【摘要】 针对网络模型的不确定性和参数的时变性,该文提出了一种基于输入速率和队列长度变化的模糊控制主动队列管理算法。采用模糊控制方法,不依赖于网络的精确数学模型;考虑了负载因素和队列因素,加快了对拥塞的感知速度。仿真结果表明该算法能迅速地将队列长度收敛到目标队列长度附近,并且其丢弃概率小于随机早期检测算法以及PI控制算法。更多还原

【Abstract】 Considering the uncertainty of network model and time variability of network parameters,a fuzzy control active queue management algorithm based on input rate and queue variance is proposed.Using fuzzy control method,an exact model is not needed.Both load and queue of the network are concerned to speed up the perception of the congestion.Simulation results show that this algorithm can stabilize the queue length to the target value and its drop probability is less than random early detection algorithm and PI controller algorithm.更多还原