Active noise control system based on variable step LMS algorithm

Abstract

The invention discloses an active noise control system based on a variable step LMS algorithm. According to the system, a filter based on the variable step LMS algorithm is applied to identification of a secondary channel; a filter weight value is updated by adjusting a step value of the filter; in order to obtain a relatively fast convergence rate, the step value of self-adaptive filtering in an initial phase is relatively high; and when the self-adaptive filtering approaches a steady state, the step size is reduced, so a relatively low steady state error can be obtained. The variable step LMS algorithm is applied to the identification of the secondary channel of the active noise control system, and the contradiction among the convergence rate, the steady state error and a tracking capability of a time-varying system can be solved well.

Description

technical field [0001] The invention relates to the technical field of active noise reduction, in particular to an active noise control system based on a variable step size LMS algorithm. Background technique [0002] In recent years, active noise reduction technology has gradually become practical. Active noise control systems generally include two parts: feedforward control and secondary channel identification using the FX-LMS algorithm. The fixed-step LMS algorithm requires less calculation and is easy to implement in hardware. , so the secondary channel identification in the general control system adopts the adaptive filtering of the fixed-step LMS algorithm, but the steady-state offset of the fixed-step LMS algorithm is proportional to the filter order and step size, and inversely proportional to the convergence time, and also That is, increasing the step size can speed up the convergence of the algorithm, but the steady-state offset will become larger; reducing the ste...

AI Technical Summary

Problems solved by technology

[0002] In recent years, active noise reduction technology has gradually become practical. Active noise control systems generally include two parts: feedforward control and secondary channel identification using the FX-LMS algorithm. The fixed-step LMS algorithm has a small amount of calculation and is easy to implement in hardware. , so the secondary channel identification in the general control system adopts the adaptive filtering of the fixed-step LMS algorithm, but the steady-state misalignment of the fixed-step LMS algorithm is proportional to the filter order and step size, and is inversely proportional to the convergence time. That is, increasing the step size can speed up the convergence of the algorithm, but the steady-state misalignment will become larger; reducing the step size can reduce the steady-state misalignment, but the algorithm convergence speed will slow down

Images