Research On The Quality Of Service Technology In IP Network

The rapid development of Internet resulted in the IP services growth rapidly and diversification,At the same time as the number of users increase dramaticly , network traffic exploding, Network congestion problems become more serious, so the quality of service (QoS) has become one of the world's research focus. In order to achieve the relative performance of QoS, congestion control plays an important role in the solving Network congestion problems,while The active queue management (Active Queue Management, AQM) mechanism is one of important means to achieve congestion control.AQM has become a attention and in-depth study of the hot issues in the field of IP network congestion control, after that a variety of the corresponding theory based on different AQM algorithms emerge In this paper, base on in-depth analysis and comparison during several AQM algorithm, in order to achieve fairness and the target of stability queue length,to enhance the fairness of CHOKe, this theory proposed a new AQM algorithm based on delay-jitter CZ-CHOKe. The algorithm through the delay jitter and the average queue length to detect network congestion,if congestion approaching, it choose the source to notify the congestion randomly, reduce the transmission numbers of the data in the Network before the queue overflows, to ease the network congestion, at the same time using dynamic sampling hit and adaptive packet loss technology. the imposition of penalties for non-response flows more effectively and precisely, improve the fairness of the algorithm, drop probability algorithms use second-order function to ensure the stability of the queue length. NS2 network simulation platform experimental results show that the algorithm can not only detect network congestion more effectively,adapt to changes in flow volume, but also can protect respond flow effectively and improve the hit probability of non-response flow, improve the fairness of the algorithm and stability of queue length.