一种用于拥塞网络节点缓存队列长度控制方法

发布时间：2018-06-08 11:36

本文选题：拥塞网络 + 节点缓存　；参考：《计算机仿真》2016年08期

【摘要】：控制拥塞网络节点的缓存队列长度,在提高网络利用率方面具有重要意义。由于拥塞网络中传送的节点数据太多,使得节点缓存队列长度受到限制。采用传统的控制方法,当存储节点受限时对节点缓存队列进行剪裁来增加传输速率,难以有效的对节点缓存队列进行控制。现提出一种量子粒子群算法,并应用到拥塞网络节点缓存队列长度的控制中。利用概率接纳算法对网络的拥塞状态进行检测,以检测结果为依据对消息进行接受与丢弃处理,采用PI控制器建立PI主动队列管理模型,获取当前时刻拥塞网络节点缓存队列数据包丢弃率,引入量子粒子群算法通过对模型比例系数和积分系数的优化,实现对拥塞网络节点缓存队列长度控制。仿真结果表明,采用改进的控制方法可以有效的降低节点丢包率,提高网络吞吐率,降低控制误差。
[Abstract]:Controlling the buffer queue length of congested network nodes is of great significance in improving network utilization. Because too much node data is transmitted in congested network, the queue length of node cache is limited. Using the traditional control method, the node cache queue is clipped to increase the transmission rate when the storage node is limited, so it is difficult to effectively control the node cache queue. A quantum particle swarm optimization (QPSO) algorithm is proposed and applied to the queue length control of congestion network nodes. Probabilistic admission algorithm is used to detect the congestion state of the network. Based on the detection results, the message is received and discarded. The Pi controller is used to establish the Pi active queue management model. The packet drop rate of the congestion network node cache queue is obtained at the current time. The quantum particle swarm optimization algorithm is introduced to control the queue length of the congestion network node by optimizing the model scale coefficient and integral coefficient. The simulation results show that the improved control method can effectively reduce the node packet loss rate, improve the network throughput, and reduce the control error.
【作者单位】：南阳理工学院软件学院;
【分类号】：TP393.06

【相似文献】