Quality of Service and Quality of Control in real-time control systems

In this paper, we consider several control loops sharing a common computation or communication resource. We assume the presence of a scheduler that approximates a fluid flow temporal partitioning of the resource among the different tasks. The goal of this paper is to expose the relation between the amount of resources allocated to each loop (Quality of Service) and the Quality of Control resulting from this choice. The latter is quantified by the steady state variance of the system state. As a first step toward a more general theory, results are presented for scalar systems controlled using a time-triggered approach.