kNN-R: Building Secure and Efficient Outsourced kNN Query Service with the RASP Encryption

Xu, Huiqi. M.S., Department of Computer Science and Engineering, Wright State University, 2012. kNN-R: Building Secure and Efficient Outsourced kNN Query Service with the RASP encryption. With the wide deployment of public cloud computing infrastructures, outsourcing database services to the cloud has become an appealing solution to save operating expense. However, some databases might be so sensitive or precious that the data owner does not want to move to the cloud unless the security is guaranteed. On the other hand, a secure outsourced service should still provide efficient query processing and significantly reduce the inhouse workload to fully realize the benefits of outsourcing. We summarize these key features for an outsourced service as the CPEL criteria: data Confidentiality, query Privacy, Efficient query processing, and Low inhouse workload. Bearing the CPEL criteria in mind, we propose an encryption called RASP to provide query services for range query and k nearest neighbors in secure outsourced databases. In the RASP encryption, data confidentiality and query privacy are guaranteed when applying it for range query and kNN. Efficient query processing is achieved by two aspects: (1) all encrypted data can be indexed to speedup query processing using RTree; (2) The protocol for k nearest search in outsourced databases can find high precision kNN results, which also minimizes costs between the cloud server and the inhouse client. High precision kNN results and minimized interactions result in low inhouse workload. In addition, we have conducted a thorough security analysis on data confidentiality and query privacy.