Age of information (AoI) and reliability are two critical metrics to support real-time applications in Industrial Internet of Things (IIoT). These metrics reflect different concepts of timely delivery of sensor information. Monitoring traffic serves to maintain fresh status updates, expressed in a low AoI, which is important for proper control and actuation actions. On the other hand, safety-critical information, e.g., emergency alarms, is generated sporadically and must be delivered with high reliability within a predefined deadline. In this work, we investigate the AoI-reliability trade-off in a real-time monitoring scenario that supports two traffic flows, namely AoI-oriented traffic and deadline-oriented traffic. Both traffic flows are transmitted to a central controller over an unreliable shared channel. We derive expressions of the average AoI for the AoI-oriented traffic and reliability, represented by Packet Loss Probability (PLP), for the deadline-oriented traffic using Discrete-Time Markov Chain (DTMC). We also conduct discrete-event simulations in MATLAB to validate the analytical results and evaluate the interaction between the two types of traffic flows. The results clearly demonstrate the tradeoff between the AoI and PLP in such heterogeneous IIoT networks and give insights on how to configure the network to achieve a target pair of AoI and PLP.

翻译：信息年龄（AoI）与可靠性是支撑工业物联网（IIoT）实时应用的两项关键指标。这两项指标反映了传感器信息时效性传输的不同概念。监控流量旨在维持低AoI所表征的状态更新新鲜度，这对正确的控制与执行操作至关重要。另一方面，安全关键信息（如紧急告警）呈偶发性生成，必须在预设截止时间内以高可靠性完成传输。本研究探讨了支持两种流量类型（即面向AoI的流量与面向截止时间的流量）的实时监控场景中AoI与可靠性之间的权衡关系。两种流量均通过不可靠共享信道传输至中央控制器。我们利用离散时间马尔可夫链（DTMC）推导了面向AoI流量的平均AoI表达式，以及用丢包概率（PLP）表征的面向截止时间流量的可靠性表达式。通过MATLAB离散事件仿真验证了解析结果，并评估了两种流量之间的相互作用。结果清晰揭示了此类异构IIoT网络中AoI与PLP的权衡关系，并为如何配置网络以实现目标AoI与PLP组合提供了指导。