中文说明：声学噪声取消。 自适应滤波的算法的推导过程基于维纳过滤 theoryhave 他们的起源在解决问题的统计公式。与此相反，最小二乘方法从确定性的角度来看待方法筛选优化问题。前面提到过，在维纳滤波理论，所需的过滤器是获得个性化 MSE，那就是，一个统计量。在最小二乘法，另一方面，该方法的性能指标是对于给定的数据，那就是，一个确定性的数量加权的误差平方和。结果这种确定性的方法是最小二乘-basedalgorithms，在一般情况下，远远高于基于 LMS 算法收敛。他们也对输入信号的功率谱密度不敏感。是保障来实现这一改进的收敛性能的价格是高计算复杂性及其综合数值稳定性较差。
English Description：
Acoustic Noise Cancellation. The adaptive filtering algorithms whose derivations are based on the Wiener filter theoryhave their origin in a statistical formulation of the problem. In contrast to this, the methodof least-squares approaches the problem of filter optimization from a deterministic pointof view. As mentioned before, in the Wiener filter theory, the desired filter is obtained byminimizing the MSE, that is, a statistical quantity. In the method of least-squares, on theother hand, the performance index is the sum of weighted error squares for the given data,that is, a deterministic quantity. A consequence of this deterministic approach is that the least-squares-basedalgorithms, in general, converge much faster than the LMS-based algorithms. They arealso insensitive to the power spectral density of the input signal. The price that is paidfor achieving this improved convergence performance is higher computational complexityand poorer numerical stability.