Research On Traffic Control Of IP Network

Effective traffic control is not only the basis of steady efficient networks but the premise of QoS service models and technology. Queueing theory is applied in this thesis to study traffic control in IP network.The method which the current research on traffic control of IP network uses has very good flexibility, but it cannot often simulate and produce the definite analytic solution of the traffic control. This paper uses the simulated method to form the models based on the flow control principle in IP network, analyzes the models with queueing theory and describes the characteristic of the correlation flow control through analytic formulas.This thesis studies traffic control in IP network from the two aspects: Queueing management and congestion control disciplines. Firstly, FIFO queue is studied in queue management discipline and mathematical expression of buffer capacity is given, then tandem queueing model is studied and performance analysis is given; secondly, priority queueing policy is analyzed and performance analysis is given, preemptive priority queuing model with traffic control is discussed based on this and performance analysis is given at the same time. Mathematical models of Selective discard,Feedback preemptive priority and bucket is formed in congestion control disciplines. Queuing theory is applied in analyzing the above models and results of theoretical meaning are given.The application part of this thesis studies voice mixture of video-conference. In this part, mixing module is introduced and queueing model is formed. Steady-state distribution is worked out and some performance evaluations are given.The results of this paper can be applied to design and optimize IP network, its performance can be greatly improved by this.