own storage expenses,but also protects data privacy and security,the most effective is to avoid leakage of user data in the process of data transmission and storage.The major problem at present is that the on-chain data storage has limitations,limited storage space and scattered storage locations,which are not conducive to performing on-chain searches.In the research on encrypted data search technology on the chain,there are few existing encrypted search schemes This thesis proposes a multi-user fuzzy keyword search scheme for blockchain,which uses blockchain technology to synchronize data on the chain and off the chain.User data documents are encrypted and stored in cloud storage,and only part of the information is retained on the chain.While improving the efficiency of fuzzy keyword search,it realizes multi-user access control and data verification.The main work of this thesis includes:1.To address the problems of poor fault tolerance and low query efficiency of keywords,this thesis constructs a KMLSH hybrid index query scheme.The word feature vector is constructed using Bigram method on the original data file,and the Mini Batch K-Means clustering algorithm is introduced to calculate the cluster class center position,which is computed by the position sensitive hash function(LSH)to obtain the binary hash vector of keywords.It indexes the keyword hash vectors backwards and matches the hash vectors in the threshold range to achieve fuzzy keyword search,resolving the drawback of poor filtering effect of Bloom filter.2.In response to the problem of single user and lack of privacy protection in query verification mechanisms.Constructs a fuzzy encryption search scheme based on blockchain.The ring signature technology is adopted to solve the multi-user identity authentication on the chain and realize the user identity information hiding.Write a smart contract and package it on the chain.The fair payment of query validation and transaction execution is achieved through log files on the chain,and the integrity of the returned data documents is verified,reducing manual processing costs.3.The effectiveness of the proposed scheme is verified through experiments.We select3500 thesis and journals as the test dataset,build the system model to conduct and perform extensive simulation experiments.The scheme proposed in this thesis can achieve fuzzy keyword search of encrypted data documents and improve search efficiency while protecting user privacy and security and verifiable operations.