摘要
随着无线网络终端数量不断增加,导致网络拥塞问题日益突出。为了提高无线网络的服务质量和资源利用率,解决建立精确的数学模型难,以及传统PID拥塞控制算法的参数整定难,自适应差,适应动态网络环境难等问题,提出一种融合模糊控制和PID型神经网络技术的主动队列管理算法(FCAPIDNN)。通过神经网络在线学习实时调节PID网络参数,借助模糊控制计算神经网络的学习速率,从而提高队列长度的调节效果。NS2的仿真结果表明,相较传统PID控制,FCAPIDNN算法能使队列长度更快地收敛到预期值附近,即使对于突发数据流也有较强的适应性。
With the increasing number of wireless network terminals,the problem of network congestion is becoming more and more prominent.In order to improve the quality of service and resource utilization of wireless networks,and to solve the difficulty of establishing accurate mathematical model and the problems of traditional PID control algorithms,such as difficult parameter tuning,the poor adaptive ability and poor adaptability to dynamic network environment,an active queue management algorithm is proposed based on fuzzy control and PID neural network(FCAPIDNN).It regulated the PID network parameters in real time through the online learning of the neural network.Moreover,it used fuzzy control to calculate the learning rate of the neural network,which improved the control effect of the queue length.The simulation results based on NS2 show that compared with the traditional PID control algorithm,FCAPIDNN algorithm can make the queue length converge to the expected value faster,even for the burst data stream also has strong adaptability.
作者
徐磊
杨晓飞
王玉龙
XU Lei;YANG Xiao-fei;WANG Yu-long(School of Electronics and Information,Jiangsu University of Science and Technology,Zhenjiang Jiangsu 212003,China)
出处
《计算机仿真》
北大核心
2019年第3期285-288,421,共5页
Computer Simulation
基金
国家自然科学基金(61374063
61403170)
江苏省自然科学基金(BK20161361)
2015年度江苏省第四期"333工程"科研资助项目(RA2015358)
江苏省"六大人才高峰"第十二批高层次人才选拔培养资助项目(DZXX-025)
作者简介
徐磊(1993-),男(汉族),江苏南京人,硕士研究生,主要研究领域为网络控制;杨晓飞(1983-),男(汉族),河南襄城人,博士,副教授,主要研究领域无线传感器网络、雷达技术应用,网络控制;王玉龙(1977-),男(汉族),山东人,教授,硕士研究生导师,主要研究领域为网络控制、故障检测、船舶运动控制等。
