The Research Of PID-type ILC Based On Network Congestion Control Algorithm And Its Simulation

With the increasingly network speed and scale,network congestion control becomes more important and it therefore has been a hotspot of research.In this dissertation,iterative learning control theories were introduced into the network congestion control,proposed an open-close loop PID type iterative learning algorithm for network congestion control and proved its convergence.Moreover,parameters setting of the PID type controller and parameters optimization of the PID type ILC network congestion control were analyzed and discussed in detail,the network reference model in rapid network and its congestion control system structure was analyzed,in the mean time,designed the congestion control system architecture by making use of the above results.The main contents are described as follows:①Researched the network reference model and its congestion control system structure,made an abstract of the network congestion with time-varying nonlinear discrete-time delay equation.②Based on the controlled system equation,proposed an open-close loop PID type ILC network congestion control algorithm and proved its mathematical convergence.③Improved the PID type ILC network congestion control algorithm on parameter setting instead of the conventional experience based setting methods.The PID type ILC algorithm and STF algorithm were adopted to set parameter in the way of formula,and the formula for parameter optimization was also described by mathematical proof.④Realized the physical model construction of the PID type ILC network c ongestion control system,the model was applied to the congestion controller,w hich was designed according to the algorithm proposed in this dissertation,and produced a better result by comparing with some former congestion controllers.