A Novel Information Update Policy for Real-Time Wireless Feedback Control in IIoT

By driving automation devices to achieve allocated tasks, real-time wireless feedback control systems can be treated as a key enabler for task-orientated applications in Industrial Internet of Things (IIoT). In such a system, timely information from wireless communication aspect is one of the most important components affecting control performance, e.g., control stability and tracking error, where the effect is determined by information update policy. In this article, we propose a novel information update scheduling policy for the aforementioned systems. Specifically, based on the existing research on the Age of Information (AoI) evaluating the timeliness of the information, we analyze the relationship between AoI and control performance, where we find that minimizing AoI is not always equivalent to maximizing control performance. Then, we adopt two metrics, i.e., Age of Stale Information (AoSI) and the Value of Information (VoI), to evaluate the timeliness of information update for stochastically stable control and VoI is for completely stable control, respectively. More importantly, we analyze the relationship between AoSI and VoI, where minimizing AoSI is a subset of maximizing VoI. Then, VoI is adopted as the unique metric for both the aforementioned systems. Finally, we propose an
$\alpha $
-wait policy for maximizing VoI. In such a policy, an optimal waiting time interval
$\alpha $
can be obtained to maximize control performance and meanwhile reduce communication resource consumption. The simulation results show that the proposed method can reduce more than 68% channel occupied probability from communication perspective and meanwhile reduce almost 20% mean-square-error (MSE) from control perspective compared with conventional methods.