Traffic prediction of can network system with dual communication channels

The increasing number of electronic control units, sensors, and actuators in intelligent vehicles, and the increasing need for more intelligent functions require a network with increased capacity and real-time capability. As an example of enhancing the capacity of a controller area network (CAN) system, this paper presents a CAN system with dual communication channels as well as a traffic prediction method that predicts the traffic of each channel to allocates frames to the more appropriate channel. An experimental testbed using off-the-shelf microcontrollers with two CAN controllers was used to prove the feasibility of the traffic prediction method.