摘要
从控制理论的角度研究了一种基于速率的拥塞控制方法.采用单神经元自适应P ID控制器,可以在线调节控制参数以保证控制环路的稳定性.运用Jury稳定判定标准给出了保证被控ATM网络在平衡点附近稳定的条件.仿真结果表明,使用该控制方案能使缓冲器队列长度快速收敛到目标值,并维持小的队列振荡,与传统的P ID控制器相比具有更好的稳定性和鲁棒性.
A rate-based congestion control method is proposed using control theory. This method uses a single neural adaptive PID controller, whose parameters can be adjusted online to ensure the stability of the controlled loop. Using the Jury stability standard gives the conditions for stability of the controlled ATM system around its desired operating point. Simulation results demonstrate that this control scheme can lead to quick convergence of the queue length to the desired value and maintain small oscillation. This scheme has better robustness to disturbance than the traditional PID controller.
出处
《控制与决策》
EI
CSCD
北大核心
2005年第11期1225-1228,1234,共5页
Control and Decision
基金
国家自然科学基金项目(60274099)
高等学校博士学科点专项科研基金项目(20020145007)
