基于显式拥塞指示的自适应主动队列管理算法

Adaptive AQM Algorithm Based on Explicit Congestion Notification

目前Internet网络中间节点拥塞控制问题在网络和控制理论界已获得了广泛关注。本文提出一种基于神经元自适应PID控制器的AQM算法,针对TCP/AQM系统模型,结合中间节点队列管理和显式拥塞指示机制(Explicit Congestion Notification,ECN)机制,采用梯度学习算法来在线调整基于神经元PID的AQM控制器参数,以实现标记/丢包概率的自适应调整,从而对网络拥塞程度作出及时响应,尤其在网络参数时变的情况下仍能保证良好的动态性能,并显著改善网络的服务性能(QoS)。最后通过NS-2仿真结果表明,该算法在队列稳定性、平均丢包率等性能方面要明显优于基于常规PID的AQM算法。

Nowadays congestion control problem of the intermediate nodes in the Internet has received extensively attention in networking and control community.A novel adaptive PID（Proportional-Integral-Differential） controller based on single neuron for the problem of AQM was proposed.Considering a previously developed nonlinear dynamic model of TCP/AQM system and the queue management and explicit congestion notification（ECN） mechanism of intermediate nodes,the parameters of AQM controller were tuned online by using gradient-descent algorithm,and the probability of packet dropout was obtained adaptively to measure the degree of congestion in time,so that the quality of service（QoS） of network and the transient performance could be improved greatly especially when the network parameters are time-varying.Finally,the proposed algorithm was verified by using NS-2 simulator.The simulation results show that the integrated performance of this proposed controller is obviously superior to those of common PID controller especially on the queue stability and mean loss ratio.