Robust UAV Jittering and Task Scheduling in Mobile Edge Computing with Data Compression

TL;DR

This work tackles energy-efficient operation in UAV-enabled mobile edge computing under three-dimensional UAV jitter. It casts joint trajectory, data compression, offloading, and edge/computation resource allocation as a Markov decision process and solves it with the randomized ensembled double Q-learning (REDQ) algorithm, leveraging multiple Q-functions and entropy regularization for robust, efficient learning in continuous action spaces. The proposed REDQ-based method accounts for data compression effects and enables cooperation between a UAV and a ground BS, achieving notable energy savings compared with PPO and A2C baselines in extensive simulations. The study provides a practical framework for robust MEC in complex environments and points to promising directions for multi-UAV extensions.

Abstract

Data compression technology is able to reduce data size, which can be applied to lower the cost of task offloading in mobile edge computing (MEC). This paper addresses the practical challenges for robust trajectory and scheduling optimization based on data compression in the unmanned aerial vehicle (UAV)-assisted MEC, aiming to minimize the sum energy cost of terminal users while maintaining robust performance during UAV flight. Considering the non-convexity of the problem and the dynamic nature of the scenario, the optimization problem is reformulated as a Markov decision process. Then, a randomized ensembled double Q-learning (REDQ) algorithm is adopted to solve the issue. The algorithm allows for higher feasible update-to-data ratio, enabling more effective learning from observed data. The simulation results show that the proposed scheme effectively reduces the energy consumption while ensuring flight robustness. Compared to the PPO and A2C algorithms, energy consumption is reduced by approximately $21.9\%$ and $35.4\%$, respectively. This method demonstrates significant advantages in complex environments and holds great potential for practical applications.

Figures (9)

• Figure 1: UAV-assisted MEC system with UAV jittering.
• Figure 2: Architecture diagram of the REDQ algorithm.
• Figure 3: Convergence performances of different algorithms.
• Figure 4: Performance comparison versus different number of users.
• Figure 5: Outage probability under different positioning errors.