视频会议系统在自相似网络上的速率控制

视频会议系统中的实际数据流量具有与传统的短相关模型差别显著的长相关性 ,这使得需要更新以往的流量控制算法以适应新兴的视频通信的要求。为此 ,在局域网上运行的视频会议系统的速率控制方面 ,提出了对源端的发送速率进行双时间标度控制的算法。它采用 QOS(业务质量 )分层的思想 ,在小时间标度上 ,根据接收方的反馈信息 ,选择一定层次质量的数据进行发送 ;在大时间标度上 ,控制相邻层次变换的难易程度 ,以减小层次间的振荡 ,使接收方得到稳定质量的图像。实验表明 ,该算法可以适应网络情况 ,为系统提供尽可能好的并且稳定的业务质量

Rate control algorithm in video conferencing system for self similar network traffic

Recent analytical and empirical studies have shown that self similarity is an inherent characteristic of VBR video traffic in disagreement with traditional Markovian models. Therefore, traditional flow control algorithms are no longer applicable to such traffic. A two time scale rate control algorithm is designed to control the sending rate considing the self similar charcteristics of the network traffic in a video conferencing system. In the smaller time scale with QOS (Quality Of Service) scaling, the sender sends a portion of data and discards the rest according to the feedback message from the receiver. Therefore, the main QOS requested by the system is maintained. In the larger time scale, the algorithm adjusts the distance between the adjacent levels based on the long range dependency of the traffic. The purpose is to reduce the oscillation of the sending levels and stabilize the picture quality at the receiver side. Our simulated result shows that depending on the network traffic throughput, the system provides the best and most stable quality of service.