Digital Twin-enabled Channel Access Control in Industrial Internet of Things

In this paper, we propose a novel channel access control scheme for industrial Internet of Things (IIoT). The scheme is named digital twin-based parallel learning and optimization (DtPLO), which aims to incorporate digital twin model into the network to enhance the decision-making efficiency for channel access with the consideration of industrial wireless environment. Specifically, we analyze outage probability that an IIoT device may fail to successfully transmit data while attempting to access the channel by employing a new statistical path-loss model obtained from measurements with the consideration of varying interference effects of the industrial wireless channel. Then, utilizing the obtained outage probability, we develop an advanced Markov decision process (MDP) model with the objective of optimizing channel access. This MDP model is further integrate with a digital twin model to establish a synergistic system that not only facilitates well-informed decision-making but also ensures dependable packet transmission for industrial operations. Additionally, through the use of memory recall to improve local model robustness and paired learning and optimization scheme that synchronizes efforts between the digital twin and the physical access point, our scheme enables optimization across a wider range of scenarios by replaying underrepresented experiences, ensuring more comprehensive and robust operational strategies. Simulation results demonstrate that, even with an increase in the offered load, the DtPLO scheme can successfully manage network access, maintain packet delivery ratio, and stabilise the device queue.