摘要
针对网络拥塞控制系统在大时滞网络中产生的不利影响,提出一种基于速率和队长的大时滞网络AQM算法。该算法采用缓冲区队列长度和包到达速率作为网络拥塞的判别依据,在结合Smith预估的模糊PID控制方法中加入速率控制项。仿真表明该算法在大时滞和网络动态变化的环境中拥塞响应较快、收敛时间短,并能较好地将队列长度稳定到期望值附近,提高缓冲区的利用率。
To solve the problem of negative impact on the performance of Active Queue Management(AQM) in large delay network,an AQM algorithm is proposed based on rate and queue length in large delay network.This algorithm consideres the effect of queue length and packet arriving rate together to indicate congestion degree.An input rate control is used to speed up the responsive time in fuzzy Smith control method.Simulation result shows that the algorithm has a fast response and short convergence time in large delay network and dynamic network environment.It also can control the size of the queue in the buffer to an expected length and achieves high utilization.
出处
《计算机工程》
CAS
CSCD
北大核心
2010年第20期96-98,共3页
Computer Engineering
基金
陕西省自然科学基金资助项目(SJ08F14)