VPAMS: Verifiable and practical attribute-based multi-keyword search over encrypted cloud data

Abstract

Abstract With fast development of cloud storage, people increasingly outsource their files to the cloud server. To protect the confidentiality of outsourced files, files often are encrypted before outsourced. However, the accurate location and search on encrypted files will be greatly limited. This paper mainly considers how to realize an accurate multi-keyword fine-grained search on the encrypted files. Firstly, we use multiple keywords and an improved k-nearest neighbor (k-NN) technology to improve the search accuracy. Secondly, our scheme can protect the privacy of the file keyword indexes and the search queries and further preserve the privacy of outsourced files and users. Thirdly, unlike other existing multi-keyword search schemes without considering the verification and decryption of search results, the data users in our scheme not only can check whether the returned files contain multiple queried keywords, but also can realize the authorized decryption of the search files by using attribute-based encryption. Finally, performance evaluations show the time costs of our search algorithm is a constant value and independent of the number of user attributes. Therefore, our scheme is more efficient and practical.