An active queue management algorithm based on adaptive variable structure control using neuron learning

Considering the non-linearity of TCP model, uncertainty of RTT and fluctuation of network load, an active queue management scheme based on variable structure control by using single neuron adaptive learning is proposed. The nonlinear variable structure controller (VSC) is used to guarantee the swiftness and robustness of queue response at router. However, the jitter of VSC will cause the queue fluctuation and performance degradation.Therefore a single neuron is introduced to adjust the parameters of the variable structure controller in order to alleviate the effect of jitter and modelling uncertainty. The scheme can reduce the jitter problem and enhance the robustness for AQM control system greatly. Finally, simulation results show the effectiveness of the proposed algorithm through NS-2 simulator.