Privacy-preserving task offloading strategy for digital twin system in satellite edge computing

Satellite edge computing faces a trade-off between privacy protection and system overhead, strict task-dependency constraints, and the limited ability of conventional algorithms to capture complex dependency topologies in digital twin task offloading. To address these issues, a computation offloading strategy based on a priority graph deep reinforcement learning algorithm was proposed. Firstly, a privacy-preserving satellite-terrestrial digital twin task offloading architecture was constructed to collaboratively optimize delay, energy consumption, and privacy protection. Secondly, a priority sorting mechanism was designed to satisfy dependency constraints and simplify the solution space. Finally, a priority-based graph deep reinforcement learning algorithm was developed that extracted topological features through a graph attention network to determine the optimal offloading strategy for the sorted tasks. Experimental results demonstrate that compared with conventional algorithms, the proposed algorithm reduces system delay by 4.52%~46.16%, satellite energy consumption by 3.42%~32.31%, and the number of compromised tasks by 2.69%~57.71%.

# Telecommunication # Satellite edge computing # Digital twin system # Priority scheduling # graph attention network # deep reinforcement learning

Lin Bing, Tan Xiao, He Kai, et al. (2026). "Privacy-preserving task offloading strategy for digital twin system in satellite edge computing". Tongxin xuebao.