Abstract
Cloud-integrated Internet of Things (IoT) is emerging as the next-generation service platform that enables smart functionality worldwide. IoT applications such as smart grid and power systems, e-health, and body monitoring applications along with large-scale environmental and industrial monitoring are increasingly generating large amounts of data that can conveniently be analyzed through cloud service provisioning. However, the nature of these applications mandates the use of secure and privacy-preserving implementation of services that ensures the integrity of data without any unwarranted exposure. This article explores the unique challenges and issues within this context of enabling secure cloud-based data analytics for the IoT. Three main applications are discussed in detail, with solutions outlined based on the use of fully homomorphic encryption systems to achieve data security and privacy over cloud-based analytical phases. The limitations of existing technologies are discussed and models proposed with regard to achieving high efficiency and accuracy in the provisioning of analytic services for encrypted data over a cloud platform.