A3C-Based and Dependency-Aware Computation Offloading and Service Caching in Digital Twin Edge Networks

The combination of Mobile Edge Computing (MEC) and Digital Twin (DT) is anticipated to enhance the quality mobile application services in 6G era. However, current research often overlooks service caching task dependency, which may deteriorate system performance. Moreover, Servers (ESs) have limited computing resources capacities, require collaboration meet user demands. To address these challenges, we propose a DT-empowered MEC architecture that supports Users (MUs) offloading dependency-aware tasks, while considering edge collaboration. objective jointly optimize computation resource allocation minimize system’s energy consumption. Hence, this problem can be formulated as Mixed Integer Non-linear Programming (MINLP) addressed by utilizing Asynchronous Advantage Actor-Critic (A3C)-based method. Extensive simulation results demonstrate our approach outperforms other benchmark algorithms under various scenarios, significantly reducing