论文部分内容阅读
提出了一种新的估计端到端时延的方法——往返规划算法,它是基于已有的线性规划算法,同时对相对两个方向上的时延进行测量,把不同分组在相对方向上的时延之和看作为同一个分组的“往返时延”以消除通信双方时钟时间差对端到端时延的影响,当相对方向上经历最小时延的分组在不同时刻发送时,这个“往返时延”还受到双方时钟频率差的影响,算法采用了调整其中任一分组发送时刻的方法消除此影响,并与线性规划算法以及Paxson算法作了性能比较,理论及实验结果均表明在不增加复杂度的前提下往返规划算法估计出的端到端时延更加准确.
A new method to estimate the end-to-end delay is proposed, which is based on the existing linear programming algorithm. It also measures the delay in two opposite directions, Of the same packet as “round trip delay ” to eliminate the influence of the difference of the clock time of both communication parties on the end-to-end delay. When the packet with the smallest delay in the opposite direction is sent at different time, “Round-trip delay ” is also affected by the difference between the two clock frequencies, the algorithm used to adjust any one of the packet sending time to eliminate this effect, and linear programming algorithm and Paxson algorithm made a performance comparison, the theory and experimental results It shows that the end-to-end delay estimated by the round-trip planning algorithm is more accurate without increasing the complexity.