Minimizing long-term energy consumption in RIS-assisted AAV-enabled MEC network

Abstract

In recent years, autonomous aerial vehicles (AAVs) are increasingly becoming flight-based communicative and computing platforms, but the scarcity of communication resources can significantly hinder their performance and scalability. Therefore, this article proposes a reconfigurable intelligent surface (RIS)-assisted AAV-enabled Mobile Edge Computing (MEC) network, aiming to reduce long-term energy consumption while maintaining system stability by jointly optimizing computing resources, time slot allocation, transmit power, RIS phase angles, and AAV trajectory. By applying the Lyapunov method, we transform the long-term stochastic optimization problem into manageable deterministic online subproblems, and obtain approximate optimal solutions using successive convex approximation, penalty functions, and convex optimization techniques. Simulation results show that compared to the baseline scheme, the proposed scheme approximately reduces energy consumption by 10%, improves system stability by approximately 16%, and maintains computational efficiency.