Research Of Congestion Control Algorithm In Trustworthy Network

The 21st century is the era of the Internet, network technology development can be rapid, increasing demand for Internet services, this requires that the network can not only support low speed / high-speed data streams, images and other non-voice transmission, but also to support video streaming and other multimedia services transmission, but the influx of data greatly increase the network the network congestion situation has seriously affected the stability of the network. Therefore, network congestion control and stable operation of the Internet has become the key to maintaining one of the means of building the next generation of high-speed Internet technology one of the difficulties.The early Internet, network congestion control is mainly based on the TCP end nodes end congestion control mechanism, with the rapid growth of the network, the network appears in a large number of UDP streams such as end to end congestion control mechanisms that are not the non-adaptive streaming, occupied bandwidth, and some to the emergence of malicious traffic. Sharp increase in malicious network behavior, so the network reliability and security is facing a severe test. Appropriate active queue management (AQM) mechanism can effectively reduce network congestion, quality of service to ensure the Internet (Qos).In the congestion control mechanism, the fairness of bandwidth allocation is a measure of performance of active queue management algorithm is very important criteria. This paper studied on the basis of classical RED algorithm focuses on a stateless fair queue management algorithm (CHOKe algorithm) fairness. CHOKe algorithm is a completely stateless algorithm, it does not need other special data structure, but CHOKe fairness algorithm is not very good. Therefore, this paper presents an improved fairness stateless fair queue management algorithm (MA-CHOKe algorithm). NS2 network simulation device through a new algorithm for the simulation results show that: the improved algorithm to adapt to flow in the protection of the premise of increased efforts to punish non-adaptive streaming which has a more excellent fairness.