互联网中的不确定时滞全局滑模AQM算法

Global sliding mode AQM algorithm for Internet with uncertainty and time delay

为解决互联网中的拥塞现象,基于全局滑模理论,提出了一种主动队列管理算法。将互联网中的不确定因素等效为拥塞控制模型的参数摄动,并考虑到时变时滞因素,利用线性矩阵不等式设计了渐近稳定的全局滑模面,使系统运动的全程都是滑动模态,显著提高了系统的鲁棒性。并且所设计的控制器能够有效降低路由器中队列长度的稳态振荡。仿真对比表明,所提出的算法具有更高的稳定性和鲁棒性,能够适应复杂多变的互联网环境。

In order to solve congestion phenomenon in Internet, an active queue management （AQM） algorithm is presented based on global sliding mode control theory. The uncertain factors in Internet are equivalent to parameter perturbation of congestion control model, and time-varying delay factor is also considered in model. An asymptotically stable global sliding surface is designed by linear matrix inequality （LMI）. Because the whole system movement is sliding mode, the robustness of system is obviously improved. Moreover, the controller can obviously reduce the steady oscillation of the queue length in router. The simulation contrasts demonstrate that the p stability and robustness, and it can adapt to the complex and changing control algorithm possesses better Internet environment.