摘要
针对CAN总线调度问题,分析了CAN网络中信息帧发生传输错误的概率化最大响应时间,融合DMS算法的可预测性良好、处理器开销低以及对数分区EDF算法的灵活性强、网络利用率高的特点,提出一种考虑消息关键性的混合调度算法;在上述研究基础上,利用MATLAB/SimEvents工具箱建立CAN网络模型,并对3种调度算法进行了仿真,当网络利用率达到约85%时,混合调度算法丢包率为0.8%,但丢失的数据不包含硬实时消息,而且成功降低了处理器的额外开销时间,仿真结果表明混合调度算法既能提高系统的实时性,又能减小处理器的运算负担。
In this paper we provide a probabilistic analysis of worst case response time of message with faults in the Controller Aera net- work (CAN). By studying and analyzing static scheduling algorithm Deadline Monotonous Scheduling (DMS) and dynamic scheduling algo- rithm Earliest Deadline First (EDF) in the CAN, and combining the good predictability and cheap processor cost of DMS algorithm with the strong flexibility and high utilization of logarithm partition EDF algorithm, this paper presented a hybrid scheduling algorithm with consider- ing critical level of messages. Based on the work mentioned above, the simulation platform of CANbus is developed with MATLAB/Si- mEvents tool box, The platform was used to real--time analysis for the scheduling algorithms mentioned before, the simulation results showed that the hybrid scheduling algorithm could not only improve the real--time performance of system, but also could reduce the cost of processor.
出处
《计算机测量与控制》
北大核心
2014年第5期1502-1505,共4页
Computer Measurement &Control
基金
国家自然科学基金(51174084)
武汉市科技青年晨光计划(201271031386)