JP3537150B2 - Noise control device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a noise control device for canceling noise by outputting a signal having the same sound pressure in opposite phase to noise from a speaker. In particular, the present invention improves the noise canceling characteristics in response to a change in noise frequency. The purpose is to let them.

[0002]

2. Description of the Related Art Conventionally, a passive muffler such as a muffler has been used to reduce noise generated from an internal combustion engine or the like, but improvements have been made in view of size, muffling characteristics, and the like. On the other hand, the noise generated from the sound source
There has been proposed an active noise control device that outputs a compensation sound of equal sound pressure from a speaker and cancels noise. By the way, the frequency characteristics or the stability of the active noise control device itself are not sufficient, and its practical use has been delayed. However, in recent years, as a signal processing technique using a digital circuit has been developed and a frequency range to be handled has been expanded, a number of practical noise control devices have been proposed (for example, JP-A-63-31396).

[0003] The noise is detected by a noise source microphone installed upstream of the duct, and a signal having a reverse phase and equal sound pressure to the noise is output from a speaker installed downstream of the duct by a signal processing circuit. This is a so-called 2-microphone / 1-speaker type active noise control device that is combined with a feedback system that detects and feeds back a sound using a microphone for a silencing point.

[0004]

By the way, the noise control device employs a signal processing technique using a digital circuit.
SP (Digital Signal Processor) is used and DSP
Is configured with an adaptive filter. However, when the frequency of the noise changes variously, there is a problem that the scale of the adaptive filter becomes large and the convergence time becomes long.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a noise control device capable of improving the erasing characteristics even when the frequency of noise changes, in view of the above problems.

[0006]

The present invention solves the above problems by inputting a noise signal from a noise source (1),
A speaker (4) that outputs a sound wave having the same phase and opposite sound pressure as the noise
An adaptive filter, a filter coefficient updating means, and a noise control device having a microphone (8) for detecting an error signal generated by eliminating noise by the speaker (4).
A transfer characteristic simulation means, a frequency detection means, and a variable bandpass filter are provided.

[0007] The adaptive filter automatically adjusts the filter coefficients to form a compensating signal of opposite phase equal sound pressure. The filter coefficient updating means forms the filter coefficient updated by the noise signal and the error signal. The transfer characteristic simulating means corrects an input noise signal of the filter coefficient updating means by a characteristic simulating a transfer characteristic from the adaptive filter to forming the error signal.

The frequency detecting means detects a noise frequency distribution from the noise source. The variable bandpass filter has a frequency distribution detected by the frequency detection unit,
A pass frequency band of the noise signal is arbitrarily set, and an input signal of the adaptive filter and the transfer characteristic simulating means is controlled.

[0009]

According to the noise control device of the present invention, the filter coefficient is automatically adjusted by the adaptive filter, and a compensation signal of opposite phase equal sound pressure is formed. The filter coefficient updated by the filter coefficient updating means based on the noise signal and the error signal is formed. The transfer noise simulating means corrects the input noise signal of the filter coefficient updating means by a characteristic simulating a transfer characteristic from the formation of the error signal by the adaptive filter. A noise frequency distribution from the noise source is detected by the frequency detecting means. The pass band of the noise signal is arbitrarily set by the variable bandpass filter according to the frequency distribution detected by the frequency detection unit, and the input signals of the adaptive filter and the transfer characteristic simulation unit are controlled. Therefore, it becomes possible to pass the noise signal only to those in the desired frequency band, and the erasing characteristics can be improved.
Desired erasure characteristics can be obtained, and an increase in the scale of the adaptive filter can be suppressed.

[0010]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing a noise control device which is a premise of the first embodiment of the present invention. The noise control device of this figure physically silences noise from a noise source 1 of an engine of an automobile or the like, and removes residual noise in a silencing space 3 near a tail pipe at an exit from a muffler 2 having a constant transmission characteristic. A speaker 4 installed to mute the sound, a power amplifier 5 for driving the speaker 4, and a low-pass filter 6 installed in front of the power amplifier 5 to remove high-frequency components of an analog signal
And a D / A converter 7 (Digital to Analyze) for converting the digital signal into an analog signal for the low-pass filter 6.
og Converter), a microphone 8 installed near the speaker 4 for detecting an error signal, an amplifier 9 for amplifying an electric signal of the microphone 8, and a low-frequency component for removing a high-frequency component of the amplified signal of the amplifier 9. Pass filter 10
And an A / D converter 11 (Analog to Digita) for converting an analog signal of the low-pass filter 10 into a digital signal.
lConverter) and a signal such as the number of revolutions of the engine as a noise source signal of the noise source 1, and the A / D converter 11
From the D / A so that the error signal at the silence point becomes minimum
A digital signal processor 12 (Digital Signal Processor) for forming a compensation signal for eliminating noise in the converter 6.
r), a RAM 13 (Random Access Memory) for storing a program of the digital signal processing device 12,
A central processing unit 14 for transferring a program from the ROM 15 (Read Only Memory) to the M13 and temporarily storing the program in the RAM 16 for this purpose is provided.

FIG. 2 is a diagram showing the configuration of the digital signal processing device of FIG. The digital signal processing device shown in FIG. 1 receives the number of revolutions of the engine, which is the noise source 1, and detects the frequency of the noise. Variable bandpass filter 12 that makes frequency band variable according to information
2, an adaptive filter 123 for inputting a signal of the frequency selected by the variable bandpass filter 122 and automatically updating the filter coefficient to form a noise compensation signal; Filter coefficient updating means 124 for forming a filter coefficient of the adaptive filter 123 from an input signal to the adaptive filter 123, and means for simulating a transfer characteristic of a signal from the output of the adaptive filter 123 to the filter coefficient updating means 124. And a transfer characteristic simulating means 125 for inputting an input signal to the adaptive filter 123 to form a simulation signal and supplying the simulation signal to the filter coefficient updating means 124. Here, the transfer characteristic from the adaptive filter 123 to the microphone 8 is Hd,
Assuming that the transfer characteristic from the microphone 8 to the filter coefficient updating means 124 is Hm, the simulated transfer characteristic Hd1 of the transfer characteristic simulating means 125 is as follows: Hd1 = Hd · Hm (1)

FIG . 3 is a diagram showing the configuration of the frequency detecting means 121 of FIG. The frequency detecting means 121 shown in FIG.
The engine speed signal from the noise source 1, that is, the noise source signal S
A plurality of bandpass filters 1211-1 and 121 for inputting r
, 1211-n, and a plurality of level forming means 1212-1, 1212-2,..., 12 connected to the plurality of bandpass filters 1211 and averaging their output levels.
212-n and a maximum frequency band detecting means 1213 for comparing the level from the level forming means 1212 and detecting the frequency band of the maximum level. Each of the plurality of band filters 1211 may be constituted by a secondary digital filter described later.

FIG. 4 is a diagram showing the configuration of the variable bandpass filter of FIG. The variable bandpass filter 122 shown in the figure is a secondary digital filter, which is a delay means 1221 for delaying an input signal by the time of one sampling cycle to output a primary delay signal, and further delaying the output to obtain a second order. Delay means 1222 for outputting the next delay signal, and delay means 1 for delaying the output signal and outputting it as a primary feedback signal
223, delay means 1224 for further delaying and outputting as a secondary feedback signal, multiplication means 1225 for multiplying the input signal by a predetermined coefficient a0 and outputting the result, and a predetermined coefficient a1 for the primary delayed signal. Multiplying means 1226 for multiplying and outputting a multiplied signal; multiplying means 1227 for multiplying and outputting a secondary delayed signal by a predetermined coefficient a2; multiplying means for multiplying and outputting a primary feedback signal by a predetermined coefficient b1 Multiplying means 122 for multiplying the secondary feedback signal by a predetermined coefficient b2 and outputting the result
9, multiplication means 1225, 1226, 1227, 122
8 and 1229 are added to generate an output signal. Coefficients a0 and a1 of the above multiplication means
, A2, b1, and b2 are stored in the RAM 13 in frequency bands of various noises, and a constant of a predetermined frequency band is set in the multiplication means according to the frequency distribution detected by the frequency detection means 121.

FIG. 5 is a diagram showing the configuration of the adaptive filter of FIG. The adaptive filter 123 shown in the figure is a non-recursive FIR (Finite Impulse Response), and receives a signal from the variable bandpass filter 122 and sequentially outputs a plurality of delay units 1231-1 and 1231 with a delay time τ. -2,
, 1231- (k-1), the input signal, and the output from each of the delay means 1231, filter coefficients C0 (n) and C
Multiplying means 1232 for multiplying 1 (n), C2 (n),..., Ck (n)
-1,1232-2,1232-3, ..., 1232-k
And the output of each multiplying means 1232 to add a compensation signal Sc
, The addition means 1233-1, 1233--2,.
1233- (k-1).

FIG. 6 shows the filter coefficient updating means (LS
FIG. 3M is a diagram showing the configuration of FIG. The filter coefficient updating means 124 shown in FIG.
(n) is multiplied by a constant α, and delay means 1242-1, 1242-2,..., 1 are used to sequentially delay the signal from the transfer characteristic simulating means 125 by a delay time τ.
242- (k-1) and normalization means 1243-1 and 1243- which divide the output signal of the multiplication means 1241 by the input signal from the transfer characteristic simulation means 125 and the output signal of each delay means 1242 to normalize them. 2, ..., 1243- (k-
1) is connected to the output of each of the normalizations and is added to a signal to be described later to obtain filter coefficients C0 (n), C1 (n), C2 (n),.
.., 1244-k which form Ck (n) and output to each of the multiplying means 1232
Delay means 1245 for delaying the output of each of said adding means 1244 by a delay time τ and adding to said adding means 1244
-1, 1245-2, ..., 1245-k.

With the above configuration, a filter coefficient for the multiplying means 1232 is formed as follows. In FIG. 5, if the input data to the adaptive filter 123 is q (n), the output data of each delay means 1231 is q (n−1), q
(n−2), q (n−3),..., q (n−k + 1). Here, assuming that the filter coefficients set in the respective multiplying means 1231 are C0 (n), C1 (n), C2 (n), C3 (n),..., Ck (n) as described above. Output data Sc of the adaptive filter 151
(n) is as follows.

Sc (n) = C0 (n) · q (n) + C1 (n) · q (n−1) + C2 (n) · q (n−2) + C3 (n) · q (n−3) +,..., + Ck (n) · q (n−k + 1) (2) Next, in FIG.
241 is set as Sm (n), and the multiplication means 1
The coefficient update constant as the other data to the H.241 is α,
The input data from the transfer characteristic simulation means 125 to the delay means 1242 is set to q (n) in the same manner as described above, and each delay means 1242
Output data of q (n-1), q (n-2), q (n-3),.
(n-k + 1). For this reason, the output data from the adding means 1241 is added to each of the delay means 124 by each of the normalizing means 1243.
The signal is normalized by the signal from 2 and is formed into the following filter coefficient by each adding means 1244 and delay means 1245.

Ck (n) = Ck (n−1) + (Sm (n) · α) / q (n + k−1), (k = 1 to k) (3) Thus, according to the present embodiment, For example, the frequency of the noise is specified by detecting the engine speed by the frequency detecting means 121, and only the noise of the specified frequency passes through the variable bandpass filter 122. It is not necessary to process signals of all frequencies, that is, only the frequency band having a large influence needs to be erased, so that the processing load is reduced and the erasing characteristics can be improved.

FIG. 7 is a diagram showing a configuration of a digital signal processing device according to a second embodiment of the present invention. The digital signal processing device 12 shown in this figure is obtained by changing the variable bandpass filter 122 shown in FIG. The variable bandpass filter 131 includes a plurality of bandpass filters 132-1 and 13-3 for inputting an engine speed Sr.
2-2,..., 132-n and the frequency detecting means 121
Multiplication means 1 for making the output level of each bandpass filter 132 variable according to the result of
, 133-n and the output of each of the multiplying means 133, and the result is added to the adaptive filter 1
23, adding means 13 for outputting to transfer characteristic simulating means 125
4 is included.

FIG. 8 is a diagram showing characteristics of the bandpass filter according to the present embodiment. FIG. 3A shows a frequency spectrum of a signal input to each band filter. On the other hand, FIG. 2B shows the gain of each band filter with respect to the frequency. The frequency detecting means 121 detects the frequency fm of the maximum peak from the drawing (a), and selects the bandpass filter 132-1 (BPF1) corresponding to the drawing (b). Is set to “1” and the other multiplication means are set to “0”.

Although the detection of the frequency of the maximum peak has been described above for the sake of simplicity, it may be set so that only the desired detection frequency band is passed. Figure
FIG. 9 is a diagram showing a configuration of a digital signal processing device according to a third embodiment of the present invention. In the digital signal processing device 12 shown in FIG. 2, the variable band filter 122 is provided before the input of the adaptive filter 123 in the first embodiment of FIG.
Is provided at the position of the input stage from the A / D converter 11 of the filter coefficient updating means 124 and at the position of the input stage from the transfer characteristic simulation means 125. And these two variable bandpass filters 12
2, 122 'are controlled by the frequency detecting means 121 in the same manner as described above.

According to the above equation (3), the filter coefficient updating means 124 uses the engine speed as an input signal for normalizing the error signal. Further, it is conceivable that the error signal detected by the microphone 8 also has a component of the maximum frequency of noise remaining. For this reason, the filter coefficient updating unit 124 also extracts the component of the maximum frequency of the noise by the variable bandpass filter 122, configures the filter coefficient with the input signal and the error signal, and can most respond to the maximum frequency. Therefore, the time required to converge to the optimum point of the filter coefficient can be reduced.

FIG. 10 is a diagram showing a digital signal processing device 12 according to a fourth embodiment of the present invention. As a modification of the third embodiment, the digital signal processing device 12 shown in this figure is the same as that of the third embodiment except that the variable bandpass filter 122 is replaced with the variable bandpass filter 131 shown in FIG. The effect is obtained. The above is the signal processing system of the feedforward system. Next, the configuration of the feedback system will be described.

FIG. 11 is a diagram showing the configuration of a digital signal processing apparatus according to a fifth embodiment of the present invention. The digital signal processing device 12 shown in FIG.
A transfer characteristic simulating means 125 for forming a transfer simulated characteristic signal from an input signal of the adaptive filter 123; and an adaptive filter based on a signal from the A / D converter 11 and a signal from the first transfer characteristic simulated means 125. A filter coefficient updating unit 124 for forming a filter coefficient of the filter 123;
First transfer characteristic simulation means 125 provided on the output side
And a difference between the signal from the second transfer characteristic simulating means 126 and the signal from the A / D converter 11 to reproduce the noise signal from the feedback signal. A difference signal calculating means 127, a variable bandpass filter 122 provided between the difference signal calculating means 127 and the adaptive filter 123, and a frequency detecting means 121 for detecting a maximum frequency of noise based on a signal of an engine speed.
And

Here, assuming that the transfer characteristic from the noise source 1 to the silencing space 3 is Hnoise, the signal Sm0 detected by the microphone 8 in the silencing space 3 is Sm0 = Sn · Hnoise + Sc · Hd ( 4) Here, Sn is the noise signal of the noise source. The input signal Se of the adaptive filter 123 is as follows: Se = Sm0.Hm-Sc.Hd1 = (Sn.Hnoise + Sc.Hd) .Hm-Sc.Hd.Hm (see equation (3)) = Sn.Hnoise.Hm + Sc Hd.Hm-Sc.Hd.Hm = Sn.Hnoise.Hm, and a signal similar to the noise signal detected by the microphone 8 is obtained.

This signal Se is, as in the first embodiment,
The signal is processed by the frequency detecting means 121 and the variable band-pass filter 122, and is subjected to, for example, a processing for maximizing the noise frequency. Note that the feedforward system of the second, third, and fourth embodiments may be a feedback system as described above.

FIG. 12 is a diagram showing the configuration of a digital signal processing device 12 according to a sixth embodiment of the present invention. The digital signal processing device 12 shown in the figure is a combination of a feedforward system and a feedback system.
Is different from the first configuration in that the first engine speed is input.
And a second variable bandpass filter 129 for inputting the reproduction signal from the difference signal calculating means 127.
Addition means 130 for adding the output signals of the first and second variable bandpass filters 128 and 129 and outputting the sum to the adaptive filter 123;
And a frequency detecting means 121 for controlling the frequency bands 129 and 129 to have complementary characteristics.

First and second variable bandpass filters 128 and 1
Numeral 29 may be such as to make the coefficient of the multiplying means variable as shown in FIG. 4, or further to make the multiplying coefficient of the variable multiplying means of plural band filters variable as shown in FIG. FIG. 13 is a diagram illustrating characteristics of the variable bandpass filter according to the sixth embodiment. FIG. 6A shows the frequency detecting means 12.
1 is a spectrum to be input to FIG. In this case, assuming that a peak is generated at the frequency f1 as shown in the figure, the frequency detecting means 121 detects this and changes the gain of the first variable bandpass filter 128 as shown in the figure (b). The input signal of the engine speed is adjusted so as to pass only near the frequency f1. On the other hand, the gain of the second variable frequency bandpass filter 129 is adjusted so that only the frequency of the signal of the difference signal calculating means 127 is not passed as shown in FIG.

For this reason, the fundamental wave passes the input of the feedforward system through the first variable bandpass filter 128,
Other frequencies, that is, harmonics, are passed through the input of the feedback system by the second variable frequency bandpass filter 129 and input to the adaptive filter 123. Therefore, even when a noise frequency other than the fundamental wave is generated in the muffler 2, a noise reduction effect can be obtained.

[0030]

As described above, according to the present invention, the pass frequency band of the noise signal is arbitrarily selected according to the detected frequency distribution, and only the selected signal is processed by the adaptive filter. The erasing characteristics can be improved, and desired erasing characteristics can be obtained.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a noise control device that is a premise of a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a configuration of the digital signal processing device of FIG. 1;

FIG. 3 is a diagram illustrating a configuration of a frequency detection unit 121 in FIG. 2;

FIG. 4 is a diagram illustrating a configuration of a variable bandpass filter of FIG. 2;

FIG. 5 is a diagram showing the configuration of an adaptive filter.

FIG. 6 is a diagram showing a configuration of a filter coefficient updating means (LSM) of FIG. 2;

FIG. 7 is a diagram showing a configuration of a digital signal processing device according to a second embodiment of the present invention.

FIG. 8 is a diagram illustrating characteristics of the bandpass filter according to the embodiment.

FIG. 9 is a diagram showing a configuration of a digital signal processing device according to a third embodiment of the present invention.

FIG. 10 is a diagram showing a digital signal processing device 12 according to a fourth embodiment of the present invention.

FIG. 11 is a diagram showing a configuration of a digital signal processing device according to a fifth embodiment of the present invention.

FIG. 12 is a diagram showing a configuration of a digital signal processing device 12 according to a sixth embodiment of the present invention.

FIG. 13 is a diagram illustrating characteristics of a variable bandpass filter according to a sixth embodiment.

[Explanation of symbols]

1. Noise source 4: Speaker 8 ... Microphone 12 ... Digital signal processing device 121 ... frequency detecting means 122, 128, 129, 131 ... variable bandpass filters 123 ... Adaptive filter 124 ... Filter coefficient updating means 125, 126 ... Transfer characteristic simulation means 127 ... difference signal calculating means 130 addition means

──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Kazuhiro Sakiyama 1-2-28, Goshodori, Hyogo-ku, Kobe-shi, Hyogo Prefecture Inside Fujitsu Ten Co., Ltd. (56) References JP-A-1-117978 (JP, A) JP-A-1-314500 (JP, A) JP-A-2-241215 (JP, A) JP-A-4-359297 (JP, A) JP-A-5-11770 (JP, A) JP-A-5-209563 ( JP, A) JP-A-5-273787 (JP, A) JP-A-5-289679 (JP, A) JP-A-5-307393 (JP, A) JP-A-5-333871 (JP, A) JP-A-5-39710 (JP, A) JP-A-6-12083 (JP, A) WO 91/013429 (WO, A1) WO 91/12579 (WO, A1) (58) Fields investigated (Int. Cl) . ^7, DB name) G10K 11/178 F01N 1/00 F01N 1/06 H03H 21/00

Claims (3)

(57) [Claims] And 1. A speaker for outputting the sound waves of the inverse phase equal sound pressure with respect to the noise from the noise source (1) (4), error of the sound waves from the noise and the speaker from該騒sound source (4) A noise control device having a microphone (8) for detecting a signal, comprising: a noise source signal (Sr) output from the noise source (1) as an input; and a frequency detection unit (12) for detecting a frequency distribution thereof.
1), the noise source signal Sr is supplied, an adaptive filter (123) for adjusting a filter coefficient and generating a compensation signal for outputting the opposite phase equal sound pressure to a speaker, and the noise source signal Sr is supplied. The output of the adaptive filter (123) is used to update the filter coefficient updating means (12).
(4) a transfer characteristic simulating means (125) for simulating a transfer characteristic of a signal up to an input to the input, and the transfer characteristic simulating means using the detected frequency distribution supplied from the frequency detecting means (121). a first variable band-pass filter for setting the pass frequency band of the signal supplied from the (125) (122), using the said detected frequency distribution supplied from the frequency detecting means (121), said error signal (Sm)
The second variable bandpass filter (1
22 ′) and the filter coefficient of the adaptive filter (123) based on the signal supplied from the first variable bandpass filter ( 122 ) and the signal supplied from the second variable bandpass filter ( 122 ′ ). And a filter coefficient updating means (124) for supplying a signal representing the filter coefficient to the adaptive filter (123). 2. The two variable filters (122, 12).
2 ′) are transfer characteristic simulation means (125), respectively.
In order to control the signal and the error signal supplied from the
Bandpass filter (132-1, 132- 2, ... 132 -n) and was detected by the frequency detecting means (121) frequency
According to the number distribution, each of the plurality of filters ( 132-1 , 132-2 ,
.. 132-n ) to adjust the output level.
3) and adding means for adding the outputs of the variable multiplying means (133).
(134) and (134).
Noise control means. 3. A loudspeaker (4) for outputting a sound wave having an opposite phase equal sound pressure to noise from a noise source (1), and a difference signal between the noise from the noise source and the sound wave from the loudspeaker (4). And a microphone (8) for detecting the noise source signal, a noise source signal Sr output from the noise source (1) being input, and a frequency detecting means (12) for detecting a frequency distribution thereof.
1) and an adaptive filter (123) for adjusting a filter coefficient and generating a compensation signal for outputting an opposite-phase equal sound pressure to a speaker.
A filter coefficient updating unit (124) for supplying a signal representing the filter coefficient to the adaptive filter (123); and a signal from an output of the adaptive filter (123) to an input to the filter coefficient updating unit. A first transfer characteristic simulating means (125) for simulating transfer characteristics, and an output signal of the adaptive filter (123) as an input;
A second transfer characteristic having the same characteristics as the first transfer characteristic simulation means;
A transfer characteristic simulating means (126); a subtraction means (127) for calculating a difference between an output signal of the second transfer characteristic simulating means (126) and an error signal (Sm) obtained by digitally converting the difference signal; A first variable bandpass filter (128) that receives the noise source signal Sr and sets a pass frequency band of the noise source signal Sr using a frequency distribution detected by the frequency detection means (121); A second variable bandpass filter (129) to which an output signal of the means (127) is supplied and which sets a pass frequency band of an output signal of the subtraction means using a frequency distribution detected by the frequency detection means (121); The outputs of the first variable bandpass filter (128) and the second variable bandpass filter (129) are added, and the output is added to the adaptive filter 123 and the simulation unit (1). 25), the filter coefficient updating means (124) based on the output of the simulating means (125) and the error signal (Sm). 123) A noise control device characterized by determining the filter coefficient.