Estimation of CAN Bus reliability based on real-time performance

With the rapid development of CAN Bus, many scholars and engineers focus on how to evaluate the reliability of CAN Bus. It is widely accepted that the calculation of the probability of successful transmission is an effective method to analyze both the real-time performance and message correctness. However it was proved that the probability of successful transmission is not applicable to calculating the reliability of timely delivery because of the statistical correlation. And hence artificial neural network was induced to simulation to calculate the reliability. The method was based on the analysis of international and national standard and protocol of CAN Bus. The reliability was approximated by the artificial neural network and the simulation was adopted to both calculation of the reliability in short period in harsh electromagnetic interference environment and validation of artificial neural network. The method reduced consumption of both data and time when the simulation was only utilized for the calculation of reliability over long time.