CN105006233A - Narrowband feedforward active noise control system and target noise suppression method - Google Patents

Abstract

The present invention discloses a narrowband feedforward active noise control system and a target noise suppression method which belong to the narrowband active noise control technology field. The system comprises a frequency tracking and reference signal generating module, a reference signal delay module, a first amplitude phase regulating module, a second amplitude phase regulating module, a first accumulation module, a secondary channel, a second accumulation module, a first parameter regulating module, a second parameter regulating module and a third parameter regulating module. A reference signal of the method is generated by an AR model according to the source noise frequency picked up by a non-acoustic sensor, a part of secondary source is generated by an amplitude phase regulator, and the other part of secondary source is generated by a delay module and the other one amplitude phase regulator. When the non-acoustic sensor exists an error, the two parts of secondary source is synthesized to be superposed with the target noise, so that more than 10% of frequency mismatching amount can be replied effectively, and the system is fast in convergence speed.

Description

Arrowband feedforward active noise control system and target noise suppressing method

Technical field

The invention discloses arrowband feedforward active noise control system and target noise suppressing method, belong to the technical field of arrowband active noise controlling (ANC, Active Noise Control).

Background technology

It is good that active noise controlling method has low frequency characteristic relative to passive noise control method, and the advantages such as controller size is little, are highly suitable for the suppression of the low frequency sinusoidal type narrow band noise of cycle or approximate period.This narrow band noise often by slewing or have reciprocating device produce.In arrowband active noise controlling, for avoiding acoustic feedback, noise frequency directly or indirectly records often through non-acoustic sensor, according to surveyed synchronizing frequency, then be supplied to the secondary noise of control wave filter generation for offsetting target noise through the sinusoidal pattern reference signal of signal generator generation same frequency.

Due to non-acoustic sensor long-play, the reason such as aging, survey between synchronizing frequency and target noise actual frequency and will there is error, namely create frequency and do not mate phenomenon.At this moment, arrowband active noise control system will be difficult to successfully manage, noise suppression effect degradation.

Summary of the invention

Technical matters to be solved by this invention is the deficiency for above-mentioned background technology, provide arrowband feedforward active noise control system and target noise suppressing method, achieving arrowband active noise control system can tracking target noise frequency when there is frequency and not mating, and solves the arrowband feedforward ANC system technical matters that calculated amount is comparatively large when tackling frequency and not mating, speed of convergence is slower.

The present invention adopts following technical scheme for achieving the above object:

Arrowband feedforward active noise control system, comprise: frequency-tracking and reference signal generation module, delayed reference signal module, the first amplitude phase adjustment block, the second amplitude phase adjustment block, the first accumulator module, secondary channel, the second accumulator module, the first parameter adjustment module, the second parameter adjustment module, the 3rd parameter adjustment module, wherein

Frequency-tracking and reference signal generation module: output terminal is connected with delayed reference signal module input, the first amplitude phase adjustment block input end, adopt the frequency of AR model tracking target noise under the synchronizing frequency of non-acoustic sensor collection and generate sinusoidal pattern reference signal, exporting sinusoidal pattern reference signal to delayed reference signal module, the first amplitude phase adjustment block;

Delayed reference signal module: output terminal is connected with the second amplitude phase module input end, exports and postpones reference signal to the second amplitude phase adjustment block;

First accumulator module: an input termination first amplitude phase adjustment block output terminal, another input termination second amplitude phase adjustment block output terminal, export termination secondary channel input end, after receiving a part of secondary source signal that the first amplitude phase adjustment block exports and another part secondary source signal that the second amplitude phase adjustment block exports, export secondary source composite signal to secondary channel;

Second accumulator module: an input termination secondary channel output terminal, another input termination target noise signal, exports residual noise signal after the secondary noise signal that receiving target noise signal and secondary channel export;

First parameter adjustment module: the filtering signal of an input termination sinusoidal pattern reference signal, another input termination residual noise signal, exports the parameter updated value of the first amplitude phase adjustment block;

Second parameter adjustment module: an input termination postpones the filtering signal of reference signal, another input termination residual noise signal, exports the parameter updated value of the second amplitude phase adjustment block;

3rd parameter adjustment module: the filtering signal of an input termination sinusoidal pattern reference signal, another input termination residual noise signal, the auto-adaptive parameter updated value of output frequency tracking and reference signal generation module.

Further, in described arrowband active noise control system, the first amplitude phase adjustment block is the first FIR filter, and the second amplitude phase adjustment block is the second FIR filter.

Further, in described arrowband active noise control system, the first parameter adjustment module, the second parameter adjustment module, the 3rd parameter adjustment module all adopt LMS algorithm undated parameter.

Further, in described arrowband active noise control system, target noise signal is generated with superposing of environment random noise in a narrow band noise produced that disappears mutually through linear primary channel propagation by narrowband noise sources.

Target noise suppressing method, adopts described system to realize, comprises the steps:

A. the acquisition of reference signal: frequency-tracking and reference signal generation module produce sinusoidal pattern reference signal under the synchronizing frequency of non-acoustic sensor collection, delayed reference signal resume module sinusoidal pattern reference signal obtains postponing reference signal;

B. the acquisition of secondary source composite signal: the first FIR filter aligns chordwise reference signal and carries out amplitude phase adjusted and obtain a part of secondary source signal, second FIR filter is carried out amplitude phase adjusted to delay reference signal and is obtained another part secondary source signal, and a part of secondary source signal, another part secondary source signal obtain secondary source composite signal after the first totalizer superposition;

C. suppress target noise signal: secondary source composite signal generates secondary noise signal through secondary channel, secondary noise signal and target noise signal superpose to disappear mutually obtain residual noise signal through the second superimposers;

D. undated parameter: adopt secondary channel estimation model process sinusoidal pattern reference signal, delay reference signal to obtain filtering-X reference signal, filtering-X postpones reference signal, first parameter adjustment module is according to the weights of filtering-X reference signal, residual noise signal update first FIR filter, second parameter adjustment module postpones the weights of reference signal, residual noise signal update second FIR filter according to filtering-X, and the 3rd parameter adjustment module is according to the auto-adaptive parameter of filtering-X reference signal, residual noise signal update frequency-tracking and reference signal generation module.

Further, in described target noise suppressing method, the sinusoidal pattern reference signal that steps A produces is:

x _i(n)＝-c _i(n)x _i(n-1)-x _i(n-2),n≥2

x _i(0)＝a _i

X _i(1)=a _icos (ω _i(0))+b _isin (ω _i(0)), wherein,

X _i(0), x _i(1), x _i(n-2), x _i(n-1), x _in () is respectively the 0th time, the 1st time, the n-th-2 times, (n-1)th time, n-th time updated value of sinusoidal pattern reference signal, c _in n-th updated value that () is auto-adaptive parameter, ω _i(0) be the synchronizing frequency that non-acoustic sensor records, auto-adaptive parameter initial value c _i(0): c _i(0)=-2cos (ω _i(0)).

Further, in described target noise suppressing method, a part of secondary source signal that step B obtains and another part secondary source signal be respectively:

$\begin{array}{c}{y}_i^{\left(0\right)}\left(n\right)=h_{i,0}^{\left(0\right)}\left(n\right)x_i\left(n\right)+h_{i,1}^{\left(0\right)}\left(n\right)x_i(n-1)\\ y_i^{(-k)}\left(n\right)=h_{i,0}^{(-k)}\left(n\right)x_i(n-k)+h_{i,1}^{(-k)}\left(n\right)x_i(n-k-1)\end{array},$ Wherein,

be n-th updated value of the first FIR filter weights, x _i(n-1), x _in (n-1)th time, n-th time updated value that () is sinusoidal pattern reference signal, be n-th updated value of the second FIR filter weights, x _i(n-k-1), x _i(n-k) for postponing (n-1)th time, n-th time updated value of reference signal.

Further, described target noise suppressing method, target noise p (n) described in step C is:

$p\left(n\right)=\underset{i=1}{\overset{q}{\mathrm{\Σ}}}[a_{p,i}{x}_{a_i}\left(n\right)+b_{p,i}{x}_{b_i}\left(n\right)]+v_p\left(n\right),$ Wherein,

Q is target noise frequency component quantity, a _p,i, b _p,ifor discrete fourier coefficient, for cosine signal component, for sinusoidal signal component, $x_{a_i}\left(n\right)=\cos \left({\mathrm{\ω}}_{p,i}n\right),x_{b_i}\left(n\right)=\sin \left({\mathrm{\ω}}_{p,i}n\right),$ ω _p,ifor target noise signal frequency, v _pn () is environment random noise.

As the further prioritization scheme of described target noise suppressing method, in step D, the renewal equation of the first FIR filter weights is: $\begin{array}{c}{h}_{i,0}^{\left(0\right)}(n+1)=h_{i,0}^{\left(0\right)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i\left(n\right)\\ h_{i,1}^{\left(0\right)}(n+1)=h_{i,1}^{\left(0\right)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i(n-1)\end{array},$ The renewal equation of the second FIR filter weights is: $\begin{array}{c}{h}_{i,0}^{(-k)}(n+1)=h_{i,0}^{(-k)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i(n-k)\\ h_{i,1}^{(-k)}(n+1)=h_{i,1}^{(-k)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i(n-k-1)\end{array},$ The renewal equation of frequency-tracking and reference signal generation module auto-adaptive parameter is: $c_i(n+1)=c_i\left(n\right)-{\mathrm{\μ}}_{c_i}{h}_{i,0}^{\left(0\right)}\left(n\right)e\left(n\right){\hat{x}}_i(n-1),$ Wherein,

be (n+1)th updated value of the first FIR filter weights, be (n+1)th updated value of the second FIR filter weights, for (n-1)th time, n-th time updated value of filtering-X reference signal, for filtering-X postpones (n-1)th time, n-th time updated value of reference signal, μ _ifor sinusoidal pattern reference signal and the renewal coefficient postponing reference signal, for the renewal coefficient of auto-adaptive parameter.

The present invention adopts technique scheme, has following beneficial effect:

(1) frequency-tracking and reference signal generation module make use of the frequency of adaptive notch filter tracking noise target, effectively also can produce the sinusoidal pattern reference signal relevant to target noise by tracking target noise frequency;

(2) the part secondary source that the delayed reference signal module increased and the second amplitude phase module generate, the part secondary source obtained through amplitude phase adjusted with reference signal to disappear the minimum residual noise of produce power through superposition, mutually with less calculated amount elevator system speed of convergence;

(3) introducing of reference delay signal module, amplitude phase adjustment block have impact on and adopts the parameter of the tracking of AR model frequency and reference signal generation module to upgrade, the frequency that whole system can be made to tackle not the amount of coupling up to more than 10%, and be easy in the active noise controlling device of reality realize;

(4) target noise suppressing method is by simple parallel-expansion, and the frequency being easy to realize under multi-frequency lane does not mate reply, reaches the object effectively suppressing narrow band noise.

Accompanying drawing explanation

Fig. 1 is the theory diagram of arrowband of the present invention active noise control system.

Number in the figure illustrates: 1, frequency-tracking and reference signal generation module, 2, delayed reference signal module, 31, the first amplitude phase adjustment block, 32, the second amplitude phase adjustment block, 41, the first parameter adjustment module, 42, the second parameter adjustment module, the 43, the 3rd parameter adjustment module, the 5, first accumulator module, 6, secondary channel, the 7, second accumulator module.

Embodiment

Be described in detail below in conjunction with the technical scheme of accompanying drawing to invention.

The invention provides the unmatched countermeasure of a kind of arrowband active noise controlling medium frequency, for making object of the present invention, technical scheme and effect clearly, clearly, with reference to accompanying drawing examples, the present invention being described in further detail.Should be appreciated that concrete enforcement described herein is only in order to explain the present invention, is not intended to limit the present invention.

Those skilled in the art will appreciate that unless otherwise defined, all terms used herein have (comprising technical term and scientific terminology) the identical meaning of the general understanding of the ordinary technical staff in the technical field of the invention.Should also be understood that those terms defined in such as general dictionary should be understood to have the meaning consistent with the meaning in the context of prior art, unless and define as here, can not explain by idealized or too formal implication.

Arrowband provided by the invention active noise control system principle as shown in Figure 1, provides one to comprise: the arrowband feedforward ANC system of frequency-tracking and reference signal generation module 1, delayed reference signal module 2, two amplitude phase adjustment block (being the first amplitude phase adjustment block 31 and the second amplitude phase adjustment block 32), three least mean square algorithm modules (being the first parameter adjustment module 41, second parameter adjustment module 42 and the 3rd parameter adjustment module 43), the first accumulator module 5, secondary channel 6, second accumulator module 7 composition.The output terminal of frequency-tracking and reference signal generation module 1 and delayed reference signal module 2 input end, first amplitude phase adjustment block 31 input end connects, delayed reference signal module 2 output terminal is connected with the second amplitude phase module 32 input end, input termination first amplitude phase adjustment block 31 output terminal of the first accumulator module 5, another input termination second amplitude phase adjustment block 32 output terminal of first accumulator module 5, output termination secondary channel 6 input end of the first accumulator module 5, input termination secondary channel 6 output terminal of the second accumulator module 7, another input termination target noise signal of second accumulator module 7, the filtering signal of an input termination sinusoidal pattern reference signal of the first parameter adjustment module 41, another input termination residual noise signal of first parameter adjustment module 41, an input termination of the second parameter adjustment module 42 postpones the filtering signal of reference signal, another input termination residual noise signal of second parameter adjustment module 42, the filtering signal of an input termination sinusoidal pattern reference signal of the 3rd parameter adjustment module 43, another input termination residual noise signal of 3rd parameter adjustment module 43.

Frequency-tracking and reference signal generation module 1, by an AR model-composing, also produce reference signal for tracking noise frequency automatic under given frequency initial value condition simultaneously.Amplitude phase adjustment block is single order FIR filter, carry out superposing with target noise signal again by secondary channel 6 after the part secondary source Signal averaging that they produce and form the inhibition that disappears mutually, delayed reference signal module 2 is set before the second amplitude phase adjustment block 32 and produces the delay reference signal inputted as the second amplitude phase adjustment block 32.Auto-adaptive parameter in frequency-tracking and reference signal generation module 1, the filter weights of amplitude phase adjustment block carry out Automatic adjusument by least mean square algorithm module 43,41 and 42 respectively;

The unmatched target noise suppressing method of reply frequency is according to following steps:

Step 1, frequency-tracking and reference signal generation module 1, produce sinusoidal pattern reference signal according to the synchronous frequency signal obtained by non-acoustic sensor, be supplied to the first amplitude phase adjustment block 31;

Described reference signal is formed through Postponement module 2 and postpones reference signal, is supplied to the second amplitude phase adjustment block 32;

Target noise p (n) at a noise formed that disappears mutually, can be expressed as after to be narrow-band source noise propagate via linear primary channel

$p\left(n\right)=\underset{i=1}{\overset{q}{\mathrm{\Σ}}}[a_{p,i}{x}_{a_i}\left(n\right)+b_{p,i}{x}_{b_i}\left(n\right)]+v_p\left(n\right)$

Wherein, q is target noise frequency component quantity, a _p,i, b _p,ifor discrete fourier coefficient, for cosine signal component, for sinusoidal signal component, $x_{a_i}\left(n\right)=\cos \left({\mathrm{\ω}}_{p,i}n\right),x_{b_i}\left(n\right)=\sin \left({\mathrm{\ω}}_{p,i}n\right),$ ω _p,ifor target noise frequency, v _pn () is additivity neighbourhood noise;

The sinusoidal pattern reference signal that frequency-tracking and reference signal generation module 1 generate is:

x _i(n)＝-c _i(n)x _i(n-1)-x _i(n-2),n≥2

x _i(0)＝a _i

x _i(1)＝a _icos(ω _i(0))+b _isin(ω _i(0))

Wherein, x _i(0), x _i(1), x _i(n-2), x _i(n-1), x _in () is respectively the 0th time, the 1st time, the n-th-2 times, (n-1)th time, n-th time updated value of sinusoidal pattern reference signal, c _in n-th updated value that () is auto-adaptive parameter; ω _i(0) be the synchronizing frequency that non-acoustic sensor records, when there is error in non-acoustic sensor, this frequency values and target noise frequencies omega _p,ibetween there is deviation.Parameter c _in () reflects frequency-tracking effect, its initial value can be set to: c _i(0)=c _i(1)=-2cos (ω _i(0)).

Step 2, the first amplitude phase adjustment block 31 pairs reference signal carries out amplitude phase adjusted, generates a part of secondary source signal

Described delay reference signal carries out amplitude phase adjusted via the second amplitude phase adjustment block 32, generates another part secondary source signal

Amplitude phase adjustment block 31 and 32 is single order FIR filter, the part secondary source signal of output with for:

$y_i^{\left(0\right)}\left(n\right)=h_{i,0}^{\left(0\right)}\left(n\right)x_i\left(n\right)+h_{i,1}^{\left(0\right)}\left(n\right)x_i(n-1)$

$y_i^{(-k)}\left(n\right)=h_{i,0}^{(-k)}\left(n\right)x_i(n-k)+h_{i,1}^{(-k)}\left(n\right)x_i(n-k-1)$

be n-th updated value of the first FIR filter weights, x _i(n-1), x _in (n-1)th time, n-th time updated value that () is sinusoidal pattern reference signal, be n-th updated value of the second FIR filter weights, x _i(n-k-1), x _i(n-k) for postponing (n-1)th time, n-th time updated value of reference signal;

The method also provides three least mean square algorithm modules 41,42 and 43, and secondary channel S (z) adopts offline identification method (also can adopt on-line identification method) to estimate, and obtains estimation model sinusoidal pattern reference signal is through estimation model process obtains filtering-X reference signal, together with residual noise e (n) cooperatively as the input signal of least mean square algorithm module 41; Postpone reference signal through estimation model process obtains filtering-X and postpones reference signal, together with residual noise e (n) cooperatively as the input signal of least mean square algorithm module 42; Sinusoidal pattern reference signal is through estimation model process obtains filtering-X reference signal, together with residual noise e (n) cooperatively as the input signal of least mean square algorithm module 43;

Least mean square algorithm module 41 is update controller of the first amplitude phase adjustment block 31 median filter weights; Least mean square algorithm module 42 is update controller of the second amplitude phase adjustment block 32 median filter weights; Least mean square algorithm module 43 is update controller of auto-adaptive parameter in frequency-tracking and reference signal generation module 1; Use gradient descent method to upgrade weights and parameter, after the secondary noise of synthesis and target noise being disappeared by regulating the weights upgrading wave filter with parameter mutually, the residue noise power of gained is minimum;

Upgrade filter weights and AR model adaptation parameter by filtering-X lowest mean square (FXLMS) algorithm, comprise and c _in (), makes the secondary noise y of generation _sn () effectively can suppress target noise p (n);

The filter weights of the amplitude phase adjustment block 31 and 32 in figure and the AR model adaptation parameter of frequency-tracking and reference signal generation module 1 upgrade by following equation:

$h_{i,0}^{\left(0\right)}(n+1)=h_{i,0}^{\left(0\right)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i\left(n\right)$

$h_{i,1}^{\left(0\right)}(n+1)=h_{i,1}^{\left(0\right)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i(n-1)$

$h_{i,0}^{(-k)}(n+1)=h_{i,0}^{(-k)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i(n-k)$

$h_{i,1}^{(-k)}(n+1)=h_{i,1}^{(-k)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i(n-k-1)$

$c_i(n+1)=c_i\left(n\right)-{\mathrm{\μ}}_{c_i}{h}_{i,0}^{\left(0\right)}\left(n\right)e\left(n\right){\hat{x}}_i(n-1)$

Wherein, be (n+1)th updated value of the first FIR filter weights, be (n+1)th updated value of the second FIR filter weights, for (n-1)th time, n-th time updated value of filtering-X reference signal, for filtering-X postpones (n-1)th time, n-th time updated value of reference signal, μ _ifor sinusoidal pattern reference signal and the renewal coefficient postponing reference signal, for the renewal coefficient of auto-adaptive parameter;

True secondary channel S (z) and estimation thereof normally be made up of FIR filter, that is:

$S\left(z\right)=\underset{j=0}{\overset{M-1}{\mathrm{\Σ}}}{s}_j{z}^{-j}$

$\hat{S}\left(z\right)=\underset{m=0}{\overset{\hat{M}-1}{\mathrm{\Σ}}}{\hat{s}}_m{z}^{-m}$

Estimate through secondary channel filtered filtering-X reference signal is:

${\hat{x}}_i\left(n\right)=\underset{m=0}{\overset{\hat{M}-1}{\mathrm{\Σ}}}{\hat{s}}_m{x}_i(n-m)$

M, for filter length.

Step 3, described secondary source with secondary source superposition, then through secondary channel S (z), exports secondary noise signal y _s(n); Secondary noise signal y _sn () and target noise signal p (n) disappear mutually and superpose, residual noise e (n) after being disappeared mutually;

Residual noise e (n) in accompanying drawing after target noise counteracting can be calculated as:

e(n)＝p(n)-y _s(n)

$y_s\left(n\right)=p\left(n\right)-\underset{i=1}{\overset{q}{\mathrm{\Σ}}}\underset{j=0}{\overset{M-1}{\mathrm{\Σ}}}{s}_j{y}_i(n-j)$

Wherein, y _in () is served as reasons with the secondary source composite signal be formed by stacking, that is:

$y_i\left(n\right)=y_i^{\left(0\right)}\left(n\right)+y_i^{(-k)}\left(n\right).$

In sum, the present invention adopts technique scheme, has following beneficial effect:

(1) frequency-tracking and reference signal generation module make use of the frequency of adaptive notch filter tracking noise target, effectively also can produce the sinusoidal pattern reference signal relevant to target noise by tracking target noise frequency;

(2) the part secondary source that the delayed reference signal module increased and the second amplitude phase module generate, the part secondary source obtained through amplitude phase adjusted with reference signal to disappear the minimum residual noise of produce power through superposition, mutually with less calculated amount elevator system speed of convergence;

(3) introducing of reference delay signal module, amplitude phase adjustment block have impact on and adopts the parameter of the tracking of AR model frequency and reference signal generation module to upgrade, the frequency that whole system can be made to tackle not the amount of coupling up to more than 10%, and be easy in the active noise controlling device of reality realize;

(4) target noise suppressing method is by simple parallel-expansion, and the frequency being easy to realize under multi-frequency lane does not mate reply, reaches the object effectively suppressing narrow band noise.

Be understandable that; for those of ordinary skills; accompanying drawing is the schematic diagram of an embodiment; module in accompanying drawing or flow process might not be that enforcement the present invention is necessary; can be equal to according to technical scheme of the present invention and inventive concept thereof and replace or change, and all these change or replace the protection domain that all should belong to the claim appended by the present invention.

Claims (9)

1. arrowband feedforward active noise control system, it is characterized in that, comprise: frequency-tracking and reference signal generation module (1), delayed reference signal module (2), the first amplitude phase adjustment block (31), the second amplitude phase adjustment block (32), the first accumulator module (5), secondary channel (6), the second accumulator module (7), the first parameter adjustment module (41), the second parameter adjustment module (42), the 3rd parameter adjustment module (43), wherein

Frequency-tracking and reference signal generation module (1): output terminal is connected with delayed reference signal module (2) input end, the first amplitude phase adjustment block (31) input end, adopt the frequency of AR model tracking target noise under the synchronizing frequency of non-acoustic sensor collection and generate sinusoidal pattern reference signal, exporting sinusoidal pattern reference signal to delayed reference signal module (2), the first amplitude phase adjustment block (31);

Delayed reference signal module (2): output terminal is connected with the second amplitude phase module (32) input end, exports and postpones reference signal to the second amplitude phase adjustment block (32);

First accumulator module (5): input termination first amplitude phase adjustment block (31) output terminal, another input termination second amplitude phase adjustment block (32) output terminal, export termination secondary channel (6) input end, after receiving a part of secondary source signal that the first amplitude phase adjustment block (31) exports and another part secondary source signal that the second amplitude phase adjustment block (32) exports, export secondary source composite signal to secondary channel (6);

Second accumulator module (7): input termination secondary channel (6) output terminal, another input termination target noise signal, exports residual noise signal after the secondary noise signal that receiving target noise signal and secondary channel export;

First parameter adjustment module (41): the filtering signal of an input termination sinusoidal pattern reference signal, another input termination residual noise signal, exports the parameter updated value of the first amplitude phase adjustment block (31);

Second parameter adjustment module (42): an input termination postpones the filtering signal of reference signal, another input termination residual noise signal, exports the parameter updated value of the second amplitude phase adjustment block (32);

3rd parameter adjustment module (43): the filtering signal of an input termination sinusoidal pattern reference signal, another input termination residual noise signal, the auto-adaptive parameter updated value of output frequency tracking and reference signal generation module (1).

2. arrowband according to claim 1 active noise control system, is characterized in that, described first amplitude phase adjustment block (31) is the first FIR filter, and the second amplitude phase adjustment block (32) is the second FIR filter.

3. arrowband according to claim 2 active noise control system, it is characterized in that, described first parameter adjustment module (41), the second parameter adjustment module (42), the 3rd parameter adjustment module (43) all adopt LMS algorithm undated parameter.

4. arrowband as claimed in any of claims 1 to 3 active noise control system, it is characterized in that, described target noise signal is generated with superposing of environment random noise in a narrow band noise produced that disappears mutually through linear primary channel propagation by narrowband noise sources.

5. target noise suppressing method, adopts system according to claim 3 to realize, it is characterized in that, comprise the steps:

A. the acquisition of reference signal: frequency-tracking and reference signal generation module produce sinusoidal pattern reference signal under the synchronizing frequency of non-acoustic sensor collection, delayed reference signal resume module sinusoidal pattern reference signal obtains postponing reference signal;

B. the acquisition of secondary source composite signal: the first FIR filter aligns chordwise reference signal and carries out amplitude phase adjusted and obtain a part of secondary source signal, second FIR filter is carried out amplitude phase adjusted to delay reference signal and is obtained another part secondary source signal, and a part of secondary source signal, another part secondary source signal obtain secondary source composite signal after the first totalizer superposition;

C. suppress target noise signal: secondary source composite signal generates secondary noise signal through secondary channel, secondary noise signal and target noise signal superpose to disappear mutually obtain residual noise signal through the second superimposers;

D. undated parameter: adopt secondary channel estimation model process sinusoidal pattern reference signal, delay reference signal to obtain filtering-X reference signal, filtering-X postpones reference signal, first parameter adjustment module is according to the weights of filtering-X reference signal, residual noise signal update first FIR filter, second parameter adjustment module postpones the weights of reference signal, residual noise signal update second FIR filter according to filtering-X, and the 3rd parameter adjustment module is according to the auto-adaptive parameter of filtering-X reference signal, residual noise signal update frequency-tracking and reference signal generation module.

6. target noise suppressing method according to claim 5, is characterized in that, the sinusoidal pattern reference signal that steps A produces is:

x _i(n)＝-c _i(n)x _i(n-1)-x _i(n-2),n≥2

x _i(0)＝a _i

X _i(1)=a _icos (ω _i(0))+b _isin (ω _i(0)), wherein,

X _i(0), x _i(1), x _i(n-2), x _i(n-1), x _in () is respectively the 0th time, the 1st time, the n-th-2 times, (n-1)th time, n-th time updated value of sinusoidal pattern reference signal, c _in n-th updated value that () is auto-adaptive parameter, ω _i(0) be the synchronizing frequency that non-acoustic sensor records, auto-adaptive parameter initial value c _i(0) _:c _i(0)=-2cos (ω _i(0)).

7. target noise suppressing method according to claim 6, is characterized in that, a part of secondary source signal that step B obtains and another part secondary source signal be respectively:

$\begin{array}{c}{y}_i^{\left(0\right)}\left(n\right)=h_{i,0}^{\left(0\right)}\left(n\right)x_i\left(n\right)+h_{i,1}^{\left(0\right)}\left(n\right)x_i(n-1)\\ y_i^{(-k)}\left(n\right)=h_{i,0}^{(-k)}\left(n\right)x_i(n-k)+h_{i,1}^{(-k)}\left(n\right)x_i(n-k-1)\end{array},$ Wherein,

be n-th updated value of the first FIR filter weights, x _i(n-1), x _in (n-1)th time, n-th time updated value that () is sinusoidal pattern reference signal, be n-th updated value of the second FIR filter weights, x _i(n-k-1), x _i(n-k) for postponing (n-1)th time, n-th time updated value of reference signal.

8. target noise suppressing method according to claim 7, is characterized in that, target noise p (n) described in step C is:

$p\left(n\right)=\underset{i=1}{\overset{q}{\mathrm{\Σ}}}[a_{p,i}{x}_{a_i}\left(n\right)+b_{p,i}{x}_{b_i}\left(n\right)]+v_p\left(n\right),$ Wherein,

Q is target noise frequency component quantity, a _p,i, b _p,ifor discrete fourier coefficient, for cosine signal component, for sinusoidal signal component, ω _p,ifor target noise signal frequency, v _pn () is environment random noise.

9. target noise suppressing method according to claim 8, is characterized in that, in step D, the renewal equation of the first FIR filter weights is: $\begin{array}{c}{h}_{i,0}^{\left(0\right)}(n+1)=h_{i,0}^{\left(0\right)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i\left(n\right)\\ h_{i,1}^{\left(0\right)}(n+1)=h_{i,1}^{\left(0\right)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i(n-1)\end{array},$ The renewal equation of the second FIR filter weights is: $\begin{array}{c}{h}_{i,0}^{(-k)}(n+1)=h_{i,0}^{(-k)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i(n-k)\\ h_{i,1}^{(-k)}(n+1)=h_{i,1}^{(-k)}\left(n\right)+{\mathrm{\μ}}_{i}e\left(n\right){\hat{x}}_i(n-k-1)\end{array},$ The renewal equation of frequency-tracking and reference signal generation module auto-adaptive parameter is: wherein,

be (n+1)th updated value of the first FIR filter weights, be (n+1)th updated value of the second FIR filter weights, for (n-1)th time, n-th time updated value of filtering-X reference signal, for filtering-X postpones (n-1)th time, n-th time updated value of reference signal, μ _ifor sinusoidal pattern reference signal and the renewal coefficient postponing reference signal, for the renewal coefficient of auto-adaptive parameter.