单神经元自适应PID控制器在AQM中的应用

The Application of Single Neuron Adaptive PID in AQM

主动队列管理(AQM)对于保持TCP网络高性能是一种有效的策略。AQM的基本机制是调节数据包进入路由器缓冲区的速率,从而避免网络拥塞。针对传统PID控制器在拥塞控制过程中参数固定的缺点,将神经网络理论引入AQM的研究中,设计了一种改进的单神经元自适应PID主动队列管理器。基于NS2平台的仿真结果表明,与PID控制器及传统的单神经元自适应PID控制器相比,改进的控制器对网络环境的变化有更强的适应能力和更好的稳定性。

Active Queue Management（AQM） is an efficient strategy to maintain high utilization of Transmission Control Protocol（TCP） network resources.The fundamental mechanism is to manage incoming packet rates at a router to prevent incipient network congestion.The parameters of the traditional PID controller are fixed in whole process,so it will have poor performance when the process changes.This paper presented an improved single neuron adaptive PID controller.The simulation results based on NS2 demonstrated that the improved adaptive neural AQM outperforms PID controller and traditional single neuron PID adaptive controller.