Secure rank correlation computation for IoT applications

Abstract

The internet of things (IoT) is emerging as the next mega-trend technology across the business spectrum. Due to significant advances in sensor and RFID technologies, smart devices are now connected via the internet and able to communicate with each other without human intervention. Hence, the amount of data collected about individuals has increased dramatically. The ability of machines and sensor devices to collect, transmit and share data has raised privacy concerns in IoT applications. Recently, privacy concerns about data collected by IoT applications have received an increasing amount of attention. Therefore, a series of security mechanisms must be available for data privacy protection. In this paper, we propose two secure correlation computation protocols for IoT applications. We concentrate on the asymmetric setting where only one party receives the computation result. We incorporate homomorphic cryptosystem into our computation to support operations in an encrypted form. Two experimental results in this paper demonstrate the performance of both computation protocols.