Real-Time Elastic Network Scheduling of Networked Control Systems

Limited network bandwidth and dynamic changes in workload have serious impact on the performances of networked control systems. For a multiple loops networked control system in a dynamic environment, this paper proposes a real-time network scheduling method based on the idea of feedback scheduling. System packet loss rate can be reduced by dynamically adjusting the sampling period of tasks, thus achieving higher utilization of network resources. The implementation mechanisms for the model structure and the algorithm are analyzed in detail, and the algorithm is verified by simulated experiment. Results show that this algorithm can improve resource utilization and guarantee good stability at the same time.