Research On Multi-keyword Searchable Encryption Technology Based On Edge Computing

With the advent of the Internet of Things era,numerous devices have been incorporated into the Internet,generating massive amounts of data.Edge computing relies on the advantages of low latency,on-demand allocation,and efficient use of resources,which complements the shortcomings of the original cloud computing.Data is quickly processed at the network edge,making the response speed faster.Outsourcing local data to cloud servers will cause data security problems.If we store private data on cloud servers in plaintext,once the data is leaked,it may be exploited by malicious attackers.In the past two years,the epidemic has brought widespread attention to the medical industry,and the leakage of sensitive patient data has also emerged one after another.Although there are many privacy protection schemes for sensitive data storage and sharing,they are mainly based on cloud computing and cannot be applied to edgenetwork environments.The edgenetwork needs to quickly respond to requests from edge devices.Most of the existing solutions are mainly based on the traditional cloud computing model,and the communication delay between users and the cloud is high.The network environment of the edge computing model is complex,and data storage needs to meet higher security requirements.Aiming at these problems,this thesis proposes two indexsplittable sensitive data storage and sharing privacy protection schemes.This thesis proposes a storage and sharing privacy protection scheme based on medical sensitive data(SPSS).A newindex structure is constructed,which splits the index from a complete matrix index into multiple sub-indexes and stores them on edge nodes,and assigns trapdoors to them through the trusted authentication center,realizing range query first and then accurate search,which effectively reduces the computational cost of retrieval.At the same time,the length of the trapdoor in the SPSS is fixed,which is only related to the total number of keywords,and its query matching will not change with the increase of query keywords.In terms of security,since the cloud server only stores encrypted outsourced data and does not touch the data index,and users need TA to perform intermediate decryption when obtaining encrypted data,the relevant plaintext information of encrypted data cannot be obtained.In order to improve the security of the scheme and increase the functionality of the scheme,we further propose a privacy protection scheme(VSPSS)for the integrity verification of the retrieval results of sensitive medical data based on the SPSS scheme.First of all,the VSPSS changes the original matrix index to an encrypted decimal value,which improves the index security without affecting the index splitting operation.The index splitting can still be allocated to each edge node for storage.The inner product becomes the addition and multiplication operations on encrypted data.Although some retrieval efficiency is lost,the security of the scheme is improved.At the same time,the integrity verification of the query results and the dynamic update of files is also realized.In the SPSS program and VSPSS program proposed in this thesis,the corresponding security proof and performance analysis for the simulation results are given.Compared with the existing schemes,the SPSS and the VSPSS have higher security and lower computational overhead.