Research On Multi-party Secure Searching Method For Metadata In Mobile Edge Computing

The popularity of intelligent mobile devices makes mobile edge computing become one of the defining technologies for envisioning the Internet of Things.The data generated by the devices have the characteristics of massive multi-source heterogeneity,making data management and data security challenges.As the basis of data governance,metadata is an important aid to make full use of raw data and has been widely used in academia and industry.Nevertheless,the security and availability requirements of metadata are contradictory in the decentralized environment of mobile edge computing.Existing searchable encryption solutions to solve such problems are built in a centralized cloud environment,which usually requires the existence of a trusted third party and is suitable for ordinary files.Due to the high cost of the trusted third party,limited equipment resources,and more potential attacks in mobile edge computing,traditional searchable encryption schemes are not competent for the secure sharing of metadata in mobile edge computing.To solve the above problems,we proposed a blockchain-based multi-party secure searching method for metadata in mobile edge computing.By providing trusted metadata,the massive,multi-source,and heterogeneous raw data can be used reliably and effectively.Firstly,based on the lightweight proxy re-encryption construction,the algorithm of metadata upload,search,and access control is designed,and the blockchain smart contract is introduced to realize the trusted storage and retrieval of metadata.To get rid of the dependence on traditional trusted third-party key management,the grouping multi-party key generation method is designed to improve the security of metadata.At the same time,to reduce the performance overhead caused by the introduction of blockchain,we analyze the existing blockchain platform design,and the optimization strategy of buffer upload is designed for the upload process,and a master-slave smart contract strategy is designed for the searching process,and a dynamic expansion method of the slave smart contract is proposed to synergistically improve the performance on the blockchain.Experiments show that compared with plaintext sharing,the proposed method only increases a small amount of time overhead.Compared with the benchmark method,the buffer upload strategy can greatly reduce the upload time of metadata and the storage overhead on the blockchain.The experimental data proves that the master-slave smart contract strategy avoids the risk of timeout of search transactions and reduces the response time.Compared with the baseline,the throughput and success rate of search transactions under concurrent conditions are improved.Security analysis shows that the proposed method has achieved good results in metadata integrity,privacy,access control,and availability,which proves the feasibility of blockchain for the sharing of key metadata.