Attribute-based multiparty searchable encryption model for privacy protection of text data

The problems of data storage and sharing have been well solved with cloud storage. However, the disadvantage is that users' messages are stored in the cloud without encryption protection. Cloud storage managers can access or even obtain user data, which brings significant security risks to data owners. Therefore, some cloud storage centers adopt ciphertext storage, which can protect them from the risk of user privacy disclosure. However, there are also problems, that is, how to find the desired data achieved in the cloud server search and how to ensure information integrity in the search process. Moreover, the encryption operation destroys the value and size relation of the original plaintext data. It does not have semantic and statistical characteristics when retrieving. Combining the hidden access structure in an attribute-based file searchable encryption model with proxy Re-Encryption technology, we propose an attribute-based file retrieval model with a partially hidden access structure supporting Proxy Re-Encryption. The model solves the above problems effectively and supports keyword updating. Finally, the security of the model is proved based on the hypothesis of D-linear and decisional q-parallel bilinear Diffie-Hellman exponent under the random oracle model.