Design and Performance of DDS-based Middleware for Real- Time Control Systems

Abstract: Data-centric design is emerging as a key tenet for building advanced data-critical distributed real-time and embedded systems. These systems must find the right data, know where to send it, and deliver it to the right place at the right time. Data Distribution Service (DDS) specifies an API designed for enabling real-time data distribution and is well suited for such complex distributed systems and QoS-enabled applications. It is also, widely known that Control Area Networks (CAN) are used in real-time, distributed and parallel processing.

Thus, the goal idea of this paper is to study an implementation of publish-subscribe messaging middleware that supports the DDS specifications and that is customized for real-time networking. This implementation introduces an efficient approach of data temporal consistency and real-time network-scheduler that schedules network traffic based upon DDS QoS-policies. A simulator has been developed to demonstrate that our implementation fulfills the guarantees predicted by the theoretical results.