Multi-keyword Ranked Search with Fine-grained Access Control over Encrypted Cloud Data

Abstract

Abstract. With the development of cloud computing, people have become accustomed to using these cheap and convenient services provided by the cloud server for data storage and processing. For protecting data privacy from leaking information to unauthorized users, probably including the cloud service provider (CSP), the data must be stored in an encrypted form before outsourcing, which leads to some traditional data services being unavailable, for example, data search. Therefore, to design a privacy-preserving data search scheme is an extremely challenging problem. Meanwhile in order to keep data from using (and even searching) by unauthorized users, it also needs to make sure that the search scheme could achieve fine-grained access control on data users, according to the access policies made by the date owner (DO).In this paper, considering both the control of user's access authority and the privacy protection in data search, we have designed a multi-keyword searchable scheme with access control over encrypted cloud data. By using the technology of inverted indexes, we can rapidly filter out data files that users own authority to access, while methods of coordinate matching and secure inner product computation have also been introduced to realize the privacy-preserving multi-keyword search. Through analysis of privacy and efficiency, we prove that our scheme achieves the requirements of privacy protection under two different adversary models, and that it has relatively high efficiency from the perspective of time complexity.