LMS

  最小均方（LMS）自适应算法就是一中已期望响应和滤波输出信号之间误差的均方值最小为准的，依据输入信号在迭代过程中估计梯度矢量，并更新权系数以达到最优的自适应迭代算法。LMS算法是一种梯度最速下降方法，其显著的特点是它的简单性。这算法不需要计算相应的相关函数，也不需要进行矩阵运算。(Minimum mean-square (LMS) adaptive algorithm is that one has to respond to the expectations and filtering the output signal of the mean square error between the value of the minimum-based, based on the input signal is estimated in the iterative process of gradient vector, and to update the weights in order to achieve the most Adaptive iterative algorithm gifted. LMS algorithm is a steepest descent gradient method, and its significant features is its simplicity. This algorithm does not require calculation of the corresponding correlation function, nor the need for matrix calculation.)