Active noise reduction method for off-line modeling of secondary channel based on EMFNL filter

Abstract

The invention relates to the field of active noise control. The invention discloses an active noise reduction method for off-line modeling of a secondary channel based on an EMFNL (Even Mirror FourierNonlinear with Linear section) filter with a linear part. The method comprises the following steps: S1, constructing an EMFNL filter tap, and identifying a coefficient of the EMFNL filter tap by adopting an adaptive algorithm to realize offline modeling of a secondary channel; S2, calculating secondary channel estimation corresponding to the secondary channel coefficient identified in S1; and S3,realizing an active noise reduction method by using secondary channel estimation. Compared with the prior art, a secondary channel is modeled nonlinearly based on the second-order EMFNL filter, and meanwhile, the EMFNL filter is used as a control filter to perform active noise control. Secondary channel estimation modeled through the method is a nonlinear model, and compared with a linear secondary channel, the nonlinear processing capacity is higher.

Description

technical field [0001] The present invention relates to the field of active noise control, to an active noise control method and a nonlinear secondary channel modeling method, and in particular to a Fourier nonlinear (EMFNL, Even Mirror Fourier Nonlinear with Linear section) filter based on an even mirror with a linear section Active noise reduction method for off-line modeling of the secondary channel. Background technique [0002] The active noise control (ANC, Active Noise Control) technology based on the superposition principle has been widely researched and applied due to its advantages of low cost, significant low-frequency effect, and simple deployment and control. It is very likely to become the standard technology for noise reduction in enclosed spaces in the future. [0003] Active noise control models are divided into models with secondary channels and models without secondary channels. In terms of no secondary channel model, Chinese patent CN 101393736B disclose...

AI Technical Summary

Problems solved by technology

In terms of no secondary channel model, Chinese patent CN 101393736B discloses an active noise control method without secondary channel modeling, using four update direction search methods to find optimal coefficients, poor real-time performance

Chinese patent CN 103915091 A discloses a non-secondary channel modeling method. This method needs to count the noise source signal and error signal power. It is essentially a statistical method. It is difficult to achieve real-time in the initial stage of the system, and when the noise source changes , the system is difficult to respond quickly

International patent WO2017 / 048480EN 2017.03.23 (Chinese patent CN 108352156 A) and international patent WO2017 / 048481 EN 2017.03.23 (Chinese patent CN 108352157 A) disclose the estimation method of the amplitude and phase of the secondary channel, which requires different frequency components To estimate, the algorithm is complex

Chinese patent CN 109448686 A discloses an online secondary modeling active noise control system, which uses a linear secondary channel model, which is difficult to deal with nonlinear secondary channel situations

Chinese patent CN 109379652 A discloses a secondary channel offline identification method and system for earphone active noise control. The secondary channel uses an infinite impulse response (IIR, Infinite Impulse Response) filter. Fewer coefficients approach linear filters, but the IIR model is unstable under the Bounded Input Bounded Output (BIBO, Bounded Input Bounded Output) criterion

[0004] In terms of control filters, commonly used filters include function extension filters, convex combination filters, recursive filters, etc., as disclosed in patent CN 107181474 A Chebyshev and Legendre filter, patent CN 106603038 B discloses a finite impulse response (Finite Impulse Response, FIR) convex combination filter, patent CN109617537 A discloses a function expansion bilinear filter, but the above filter Both can only realize the secondary channel estimation as a linear model, compared with the nonlinear secondary channel, the nonlinear processing ability is greatly reduced

Images