论文部分内容阅读
为了在航空电子全双工交换式以太网(AFDX)中更加灵活地对虚拟链路(VL)实行接纳控制,将OpenFlow引入AFDX网络中,建立了相应的网络模型、消息模型和流量模型。利用网络演算方法分析了优先级VL在基于OpenFlow的AFDX网络中的端到端延迟上界,结合粒子群优化算法,提出了动态优先级接纳控制算法。理论分析结果表明:与动态非优先级接纳控制方法和传统静态优先级分配方法相比,本算法的消息延迟分别降低了49.2%和26.4%,并且本算法能够更加灵活地对VL实行接纳控制,提高网络资源的利用率。最后通过仿真对理论分析的结果进行了验证。本算法为提高接纳控制方法的性能提供了参考,增强了AFDX网络的通信效率与实时性能。
In order to more flexibly control the virtual link (VL) in Avionics full-duplex switched Ethernet (AFDX), OpenFlow is introduced into the AFDX network, and the corresponding network model, message model and traffic model are established. The end-to-end delay upper bound of priority VL in OpenDlow based AFDX network is analyzed by using network calculus method. Combined with particle swarm optimization algorithm, a dynamic priority admission control algorithm is proposed. The theoretical analysis shows that compared with the dynamic non-priority admission control method and the traditional static priority assignment method, the message delay of this algorithm is reduced by 49.2% and 26.4% respectively, and the proposed algorithm is more flexible in admission control, Improve the utilization of network resources. Finally, the simulation results verify the theoretical analysis. The algorithm provides a reference for improving the performance of the admission control method and enhances the communication efficiency and real-time performance of the AFDX network.