摘要： 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。

关键词: 网络突发业务；自相似建模；Hurst 系数；参数估计

Abstract: This paper proposes a Hurst index estimation method based on autoregressive(e.g. AR) model, and presents mathematic inference process. Simulation results based on real traffic data show that the proposed method is better than those methods such as R/S, at the same time, using this method to estimate Hurst index can reveal the self-similar of real network traffic well. Thus this method can be applied to the network traffic management and network congestion control.

Key words: Network traffic; Self-similar modeling; Hurst index; Parameter estimation