Secure Edge Computing in IoT via Online Learning

Abstract

To accommodate heterogeneous tasks in Internet of Things (IoT), a new communication and computing paradigm termed mobile edge computing emerges, extending the computing service from the cloud to edge, but at the same time exposing new challenges on security. The present paper studies online security-aware edge computing under jamming attacks. Leveraging online learning tools, we develop SAVE-S algorithm that is tailored for the stochastic jamming. Without utilizing extra resources such as spectrum and transmission power to evade jamming attacks, SAVE-S can select the most reliable server to offload computing tasks with minimal privacy and security concerns. It is analytically established that without any prior knowledge of future jamming information and server security risks, the proposed scheme can achieve O(√T) regret. Information sharing among devices can accelerate the security-aware computing tasks. Incorporating the information shared by other devices, SAVE-S has significant improvements on the sublinear regret, which is guaranteed by what we term value of cooperation. The effectiveness of proposed schemes are tested on both synthetic and real datasets.