Deep Reinforcement Learning based Deterministic Routing and Scheduling for Mixed-Criticality Flows

Deterministic networking (DetNet) has recently drawn much attention by investigating deterministic flow scheduling. Combined with artificial intelligent (AI) technologies, it can be leveraged as a promising network technology for facilitating automated configuration in the Industrial Internet of Things (IIoT). However, stricter requirements IIoT have posed significant challenges, that is, and bounded latency time-critical applications. This paper incorporates deep reinforcement learning (DRL) Cycle Specified Queuing Forwarding (CSQF) proposes DRL-based Flow Scheduler (Deep-DFS) to solve Routing Scheduling (DFRS) problem. Novel delay aware representations, action masking criticality reward function design are proposed make Deep-DFS more scalable efficient. Simulation experiments conducted evaluate performances Deep-DFS, results show schedule flows than other benchmark methods (heuristic-based AI-based methods).