An enhanced energy efficiency scheme for secure computing in UAV-MEC networks

Mobile edge computing (MEC) mitigates terminal device computing demands by deploying cloud resources at the network's edge. In this MEC framework, unmanned aerial vehicles (UAVs) equipped with MEC servers enhance both uplink and downlink offloading due to their exceptional maneuverability and line-of-sight (LoS) connectivity. However, the wireless nature of UAV-MEC systems exposes sensitive data to potential eavesdropping. To address this concern, we formulate an optimisation challenge aimed at maximising data secrecy energy efficiency. This optimisation balances data and energy efficiency while preserving communication security. Due to the problem's time-varying and non-convex nature, we decompose it into four subproblems: terminal scheduling, local computing ratio, UAV transmit power, and UAV trajectory optimisation. Subsequently, we develop a hybrid iterative algorithm to maximise data secrecy energy efficiency during offloading. Simulations illustrate the algorithm can efficiently utilise terminal and MEC server computation capabilities, enhance system security, and improve energy efficiency while reducing energy consumption in task offloading.