Real-Time Task Scheduling for Digital Twin Edge Network.

The deployment of digital twins (DTs) at the edge of a wireless network can facilitate low-latency and high-throughput DT autonomous and real-time Internet of everything (IoE) applications. In such DT edge networks (DTENs), each DT has two types of real-time tasks that require timely processing: DT update tasks and DT inference tasks. However, the joint scheduling of these two types of tasks has been overlooked in prior works. In this paper, the first joint real-time scheduling scheme for DT update and inference tasks in a DTEN is proposed. Moreover, a novel performance metric called freshness is introduced to capture the effectiveness and synchronization performance of scheduling. Also, a new scheduling scheme is proposed to efficiently solve a freshness maximization problem for DTENs. Simulation results show that the performance of the proposed scheme is within 4% of the upper bound for DTENs with 20 physical objects, and within 12% of the upper bound in worst cases for DTENs with more than 30 physical objects. The results also show that the proposed approach reduces the maximum de-synchronization time by 63% compared to existing real-time scheduling algorithms.