一种面向车内噪声控制的改进变步长LMS算法  

An improved variable step-size LMS algorithm for vehicle interior noise control

作　　者：钱梦男 卢剑伟 晏桂喜 郭嘉豪 QIAN Mengnan;LU Jianwei;YAN Guixi;GUO Jiahao(School of Automobile and Traffic Engineering, Hefei University of Technology, Hefei 230009, China;School of Mechanical Engineering, Hefei University of Technology, Hefei 230009, China)

机构地区：[1]合肥工业大学汽车与交通工程学院,安徽合肥230009 [2]合肥工业大学机械工程学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2021年第10期1306-1310,共5页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(51875150)。

摘　　要：车载有源噪声控制(active noise control,ANC)系统多基于最小均方(least mean square,LMS)算法来实现,但定步长LMS算法在收敛速度和稳态误差之间存在矛盾。文章基于Softsign函数建立误差与步长之间的非线性关系,提出了一种改进变步长LMS算法。仿真算例结果表明,与定步长LMS算法及Tanh-LMS算法相比,该算法具有更快的收敛速度和更小的稳态误差,且比Tanh-LMS算法运算量更小,因而在车载ANC系统上应用更具优势。Active noise control(ANC)for vehicle interior noise is mostly based on the least mean square(LMS)algorithm,but the contradiction of convergence speed and the steady-state error exists in the LMS algorithm.In order to deal with this problem,this paper presents an improved variable step-size LMS algorithm based on Softsign function.This algorithm establishes the nonlinear relationship between error and step size.The simulation results show that compared with the LMS algorithm and Tanh-LMS algorithm,this algorithm has quicker convergence speed and smaller steady-state error.Moreover,this improved algorithm has smaller computation compared with Tanh-LMS,which makes it more applicable in the vehicle ANC system.

关 键 词：Softsign函数 最小均方(LMS)算法 变步长 车内噪声 

分 类 号：TB535.2[理学—物理]