欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对网络拥塞问题提出了一种非线性主动队列管理（AQM）算法,设计了一种标注概率可以保证网络的最大传输速率和最小队列长度。通过比较引理和反步设计法设计了非线性输出反馈控制算法,并提出了控制参数的取值范围。在考虑实际网络中往返时间（RTT）与队列长度的函数关系的基础上提出了基于变RTT的非线性网络模型,并设计了新的TCP窗口大小观测器,证明了当标注概率取值范围为0~1时,TCP窗口观测误差渐近地收敛到0。仿真表明,该文基于反步设计法的非线性AQM算法在变化的网络环境下具有较好的鲁棒稳定性,实验中的链路利用率为95.41%,丢包率为0.287%。
英文摘要：
A nonlinear active queue management（AQM） algorithm is presented for network congestion.The packet drop ratio is designed for ensuring both the most transmission rate and the least queue length.The nonlinear output feedback controll algorithm is obtained based on comparison lemma and backstepping technique,and the value range of parameters is also proposed.Considering the functional relationship between round trip time（RTT） and queue length,a nonlinear network model based on variable RTT is designed.A novel TCP window size observer is proposed.The observer error converges to zero asymptotically when the packet drop ratio falls between 0 and 1.Simulation results indicate the proposed nonlinear active queue management-backstepping technique has high robust stabilization under diverse network conditions,the link utilization is 95.41%and the packet loss ratio is 0.287%.
同期刊论文项目
同项目期刊论文
ROBUST STABILIZATION OF STATE DELAYED T-S FUZZY SYSTEMS WITH INPUT SATURATION VIA DYNAMIC ANTI-WINDU
Economic and economic-statistical designs of an (X)over-bar control chart for two-unit series system
Optimizing production and inventory decisions in a supply chain with lot size, production rate and l
A Bayesian Meta-Modeling Approach for Gaussian Stochastic Process Models Using a Non Informative Pri
期刊信息
位置：