论文部分内容阅读
提出了一种基于自回归AR模型的Hurst系数的估计方法,并给出了数学推导过程。采用真实网络突发业务的仿真结果表明,该文所提出的方法比传统的R/S法等估计方法具有更高的估计精度,能更好地反映真实网络业务流量的自相似性。该方法可望用于网络业务流量的管理和网络拥塞控制。
A method of estimating Hurst coefficients based on autoregressive AR model is proposed and the mathematical derivation process is given. The simulation results using real network burst service show that the proposed method has higher estimation accuracy than the traditional R / S method and can better reflect the self-similarity of real network traffic. This method is expected to be used for network traffic management and network congestion control.