Challenges in the Implementation, the Configuration and the Evaluation of a QoS-enabled Middleware for Real-Time Embedded Systems

Distributed computing in complex embedded systems gain complexity, when these systems are equipped with many microcontrollers which oversee diverse Electronic Control Units (ECU) connecting hundreds or thousands of analogue and digital sensors and actuators.

The Publish/Subscribe paradigm matches well with these systems. Data Distribution Service (DDS) is a publish/subscribe data-centric middleware. It specifies an API designed for enabling real-time data distribution and is well suited for such complex distributed systems and QoS-enabled applications.

Unfortunately, the need to transmit a large number of sensor measurements over a network negatively affects the timing parameters of the control loops. The CAN-bus enables the information from a large number of sensor measurements to be conveyed within a few messages. Its priority-based medium access control is used to select the sensor messages with high timing constraints. This approach greatly reduces the time for obtaining a snapshot of the environment state and therefore supports the real-time requirements of feedback control loops.

The main objective of this paper is to demonstrate how DDS API is implemented on a CAN-Bus and gives performance evaluation related to delivery and transport QoS parameters.