预测PI时滞网络拥塞控制算法设计及性能分析

Design and analysis of predictive PI algorithm for congestion control in time-delay network

针对网络中存在的大时滞给主动队列管理算法性能带来的不利影响,将Sm ith预估器与Dahlin算法相结合,提出了一种预测PI拥塞控制算法,首先利用Sm ith预估器补偿时延滞后,克服了大时滞给系统性能带来的影响;然后按Dahlin算法设计控制器,把控制器参数和预估对象模型参数相结合,既减少了整定参数,也避免了参数整定时的相互影响.同时,利用经典控制理论方法分析了系统稳定性和存在链路容量干扰时瓶颈队列的暂态、稳态特性.仿真结果显示预测PI算法控制性能优于RED,PI算法及具有较强的鲁棒性.

A predictive proportional integral （PPI） congestion control algorithm is proposed which combined Smith predictor and Dahlin algorithm to alleviate the time-delay influence on the performance of active queue management （AQM） in network. The Smith predictor is utilized to compensate the time-delay, while Dahlin principle is employed to design controller to reduce the number and interaction of tuning parameters. A classical control method is also introduced to analyze the stability of the system and the transients as well as steady-state behaviors of bottleneck queue with link capacity disturbances. Finally, simulation results are given to show the PPI algorithm＇s advantages through experiments with disturbances, and comparison between random early detection （RED） and proportional integral （PI） algorithms are also made.