通过建立步长因子μ与误差信号e之间的非线性关系,将新的变步长LMS算法运用到自适应噪声对消(ANC)中。分析比较了该算法与固定步长LMS算法及箕舌线变步长算法的统计性能。该算法除了具有固定步长LMS算法计算量小、稳定性好、简单等优点外,其收敛速度及稳定性优于箕舌线变步长算法。仿真结果表明:该算法用于降噪系统中能够取得较好的效果。
An variable step size adaptive filtering algorithm is presented by building a nonlinear function relationship between the step size factor μ and the error signal e,and it is applied to the ANC (Adaptive Noise Cancellation) system.The comparison of the statistical performance between variable step size,LMS (least mean square) algorithms and LMS algorithm based on tongue-like curve is analyzed.The algorithm is simple and easy to be implemented.Computer simulation shows that the performances of the algorithm such as convergence speed,steady state error are better than the LMS algorithm based on tongue-like curve,and good results in ANC are obtained.