PIMRS: achieving privacy and integrity-preserving multi-owner ranked-keyword search over encrypted cloud data

Abstract

Because of the flexibility and convenience brought by cloud computing, it has been adopted in many applications. To preserve the privacy of cloud data, data owner often encrypts all sensitive data files, which makes the keyword search application based on plaintext a very challenging task. Therefore, several privacy‐preserving keyword search algorithms have been developed recently, and most of these works support only single‐data‐owner settings. However, there are always more than one data owners in real applications, which are much more complex and challenging than single‐owner scenario. To support multi‐owner keyword search, those prior search algorithms need to be repeated several times, because each data owner intends to encrypt his own files with a unique private‐key separately. It is absolutely not an efficient way. In this paper, we propose a privacy and integrity‐preserving multi‐owner ranked‐keyword search scheme termed PIMRS. In the PIMRS, we exploit an asymmetric scalar‐product encryption function based on the TF × IDF rule to preserve data privacy and to obtain more precise search results. Furthermore, a circular bi‐direction‐linked list based scheme is proposed to preserve the integrity of search results, which also enables the misbehaviors of cloud server to be detected. The security analysis of PIMRS shows its privacy and integrity property, and extensive experiments based on real‐world data set confirm the high efficiency of proposed schemes. Copyright © 2016 John Wiley & Sons, Ltd.