Privacy-Aware Online Task Offloading for Mobile-Edge Computing

Mobile edge computing (MEC) has been envisaged as one of the most promising technologies in fifth generation (5G) mobile networks. It allows devices to offload their computation-demanding and latency-critical tasks resource-rich MEC servers. Accordingly, can significantly improve latency performance reduce energy consumption for devices. Nonetheless, privacy leakage may occur during task offloading process. Most existing works ignored these issues or just investigated system-level solution MEC. Privacy-aware user-level optimization problems receive much less attention. In order tackle challenges, a privacy-preserving device-managed scheme is proposed this paper This achieve near-optimal while protecting location usage pattern users. Firstly, we formulate joint problem preservation semiparametric contextual multi-armed bandit (MAB) problem, which relaxed reward model. Then, propose privacy-aware online (PAOTO) algorithm based on transformed Thompson sampling (TS) architecture, through (1) best possible delay performance, (2) goal preserving privacy, (3) obtain an policy without requiring any information. Simulation results demonstrate that outperforms methods terms minimizing system cost