Generators for synthesis of QoS adaptation in distributed real-time embedded systems

This paper presents a model-driven approach for generating Quality-of-Service (QoS) adaptation in Distributed Real-Time Embedded (DRE) systems. The approach involves the creation of high-level graphical models representing the QoS adaptation policies. The models are constructed using a domain-specific modeling language - the Adaptive Quality Modeling Language. Multiple generators have been developed using the Model-Integrated Computing framework to create low-level artifacts for simulation and implementation of the adaptation policies that are captured in the models. A simulation generator tool synthesizes artifacts for Matlab Simulink/Stateflow(R) (a popular commercial tool), providing the ability to simulate and analyze the QoS adaptation policy. An implementation generator creates artifacts for Quality Objects, a QoS adaptation software infrastructure developed at BBN, for execution of QoS adaptation in DRE systems. A case study in applying this approach to an Unmanned Aerial Vehicle - Video Streaming application is presented. This approach has goals that are similar to those specified in the OMG's Model-Driven Architecture initiative.