The Design And Optimization Of Searchable Encryption Scheme Based On Homomorphic Encryption And CP-ABE

With the rapid development of cloud technology,a large number of users choose to outsource their data to the public cloud in order to save costs.Because the public cloud has the characteristics of honest but curious,in order to ensure the cloud data confidentiality,data owners usually to encrypt the data,which will cause the legitimate visitors to perform ciphertext search operation in the cloud very challenging,so searchable encryption has become a hot research topic in the field of cloud computing and information security.However,the existing searchable encryption schemes support a limited number of search keywords,and cannot accurately calculate the relevance and matching degree between the search items and the documents to be searched,and the user experience is poor.In addition,some existing schemes also have the defects of low search efficiency and poor practicability.For the search permission of the cloud privacy data,most existing schemes do not provide flexible access policies.Therefore,with respect to searchable encryption,there is not a scheme that satisfies reliability,efficiency,and flexibility at the same time.This paper will analyze and explore in this direction,conduct in-depth research on searchable encryption schemes,and achieve the following results:(1)First,a searchable encryption initial scheme is designed.By using integer DGHV full homomorphic encryption as the encryption algorithm,the CP-ABE as the flexible search permission control algorithm,the TF-IDF-based vector space model as the similarity calculation and result ranking model,and finally relying on Hadoop platform to achieve the initial searchable encryption scheme.After security analysis,we can see that the initial scheme has data confidentiality,privacy and confidentiality of the index and keywords,query request and ciphertext search privacy.The experimental results show that the initial scheme has higher ciphertext searchefficiency than the existing schemes,and has the search result correctness that is equivalent to the plaintext search Lucene.(2)In order to improve the ciphertext search performance and optimize the access control strategy,based on the initial scheme,performance optimization and function expansion are carried out,and an optimized searchable encryption scheme is proposed.The optimization scheme uses Random Asymmetric Splitting(RAS)algorithm instead of homomorphic encryption algorithm to encrypt the weight vector,which also replaces the original CP-ABE with a more flexible R-CP-ABE supporting the revocation mechanism,and further optimizes the similarity computation based on TF-IDF vector space model(VSM),finally uses Spark instead of Hadoop's Map Reduce to implement the optimized searchable encryption scheme.After security analysis,it can be seen that the optimization scheme not only has the security features of the initial scheme,but also has the non-repudiation of the query request,the disassociation of the query,and the higher reliability.The experimental results show that the optimization scheme has a higher search efficiency than the initial scheme on the premise of ensuring the correctness of the search.