The Standard (conventional) adaptive algorithms exhibits low convergence rate and minimum noise suppression, or else the system becomes unstable under Gaussian and non-Gaussian (impulsive noise SαS distributions) noise environments. In order to overcome the drawback of traditional algorithms (i.e., to eliminate unwanted noise), the popular algorithm Filtered Cross Minimum Square (FxLMS) is used in Active Noise Control (ANC), not only to improve its efficiency but also to improve its performance. In this paper, we proposed two improvements: first, we proposed a novel method Active threshold function FxLMS (ATFxLMS) being employed in ANC in the paths of primary (reference) and error signals; a second proposal is employing the Variable Step-Size based on Absolute Harmonic Mean (AHMVSS) of error signal. The idea behind this method is that the step-size of the algorithm varies depending on the harmonic mean of error signal obtained from the error location. In comparison to the fixed step-size algorithm, the proposed ATF-AHMVSS provided an improved convergence rate for the desired ANC efficiency. Moreover computational complication of the proposed method was examined as it was found that the proposed algorithm provided stable condition for ANC systems. Computer simulation results are revealed that the proposed (AT & AHMVSS-FxLMS) algorithm have attained excellent performance in terms of convergence speed, noise reduction and minimum steady state error as compared to other existing algorithms under different noise inputs. The results obtained from the proposed algorithm show outperformance compared to traditional adaptive algorithms.

标准（常规）自适应算法显示出低收敛速度和最小的噪声抑制，否则系统在高斯和非高斯（脉冲噪声SαS分布）噪声环境下会变得不稳定。为了克服传统算法的缺陷（即消除不必要的噪声），在有源噪声控制（ANC）中使用了流行的算法“最小交叉平方滤波”（FxLMS），不仅提高了效率，而且提高了性能。在本文中，我们提出了两项​​改进：首先，我们提出了一种新的方法：主动阈值函数FxLMS（ATFxLMS）在ANC中的主要（参考）信号和误差信号路径中使用；第二种建议是采用基于误差信号的绝对谐波均值（AHMVSS）的可变步长。该方法背后的思想是算法的步长大小取决于从错误位置获得的错误信号的谐波均值。与固定步长算法相比，提出的ATF-AHMVSS为期望的ANC效率提供了更高的收敛速度。此外，检查了所提出的方法的计算复杂性，因为发现所提出的算法为ANC系统提供了稳定的条件。计算机仿真结果表明，与其他现有算法在不同噪声输入下相比，所提出的（AT＆AHMVSS-FxLMS）算法在收敛速度，降噪效果和最小稳态误差方面均具有优异的性能。与传统的自适应算法相比，从所提出的算法获得的结果显示出了优异的性能。