Fast Multi-dimensional Range Queries on Encrypted Cloud Databases

With the adoption of cloud computing, data owners can store their datasets on clouds for lower cost and better performance. However, privacy issues compel sensitive data to be encrypted before outsourcing, which inevitably introduces challenges in terms of search functionalities. This paper considers the issue of multi-dimensional range queries on encrypted cloud databases. Prior schemes focusing on this issue are weak in either security or efficiency. In this paper, using our improved asymmetric scalar-product-preserving encryption, we present an innovative technique for the encrypted rectangle intersection problem. Based on this technique, we propose a tree-based method to handle multi-dimensional range queries in encrypted form. Thorough analysis demonstrates that our method is secure under the honest-but-curious model and the known-plaintext attack model. Experimental results on both real-life and artificial datasets and comprehensive comparisons with other schemes show the high efficiency of our proposed approach.