Research On Attribute-based Searchable Encryption Mechanism

Cloud storage,as a new type of Internet storage,is rapidly favored by enterprises and individual users because of its flexible storage methods,powerful computing capabilities,and storage space.To ensure the security of data and privacy of users,data is usually stored on the remote cloud server in the form of ciphertext,but this is inconvenient for the ciphertext retrieval.Faced with the challenge of cloud server to retrieve ciphertext,searchable encryption mechanism emerged.This mechanism supports verify whether to store the target file on the cloud server by matching the trapdoor information with the keyword ciphertext without decryption.Searchable encryption are mainly classified into symmetric searchable encryption and public key-based searchable encryption.Public key-based searchable encryption can support the multi-user search model due to the separation of its public and private key pairs.Data users only need to search the keyword ciphertext through the search password,avoiding the process of pre-interactive operation in the symmetric searchable encryption,thus this mechanism has practical and broader application scenarios.As a very attractive cryptographic primitive,Attribute-based encryption supports the one-to-many data sharing model.The data holder specifies the attributes of the shared data user through an access policy.The data user can decrypt the ciphertext if and only if when attribute meets the access policy.Although most of the existing attribute-based searchable encryption schemes support fine-grained access control systems,the ciphertext length and search efficiency of these schemes increase linearly with the number of attributes,and the privacy of keywords in trapdoors cannot be ensured.Therefore,this paper uses a variety of cryptographic primitives,such as fast decryption,anonymity and asymmetrical groups,to propose an efficient Key-Policy Attribute-Based Searchable Encryption in Prime-Order Groups(KP-ABSE).This scheme not only satisfies many basic security requirements such as control search,hidden query and query independence,but also gains the advantages of attribute-based key management and supports fine-grained access control systems.This thesis mainly researches and analyzes public key-based searchable encryption mechanism from aspects of security and search efficiency,and achieves the following results:(1)In terms of security,the use of linear partitioning technology to achieve the anonymity of attribute,to prevent the adversary from using a logarithmic operation to guess the keyword attacks through the asymmetric group approach,and use of the public and private key pairs of the designed cloud server to solve trapdoors keywords privacy issues.(2)In terms of search efficiency,the keywords are divided into keyword names and keyword values.Keyword ciphertexts are aggregated using the fast decryption method,and the trapdoors are preprocessed before the search algorithm.The length of the ciphertext is reduced from the linear magnitude to the constant magnitude,and the number of logarithm operations in the search algorithm is reduced to a constant number.It not only reduces the computation,transmission and storage costs of ciphertext,but also significantly improves the search efficiency.Finally,through simulation experiments and performance efficiency comparisons,it is further demonstrated that our scheme has high practicability and safety.