基于动量LMS算法的变步长次级路径在线建模ANC系统

ANC System for Online Modeling of Variable Step Size Secondary Path Based on Momentum LMS Algorithm

针对ANC系统收敛速度慢、稳定性差、次级通道建模精度低,以及降噪性能弱的问题,提出了动量LMS算法。采用宽带前馈型主动噪声控制的结构进行系统设计,有效处理非周期低频噪声信号;设计动量LMS算法,对次级通道路径进行实时在线建模;提出利用建模精度ΔS的变化来决定步长变化的2种方法,并在LMS算法基础上增加了2个动量项。仿真结果表明,所提2种方法有效地提高了ANC系统的降噪性能、算法的收敛速度及建模精度,降噪量最高分别可达17 dB,16 dB。

The momentum LMS algorithm is proposed to solve the problems of slow convergence speed,poor stability,low of secondary channel modeling accuracy and weak noise reduction performance in the ANC system.The system design using broadband feedforward active noise control is used to effectively process aperiodic low-frequency noise signals;the momentum LMS algorithm is designed to perform real-time online modeling of the secondary channel path;there are two methods to change the step size,and two momentum terms are added based on the LMS algorithm.The simulation results show that the proposed two methods can effectively improve the noise reduction performance of the ANC system and the convergence speed and modeling accuracy of the algorithm,and the highest noise reduction can reach 17dB and 16dB respectively.