Secure kNN query of outsourced spatial data using two-cloud architecture

Cloud service providers can offer virtual servers, with powerful processors and massive storage capacity for nominal per-use charges. With the underlying resources, the cloud allows data owners to trade capital expense for the variable cost. The auto-scaling features of the cloud can control the consumption of computing resources and hence the associated cost. However, by saving data on the cloud, the privacy of data becomes at risk. Data becomes exposed to the untrusted cloud. The users' queries can also be monitored by attackers and cloud operators. Furthermore, the pattern of accessing data can be identified. The solution is to encrypt data before outsourcing to the cloud. However, answering queries over encrypted data becomes a challenge. This paper proposes a novel technique, called Secure Voronoi kNN (SVK), that answers k-nearest neighbor (kNN) queries over encrypted data in a two-cloud architecture. SVK can hide data access patterns from the cloud.

Automated summary

Cloud service providers offer virtual servers with powerful processors and large storage capacity at nominal fees. However, storing data on the cloud poses privacy risks, and user queries can be monitored by attackers and cloud operators.