Private range queries on outsourced databases

Abstract

With the advent of cloud computing, data owners could upload their databases to the cloud service provider to relief the burden of data storage and management. To protect sensitive data from the cloud, the data owner could publish an encrypted version of the original data. However, this will make data utilization much harder. In this paper, we consider the problem of private range query. Specifically, the data owner wants to obtain all the data within a query region, while keeping the query private to the service provider. Previous works only provide partial security guarantee, and are inefficient to deal with large scale datasets. To solve this problem, we present a fully secure scheme based on private information retrieval (PIR) and batch codes (BC).