一种非线性变步长LMS自适应滤波算法

A Nonlinear Variable Step Size LMS Adaptive Filtering Algorithm

基于定步长LMS自适应滤波算法存在稳态误差和收敛速度无法合理协调的缺点,通过在误差信号与步长之间建立非线性关系,提出了一种新的变步长LMS算法。通过步长函数可以让步长因子在算法初始阶段具有较大值,在算法收敛阶段获得较小值,且变化平缓。理论分析和实验仿真结果表明,新算法克服了在稳态自适应阶段步长调整导致系统不稳定的问题。将新算法应用于电脑风扇噪声的有源噪声控制中,取得了良好的效果。

The fixed-step LMS adaptive filtering algorithm has the disadvantages that the steady-state error and the convergence speed cannot be reasonably coordinated.A new variable step size LMS algorithm is proposed by establishing a nonlinear relationship between the error signal and the step size.The step function can make the step factor have a larger value in the initial stage of the algorithm and obtain a smaller value in the convergence phase of the algorithm.And the step factor changes smoothly.Theoretical analysis and experimental simulation results show that the new algorithm overcomes the problem of system instability caused by step size adjustment in the steady-state adaptive phase.The new algorithm has been applied to the active noise control of computer fan noise,and good results have been achieved in the control.