Real-time communication for smart sensor networks: A CAN based solution

Abstract

Due to the increased availability of low cost network technology, the use of networks to interconnect sensors, actuators and controllers is now widely accepted. Such increased availability is one of the driving factors for the implementation of smart sensor networks. To ensure the correctness of the supported applications, the communication network must provide a reliable and timely communication service. Aside from the Medium Access Control (MAC) protocol, one of the components that has a high impact in the communication delays is the local communication stack. Therefore, the usage of an adequate communication stack is of utmost importance to guarantee the timing correctness of the supported smart sensor applications. In this paper, we analyze the real-time aspects of a CAN-based smart sensor network. We assess the use of well established real-time scheduling algorithms to manage the outgoing queue of each of the local communication stacks. We show that it is possible to improve the responsiveness of applications supported by the CAN communication protocol, by using just a light scheduling middleware to adequately schedule each of the outgoing queues. We also show that implementing such middleware upon COTS communication hardware, it is possible to reduce the occurrence of priority inversions in the communication medium. Therefore it becomes possible to decrease the number of deadline misses even for highly loaded network scenarios. As a consequence, we advocate that CAN networks can be an interesting solution to support event-triggered smart sensor networks.