本文提出了一种对叠加分数布朗运动两分量分别进行极大似然估计的算法．首先推导了重抽样情况下分数高斯噪声的小波系数方整公式，并讨论得知不同Ｈ值时方差随尺度变化存在两种截然不同的变化趋势。利用这一特点，可以对叠加的分数布朗运动的参数分别进行极大似然估计，仿真结果表明了该算法的有效性。 In this paper, an algorithm for maximum likelihood estimation of two components of superposition fractional Brownian motion is proposed respectively. Firstly, the square coefficient of wavelet coefficients of fractional Gaussian noise under the condition of re-sampling is deduced. It is discussed that there are two distinct variations of variance with the change of scale when different H values ​​are known. Using this feature, maximum likelihood estimation can be performed separately on the parameters of the fractional Brownian motion, and the simulation results show the effectiveness of the proposed algorithm.