ReCANcentrate: a replicated star topology for CAN networks

Abstract

Controller area network (CAN) is nowadays widespread in distributed embedded systems due to its electrical robustness, low price, and deterministic access delay. However, its use in safety-critical applications has been controversial due to dependability limitations. In particular, in a CAN bus there are multiple components such that a single fault of any of them can prevent the communication capabilities of several nodes and may provoke a general failure of the communication system, i.e. there are multiple severe points of failure. In Barranco, M et al., (2004) we proposed a new active star topology, called CANcentrate, that solves these limitations by means of an active hub with enhanced fault-treatment capabilities. However, the center of the star still represents a severe point of failure, thus not being suitable for more demanding safety-critical systems. In this paper, we propose a replicated star topology, called ReCANcentrate, which has no severe points of failure and is fully compatible with existing CAN controllers. The paper analyzes related work, describes the CANcentrate basics, explains the design and functionalities of ReCANcancentrate, and finally describes the implementation and test of its prototype