一种改进的LMS自适应滤波算法

An Improved LMS Adaptive Filtering Algorithm

在信号采集处理中,为了更有效地将信号中的噪声剔除,对原始信号的还原程度更高。针对现如今已经存在的LMS变步长算法提出了一种改进方案,既在基于tanh函数的LMS算法的基础上,通过建模推导并引入了一个调节因子。若系统处于大误差情况下,算法能够保持较快的收敛速度,同时又保证了系统误差较小时的收敛精度,不易发生过滤波。设计完成后通过Matlab进行了多次仿真验证,结果表明,在均值为0的高斯白噪声条件下,相比于原有LMS算法,收敛速度提高了10%,滤波精度提升了5%,MES有了0.2dB的提升,所以改进后的算法也有着良好的应用前景。

In the signal acquisition and processing,in order to eliminate the noise in the signal more effectively,the restora⁃tion degree of the original signal is higher.An improved LMS variable step size algorithm is proposed,which is based on the LNs al⁃gorithm based on the tanh function,and an adjustment factor is introduced through modeling and derivation.If the system is in the case of large error,the algorithm can maintain a fast convergence speed and ensure the convergence accuracy when the system error is small,so it is not easy to filter.The algorithm is verified by Matlab modeling system simulation.The simulation results show that compared with the original LMS algorithm,MES has an improvement of 0.2dB under the condition of Gaussian white noise with mean value of 0,so the improved algorithm also has a good application prospect.