PID type neural network control for active queue management

To study the congestion control of intermediate nodes in the changing network, PID type neural network control scheme for AQM (active queue management) is proposed. Back-propagation algorithm is used to adjust the weights of neural networks and stability of closed-loop system is proved according to the Lyapunov theory. Based on NS-2 simulation platform, the result showed that the proposed control scheme can adapt to the changing network situation, the system steady-state error and transient performance of the proposed scheme is superior to those of PID.