Feedback based network scheduling of networked control systems

Due to the bandwidth limitation as well as workload variations in today's networked control systems (NCSs), available network resources for control applications are prone to be scarce. With traditional static schedule, the quality of control (QoC) could be significantly degraded. To attack the problem of network scheduling in a NCS with several control loops, a network scheduler that can be easily implemented in existing networked control applications is presented. The objective is to improve the overall system performance via dynamic resource allocation. Following the newly emerging methodology of feedback scheduling, the control requirements are integrated in the scheduling of the shared network. As a novel approach to allocating network resources, the proposed network scheduler dynamically adjusts control loops' priorities according to their current urgency values. Preliminary simulation results highlight the advantages of the proposed approach.