Mixed Traffic Scheduling With Latency and Jitter Analysis in URLLC Industrial Automation

Ultra-reliable and low-latency communications (URLLC) has great potential in wireless industrial automation, which increases flexibility without moderating the stringent quality-of-service (QoS). In URLLC-enabled Industrial Internet of Things (IIoT), mixed traffic is essential to support applications with differentiated latency requirements, where event-triggered (ET) traffic requires bounded end-to-end (E2E) latency, and time-triggered (TT) traffic demands deterministic latency. This article proposes a mixed traffic scheduling scheme with the preemption mode, in which TT traffic with a higher priority is allowed to preempt the transmission of ET traffic to ensure TT deterministic latency. However, this preemption may cause ET latency to exceed its requirements with constrained resources. To characterize the impact of TT preemption on the ET latency and the deterministic metrics of TT traffic, we derive the ET delay violation probability bound and the TT deterministic delay probability using the stochastic network calculus. To further achieve bandwidth-saving scheduling while ensuring the QoS requirements of mixed traffic, the credit-based shaper (CBS) configurations, radio resources and delay components are assigned under constraints of ultra-reliability, ET bounded latency and TT deterministic latency. Then, a three-step method is proposed to find the global optimal solution to minimize the total bandwidth. Simulation results validate our analysis and show the bandwidth-saving performance gain by jointly optimizing uplink and downlink resources with the proposed preemption-based scheduling scheme.