The Research Of Streaming Media Network Congestion Control Algorithm On The Basis Of The Rates

Along with the wide application of streaming media and people’s pursuit of high quality service. streaming media transmission technology becomes a focus in the field of network application. Streaming media, set audio, video and graphic all rolled into one, has characteristics that it contains large amount of data, and makes low delay and low jitter. It adopts UDP rather than TCP. However, UDP does not provide any congestion control mechanism that makes no guarantee to the service quality. In addition, streaming media transmission takes up large proportion of network resource, the lack of congestion control mechanism will affect the operation of the net. Therefore, it is doomed to introduce congestion control mechanism into streaming media transmission. It has been found that the most promising congestion control algorithm should be the one that both guarantee streaming QoS and lay emphasis on the friendly coexistence with TCP data flow.Based on the above mentioned requirements, this paper analyzes the causes of congestion, the characteristics and research significance of congestion control mechanism. Then it elaborates streaming media and real-time transmission and studies the protocol that supports real-time transmission of RTP and RTCP and the basic principle of TCP congestion control. The rate of congestion control algorithm (TFRC) was researched in this paper. In view of some shortcomings of TFRC algorithm, such as volatility sending rate, long time of slow start and the lack of one-way delay jitter control, we introduce a more suitable congestion control algorithm, called FS-TFRC, for streaming media transmission, which makes improvements of TFRC on fairness and smoothness aspect. RTCP feedback information is used to calculate the parameters of the throughput rate RTT and packet loss rate so as to smoothly send rate. Especially in the stage of slow start, FS-TFRC reduces the slow start time ad treats one-way delay jitter as a feedback signal to adjust the sending rate of FS-TFRC. At the same time, it uses an adaptive threshold adjustment scheme which dynamically adjusts jitter delay threshold according to the historical situation. At last, through the simulation experiment, it verifies that the improved algorithm FS-TFRC makes improvements in reducing the jitter of delay, slowing down jitter and ensuring smoothness, and can better adapt to streaming media real-time transmission.