| 多策略选路下端到端低延迟应用层ANFIS控制器设计 |
 点此下载全文 |
| 引用本文：蔡文斌1，张焕域1，林君颖2，李博1，陈志龙1，洪杰1.多策略选路下端到端低延迟应用层ANFIS控制器设计[J].计算技术与自动化,2026,(2):54-60 |
| 摘要点击次数: 51 |
| 全文下载次数: 18 |
|
|
| 中文摘要:电力通信网络分段路由流量工程（SR-TE）实施过程中，应用常规模糊控制算法设计应用层控制器，容易因传输路径冲突发生拥堵，导致控制器应用后网络端到端时延较大。因此，提出了多策略选路下端到端低延迟应用层ANFIS控制器设计方法。从路径检测、流量调节、时延控制方面入手，建立应用层控制器结构。检测流间冲突度和冲突预留时延，构造全局代价函数，多策略选路下规划出符合要求的端到端传输路径。以模糊控制PID算法为核心，结合神经元网络结构，设计用于流量动态调节的自适应控制算法。以最小时延为目标校正当前控制参数，完成端到端低时延应用层控制器设计。实验结果表明：将该控制器应用在网络拥堵状态下，表现出的端到端时延依旧为163ms左右，与拥堵发生前相比并未发生变化，证明了控制器的优越应用性能。 |
| 中文关键词:SR-TE 模糊控制 流量分析 低时延 神经元自适应 控制器 |
| |
| Design of Application Layer ANFIS Controller for End-to-end Low Latency Under Multi Strategy Routing |
|
|
| Abstract:During the implementation of the segmented routing traffic engineering (SR-TE) for power communication networks, the use of conventional fuzzy control algorithms to design application layer controllers can easily lead to congestion due to transmission path conflicts, resulting in significant end-to-end network latency after the controller is applied. Therefore, a design method for end-to-end low latency ANFIS controller with multi strategy routing is proposed. Starting from path detection, traffic regulation, and delay control, establish an application layer controller structure. Detect the degree of inter stream conflict and conflict reservation delay, construct a global cost function, and plan an end-to-end transmission path that meets the requirements under multi strategy routing. Design an adaptive control algorithm for dynamic flow regulation based on fuzzy control PID algorithm and neural network structure. Calibrate the current control parameters with the goal of minimizing latency, and complete the end-to-end low latency application layer controller design. The experimental results show that when the controller is applied in network congestion, the end-to-end delay still remains around 163ms, which has not changed compared to before the congestion occurred, proving the superior application performance of the controller. |
| keywords:SR-TE fuzzy control traffic analysis low latency neuron adaptation controller |
| 查看全文 查看/发表评论 下载pdf阅读器 |
位访问者,今日访问