Digital twins (DTs) in mobile edge computing (MEC) networks create virtual entities for IoT devices to provide query services. The required data is routed by a service function chain (SFC)—an ordered sequence of virtual network functions (VNFs). Quality of service (QoS) depends on data freshness, measured by Age of Information (AoI), and service makespan. However, due to limited DT update budgets and edge computing resources, waiting for DT updates for fresher data leads to higher service makespan; balancing data freshness and service makespan to maximize QoS is challenging, but can be done by determining DT updates and VNF scheduling. In this paper, we consider the problem of AoI-aware VNF scheduling with DT updates. We first formulate it as an integer nonlinear programming and prove it is NP-hard. Then, we propose an improved genetic algorithm featuring three-layer chromosome encoding, hybrid initialization, adaptive crossover and mutation, and a repair mechanism. Extensive experiments demonstrate the superiority of the proposed algorithm over baseline methods.
Liao et al. (Thu,) studied this question.