Traffic Scheduling Method for Power Low-Latency Business in TSN Based on Sending Time Scheduled

Time-sensitive networks (TSN) can provide stable end-to-end latency and low jitter for power low-latency control business. Time-aware shaper (TAS) is a more commonly used traffic shaping mechanism in TSN, which controls the sending and receiving of traffic by using gated lists (GCL). However, TAS has a complex gating planning problem, so a cyclic queuing and forwarding mechanism (CQF) has been proposed to simplify the design of gated lists. However, there is still a lack of corresponding resource scheduling algorithms to reasonably allocate resources to multiple flows. This paper finds that the network resource utilization can be greatly improved by adjusting the sending time of traffic. Therefore, this paper considers the resource scheduling problem as an optimization problem under certain constraints, and on this basis, this paper proposes a greedy sorting algorithm (GSA) to schedule the traffic in the CQF model. The algorithm sorts the flow and adjusts the injection time of the flow by comprehensively considering the flow characteristics to increase resource utilization; meanwhile, this paper also proposes selecting the optimal offset time slot based on the flow density to further improve the success rate of network scheduling. Simulation results show that the algorithm can increase the number of successfully scheduled flows by about 19.9% compared to direct scheduling.