JPH08234767A - Noise eliminating device - Google Patents

Abstract

PURPOSE: To move a position to be noise-eliminated optionally and to contribute also to the comfort of housing sound environment. CONSTITUTION: A sensor microphone detects a noise coming into a house to output a noise signal. Plural error microphones 51 , 52 respectively detect residual noises that the coming noise is canceled by interference with a control sound radiated from a loud speaker 1 to be remained to output residual noise signals. Amplifiers 131 , 132 are constituted so as to be set and adjusted to prescribed amplification degrees individually independently, and they amplify the residual noise signals outputted from the corresponding error microphones 51 , 52 at respectively independently set amplification degrees. A space area to be noise- eliminated is specified from a ratio between voltage effective values of plural residual noise signals after amplification. A resident can move the space area to be noise-eliminated to another place by individually independently setting and adjusting the amplification degrees of the amplifiers 131 , 132 at need.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a muffling device, and more particularly to a muffling device for actively muffling noise arriving in a living space or the like.

[0002]

2. Description of the Related Art Conventionally, as this type of silencer, for example, one disclosed in Japanese Patent Application Laid-Open No. 5-181487 is known. As shown schematically in FIG. 7, this conventional device uses a loudspeaker 1 installed in a passenger room to generate noise (primary sound) such as an engine sound that arrives in a passenger room of an aircraft or an automobile.
Therefore, by radiating a control sound (secondary sound) having the same amplitude and opposite phase as the above noise toward the seat, the noise is canceled by interference.

A sensor microphone 2 as a noise detecting means is attached to the engine side in the passenger compartment, and the sensor microphone 2 detects the sound pressure level of the engine sound and converts it into a noise signal SM. This noise signal SM is input to the controller 3. On the other hand, at the left and right ears of the passenger who sits down on the seat 4, the sound pressure level of noise (residual noise) that is reduced by being interfered by the control sound is detected to detect the residual noise signal E.
Error microphones 5 ₁ and 5 ₂ for converting into M1 and EM2 are provided. The residual noise signals EM1 and EM2 are input to the controller 3. As shown in FIG. 8, the controller 3 includes a signal processing unit 31 that performs digital signal processing, and amplifiers 32 ₁ , 32 ₂ , 32 ₃ , 32 ₄ , which form peripheral circuits,
Low-pass filter (hereinafter referred to as LPF) 33 ₁ , 3
3 ₂ , 33 ₃ , 33 ₄ , analog / digital converter (hereinafter referred to as A / D converter) 34 ₁ , 34 ₂ , 34 ₃ ,
A noise signal SM which is composed of a digital / analog converter (hereinafter referred to as a D / A converter) 35 and is inputted.
A control sound signal for driving the loudspeaker 1 using an adaptive control algorithm including a transfer function between the loudspeaker (control sound source) 1 and the error microphones 5 ₁ and 5 ₂ based on the residual noise signals EM1 and EM2. By generating the SP, adaptive control is always performed so that the noise level at the positions of the error microphones 5 ₁ and 5 ₂ (that is, the left and right ears of the passenger) is minimized.

[0004]

However, in the fixed circuit configuration as described above, the silenced space is also fixed. In the case of passenger cabins such as airplanes and automobiles, it is sufficient to muffle around the seats, so there is no problem even if the muffled space is fixed. However, if this type of silencer is to be applied to improve the sound environment in a house, the position of the resident will not be determined daily even if one room is taken in the house. As it is, it will be a bad thing to use. The present invention has been made in view of the above circumstances, and an object thereof is to provide a muffling device that can contribute to a comfortable living sound environment by varying the position to mute.

[0005]

In order to solve the above-mentioned problems, a silencer according to the invention of claim 1 is a noise detecting means for detecting an incoming noise and outputting a noise signal, and an input control. Based on the sound signal, the control sound emitting means for emitting a control sound to a predetermined space area to be silenced, and the control sound emitting means are provided so as to sandwich or surround the predetermined space area to be silenced, and the noise is A plurality of residual noise detecting means for respectively detecting residual noises remaining by being canceled by interference with the control sound emitted from the control sound emitting means and outputting a residual noise signal, and independently set to a predetermined amplification degree A plurality of residual noise signal amplifying means that are adjustable and that amplify the residual noise signals output from the corresponding residual noise detecting means, respectively, with independently set and adjusted amplification levels; and a plurality of residual noise signals. Based on the residual noise signal mixing means for mixing the plurality of amplified residual noise signals output from the signal amplification means, and the plurality of residual noise signals after amplification and mixing based on the noise signal and the residual noise signal after amplification and mixing. Control for emitting a control sound having a phase opposite to that of the waveform component of the noise so that the residual noise is always minimized with respect to the predetermined spatial region to be silenced, which is determined from the ratio of the effective value of the residual noise signal. And a control means for generating a sound signal and supplying the sound signal to the control sound emitting means.

[0006]

In the structure of the present invention, the noise detecting means detects, for example, noise arriving in the house and outputs a noise signal. Further, the plurality of residual noise detecting means respectively detect residual noise that remains due to interference of the incoming noise with the control sound emitted from the control sound emitting means, and output a residual noise signal. The plurality of residual noise signal amplifying means are independently set and adjusted to a predetermined amplification degree, and amplify the residual noise signals output from the corresponding residual noise detecting means by the independently set and adjusted amplification degrees. .
The residual noise signal mixing means mixes the amplified residual noise signals output from the residual noise signal amplifying means. The spatial region to be silenced is specified from the ratio of the effective values of the plurality of residual noise signals after amplification and before mixing. As a result, the control means, based on the noise signal and the residual noise signal after amplification and mixing, has a control sound having a phase opposite to the waveform component of the noise so that the residual noise is always minimized with respect to the spatial region to be silenced. Is generated and supplied to the control sound emitting means. The control sound emitting means emits the control sound to a predetermined spatial region to be silenced based on the input control sound signal. If necessary, the occupant can adjust the amplification level of the residual noise signal amplifying means independently and adjust the spatial area to be silenced, so that the occupant area is surrounded by or surrounded by a plurality of residual noise detecting means. Within the range (for example, in the case where the residual noise detecting means is installed on the wall surface of the four circumferences, substantially the entire range of the room), it can be moved to another place.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic diagram showing an overall configuration of a silencer according to an embodiment of the present invention, FIG. 2 is a block diagram showing an electrical configuration of a controller used in the silencer, and FIG. 3 is incorporated in the controller. FIG. 4 is a block diagram showing the electrical configuration of the signal processing unit, and FIG. 4 is a block line showing the electrical configuration at the time of initial setting (correction filter identification) of the signal processing unit implemented by software (Filtered-x LMS algorithm). Figures and 5 show software (Filter
ed-x LMS algorithm) block diagram showing the electrical configuration of the signal processing unit at the time of muffling operation, and
FIG. 6 is an explanatory diagram for explaining how to set a spatial region to be silenced by the muffling device. In this embodiment, the frequency of noise to be silenced is set to 500 Hz or less for convenience of explanation.

The muffler of this example is used in a living sound environment. As shown in FIG. 1, a sensor microphone 2 as a noise detecting means is attached to a window canopy 6 of a house. The microphone 2 mainly detects the sound pressure level of external noise (primary sound) entering through the window opening 7 and converts it into a noise signal SM. This noise signal SM is input to the controller 8. On the other hand, the loudspeaker 1 serving as a control sound source has a partition wall 9 that defines an arbitrary room in a house.
And as a residual noise detecting means.
The error microphones 5 ₁ and 5 ₂ are arranged in pairs to be separated from each other by a predetermined space (a distance sufficiently longer than the distance between both ears of a person). The loudspeaker 1 is a control sound (secondary sound) that should cancel noise due to sound wave interference at an arbitrary spatial position (described later) S1 or S2 that connects _one error microphone 5 ₁ and the other error microphone 5 ₂ with a straight line. Radiates. The loudspeaker 1 is connected to the controller 8. Further, the error microphones 5 ₁ and 5 ₂ detect the sound pressure level of the external noise (residual noise) that is reduced by the interference with the control sound and convert it into residual noise signals EM1 and EM2. The residual noise signals EM1 and EM2 are input to the controller 8.

Next, the electrical configuration of the controller 8 will be described with reference to FIG. In the figure, the noise signal SM output from the sensor microphone 2 is input to the amplifier 9. The amplifier 9 amplifies the noise signal SM. LPF
10 has a cutoff frequency of 500 Hz and is configured by an anti-aliasing filter, and passes the low frequency component of the output signal of the amplifier 9. The A / D converter 11 uses an LPF with a sampling frequency of 1.5 kHz.
The low frequency component of the noise signal SM output from 10 is converted into a digital noise signal and then input to the signal processing unit 12.

Residual noise signals EM1 and EM2 output from the error microphones 5 ₁ and 5 ₂ are output from the corresponding amplifier 1
Input to 3 ₁ , 13 ₂ . Each amplifier 13 ₁ (13 ₂ )
The gain can be freely adjusted by the operator (resident) independently of the other amplifier 13 ₂ (13 ₁ ).
The residual noise signal EM1 (EM2) is amplified with the adjusted gain. The mixer 14 mixes the amplified residual noise signals EM1 and EM2 of the two systems to obtain a mixed residual noise signal (EM
Output as 1 + EM2). The LPF 15 has a cutoff frequency of 500 Hz and passes the low frequency component of the output signal (EM1 + EM2) of the mixer 14. A / D
The converter 16 outputs the mixed residual noise signal (EM1 +) output from the LPF 15 at a sampling frequency of 1.5 kHz.
EM2) low-frequency component of digital residual noise signal ε
After conversion into (i), the signal is input to the signal processing unit 12.

The D / A converter 17 converts the digital signal output from the signal processing unit 12 into an analog signal at a sampling frequency of 1.5 kHz. Also, LPF18
Has a cut-off frequency of 500 Hz and is configured by an anti-aliasing filter, and passes the low frequency component of the output signal of the D / A converter 17. The amplifier 19 amplifies the output signal of the LPF 18 and then inputs it to the loudspeaker 1.

Further, the effective value ratio calculation circuit 20 includes an amplifier 1
Three₁, 13₂The output signals of
Calculate the voltage ratio. For example, as shown in FIG.
Bowl 13₁The effective value of the output voltage of the₂Output
When the effective value of the voltage is a, a / b is calculated. This
The calculation result of is input to the signal processing unit 12, and
It is displayed on the indicator 21. The display 21 is operated by the operator (resident
Each person) 13 ₁(13₂) Depending on the gain adjustment
The ratio of the effective voltage that fluctuates with
The spatial regions S1 and S2 to be silenced specified by the pressure ratio
It also has a function to register settings in conjunction with operating means (not shown).
There is.

Here, it is important to note that the amplifier 13
By adjusting the gains of ₁ and 13 ₂ , it is possible to arbitrarily set the space to be silenced. Depending on the setting method, in a room with few reflections, as shown in FIG.
The ratio a / b of the effective value of the output voltage of 3 ₁ , 13 ₂ and the spatial position where the sound is muted (the spatial position where noise is reduced to 10 dB or more)
The inventor of the present application found through experiments that a: b = Y: X approximately holds. Note that a
Is the effective value of the output voltage of the amplifier 13 ₂ , b is the amplifier 1
3 effective value of the _first output voltage, X is spatial position S1 that is muted from the error microphone 5 ₂ amplifier 13 ₂ is connected (S
The distance up to 2), Y, indicates the distance from the error microphone 5 ₁ to which the amplifier 13 ₁ is connected to the spatial position S1 (S2) where the sound is muted. X + Y is two error microphones 5
_1, 5 is the distance between the _two, the space is muted position S1 (S
2) is on a straight line connecting the error microphone 5 ₁ and the error microphone 5 ₂ . Therefore, while the operator (resident) is watching the RMS value ratio of the amplifier output voltage calculated by the RMS value calculating circuit 20 and displayed on the display 21, each amplifier 13 ₁ , 1
By adjusting the 3 _second gain, two error microphones 5 _1, 5 restriction that between _2, but which the spatial position S1, S2 to be muted may be set to a desired position. The spatial position to be silenced specified in this way can be considered as the center of gravity of the error microphones 5 ₁ and 5 ₂ weighted by the reciprocal of the effective value ratio. This is also called the center of gravity of the error microphones 5 ₁ and 5 ₂ .

As shown in FIG. 3, the signal processing unit 12 controls the timing of an arithmetic processor (digital signal processor) 121 for performing digital signal processing, a storage device 122 constituted by an external RAM, and peripheral circuits. It is composed of an input / output module 123 for adjusting. In the storage device 122, the spatial regions S1 and S2 to be silenced specified from the effective value ratio of the output voltages of the amplifiers 13 ₁ and 13 ₂ are stored in association with the filter coefficients of the correction filters 83 and 84 described later. To be done.

Next, with reference to FIGS. 4 and 5, an electrical configuration of the signal processing unit 12 realized by software (Filtered-x LMS algorithm) will be described. In the figure, the signal processing unit 12 includes various types of adaptive digital filters (hereinafter referred to as ADFs) 81, 81a and 81b, a controlled filter (hereinafter referred to as CNF) 82, a correction filter 83, a correction filter 84 for feedback, and the like. It is composed of a filter, a differential amplifier 85, a white noise generator 86, and subtractors 87a and 87b. As an initial operation, the correction filters 83 and 84 are identified. The identification of the correction filter 83 is to identify the loudspeaker 1 before the muffling operation.
From the center of gravity of the error microphones 5 ₁ and 5 _{2 to} the transfer function G
The operation of calculating 11 and the identification of the correction filter 84 for feedback are operations of calculating the transfer function G12 between the loudspeaker 1 and the sensor microphone 2, and correspond to the calculated transfer characteristics G11 and G12. The filter coefficient to be set is set in each of the correction filters A 83 and 84 to ensure the silencing accuracy. Next, as an adaptive control operation at the time of muffling, the inverse characteristic F21 of the transfer function G21 between the sensor microphone 2 and the center of gravity of the error microphones 5 ₁ and 5 ₂ is calculated for each sampling cycle, and the space to be muffled ( At the center of gravity of the error microphones 5 ₁ and 5 _2, a digital control sound signal for canceling noise is generated and output.

The noise signal output from the A / D converter 11 (FIG. 2) is input to the non-inverting input terminal of the differential amplifier 85. The output signal of the differential amplifier 85 is input to the CNF 82 and the correction filter 83. On the other hand, the digital residual noise signal ε (i) output from the A / D converter 16 (FIG. 2) is input to the ADF 81.

The ADF 81 outputs the noise signal XHC (i) output from the correction filter 83 and corrected by the correction filter 84 for feedback and the correction filter 83.
Filtered-x L, which is one of the steepest descent adaptive control algorithms, using the residual noise signal ε (i) based on the value of
Using the MS (Least Mean Square) algorithm, a necessary transfer function (filter coefficient) is adapted so that residual noise is always minimized at the positions of the error microphones 5 ₁ and 5 ₂ . The CNF 82 has an inverse characteristic F of the transfer function G21 from the sensor microphone 2 to the center of gravity of the error microphones 5 ₁ and 5 _2.
The filter coefficient j1 corresponding to 21 is set. The filter coefficient j1 is updated by the ADF 81 every sampling period.

The correction filter 83 has an nth-order FIR (fini
te impulse response) filter,
The transfer characteristic from the loudspeaker 1 to the center of gravity of the error microphones 5 ₁ and 5 ₂ , more precisely, the transfer path (point J →→ D / A converter 1) shown in FIGS. 1, 2 and 5 as a whole.
7-> LPF 18-> amplifier 19-> loudspeaker 1-> error microphone 5 _1- > amplifier 13 _1- > mixer 14-> LPF 15->
A / D converter 16 →→→ point K) and transmission path (point J →→ D / A converter 17 → LPF1)
8 → amplifier 19 → loudspeaker 1 → error microphone 5 ₂
→ amplifier 13 ₂ → mixer 14 → LPF 15 → A / D converter 16 → → → point K) and 500
It is a filter for correcting transfer characteristics of Hz or less.

The correction filter 84 for feedback is
The control sound radiated from the loudspeaker 1 is composed of the nth-order FIR digital filter, and the sensor microphone 2
It is intended to prevent the sound from being converted into an electric signal and causing acoustic feedback. To be precise, between the loudspeaker 1 and the sensor microphone 2, to be precise, the transmission path shown in FIG. 1, FIG. 2 and FIG. (Point J → → D / A converter 17 → LPF 18 → amplifier 19 → loudspeaker 1 →
This is a filter for correcting the transfer characteristics of the sensor microphone 2 → amplifier 9 → LPF10 → A / D converter 11 →→ point L) below 500 Hz. The control sound signal output from the CNF 82 passes through the correction filter 84 for feedback, is input to the inverting input terminal of the differential amplifier 85, and is subtracted from the noise signal in the differential amplifier 85, whereby feedback correction is performed. .

Next, the operation of this example will be described. In the muffling device having the above-described configuration, first, before performing the muffling operation, the correction filters 83,
The filter coefficients h1 and h2 of 4 are obtained. In the initial setting operation, as shown in FIG. 4, an ADF 81a for obtaining the filter coefficient h1 of the correction filter 83 and an ADF 81b for obtaining the filter coefficient h2 of the feedback correction filter 84 are required. As shown in the figure, the white noise generator 86 creates an M-sequence white noise signal X (i) (i is a sampling time) whose band is limited to 500 Hz or less, and corrects the correction filters 83, 84 and the ADF.
81a, 81b and D / A converter 1
7 (Fig. 2).

As a result, the white noise signal X (i) input to the D / A converter 17 is
After being converted into an analog signal, the signal is input to the amplifier 19 via the LPF 18, amplified, and input to the loudspeaker 1 (FIG. 1). Therefore, the white noise corresponding to the white noise signal X (i), which is band-limited to 500 Hz or less, is radiated into the space as a sound wave from the loudspeaker 1 and detected by the error microphones 5 ₁ and 5 ₂ . _{At 1} and 5 ₂ , the detected white noise level is converted into a noise signal and output.

Error microphone 5₁, 5₂Respectively output from
Noise signal of two systems corresponding to the white noise signal X (i)
Is the amplifier 13₁, 13₂Predetermined in (Fig. 2)
Are amplified with a gain of and mixed together in the mixer 14.
After passing through the LPF 15, the A / D converter 16
Converted into a digital mixed noise signal,
It is input to the non-inverting input terminal. It should be noted that the initial setting
The error microphone 5 ₁, 5₂The output signal from the
Since it is not a sound, it is simply called a noise signal. In the following,
It is the same. Here, the operator (resident) is the display 21
Amplifier displayed on (13₁, 13₂) Output voltage effective
Each amplifier 13 while watching the value ratio₁, 13₂Adjust the gain of
The spatial position you want to mute (that is, the error
Iku 5₁, 5₂Center of gravity) S1 is set. For example, in FIG.
Error microphone 5 as shown₁, 5₂Divide the space into Y: X
If you want to use the spatial position S1 as the sound deadening point,
Each amplifier is adjusted so that the effective value ratio a / b becomes equal to Y / X.
13₁, 13₂Adjust the gain of. In addition, b is the amplifier 1
Three₁Output voltage of the amplifier, a is the amplifier 13₂Of output voltage
It is an effective value.

On the other hand, the ADF 81a and the correction filter 83
The white noise signal X (i) input to is subjected to a convolution operation with the filter coefficient h1 (initial value is random or 0) of the correction filter 83 obtained by the ADF 81a, and then the obtained operation result YC (i ) Is output from the correction filter 83 and input to the inverting input terminal of the subtractor 87a. In the subtractor 87a, the convolution operation result YC (i) processed by the correction filter 83 is subjected to a difference operation from the mixed noise signal output from the A / D converter 16, and the obtained operation result is the error signal εC (i ) Is output as AD
Input to F81a.

As a result, in the ADF 81a, based on the error signal εC (i) and the white noise signal X (i),
The Filtered-x LMS algorithm processing shown in Expression (1) is executed, and the loudspeaker 1 outputs the error microphones 5 ₁ , 5 ₂
Transfer function G11 up to the center of gravity of
All the filter coefficients of a are updated to the filter coefficient h1 corresponding to the obtained transfer function G11. Then, this filter coefficient h1 is set in the correction filter 83 as a filter coefficient h1 for correction during the actual muffling operation. h1 (i + 1) (n) = h1 (i) (n) -2μX (i−n) εC (i) (1) In the equation (1), i is a sampling time and n is an ADF.
The tap number of 81a (filter order), μ is a convergence coefficient.

On the other hand, the white noise corresponding to the white noise signal X (i) radiated as a sound wave from the loudspeaker 1 (FIG. 1) is also detected by the sensor microphone 2. Therefore, the white noise is detected by the sensor microphone 2. The level of white noise is converted into a noise signal and output. Next, the noise signal corresponding to the white noise signal X (i) output from the sensor microphone 2 is input to the controller 8 and the amplifier 9 (FIG. 2).
After being amplified in, the signal is converted into a digital noise signal in the A / D converter 11 via the LPF 10 and input to the non-inverting input terminal of the subtractor 87b.

On the other hand, the white noise signal X (i) input to the ADF 81b and the correction filter 84 for feedback.
Is calculated by performing a convolution operation with the filter coefficient h2 (initial value is random or 0) of the correction filter obtained by the ADF 81b, and then the obtained calculation result YH (i) is output from the correction filter 84 for feedback. , Subtractor 87
It is input to the inverting input terminal of b. In the subtractor 87b, the result YH (i) of the convolution operation processed by the correction filter 84 for feedback is subtracted from the noise signal output from the A / D converter 11 (FIG. 2), and the obtained result is The error signal εH (i) is output and AD
It is input to F81b.

As a result, in the ADF 81b, based on the error signal εH (i) and the white noise signal X (i),
The Filtered-x LMS algorithm processing shown in Expression (2) is executed to obtain the transfer function G12 from the loudspeaker 1 to the sensor microphone 2, and the filter coefficient of the ADF 81b is a filter corresponding to the obtained transfer function G12. All are updated to the coefficient h2. Then, this filter coefficient h2
Is set in the correction filter 84 for feedback as the filter coefficient h2 for feedback correction during the actual silencing operation. h2 (i + 1) (n) = h2 (i) (n) -2μX (i−n) εH (i) (2) In the equation (2), i is the sampling time and n is the ADF.
81b is a tap number (filter order), and μ is a convergence coefficient. The processing described above is performed during one sampling cycle. By repeating this process for about 10 to 20 seconds, the filter coefficient h1 of the correction filter 83 and the filter coefficient h2 of the feedback correction filter 84 are obtained.

After that, the operator (resident) operates the operating means (not shown) to set the obtained filter coefficients h1 and h2 as the filter coefficients h1 and h2 of the spatial region S1 to be silenced in the storage device 122. Input what should be stored in.
Thus, the obtained filter coefficients h1 and h2 are stored as the filter coefficients h1 and h2 of the spatial region S1 to be silenced, together with the effective value ratio a / b (= Y / X) of the amplifier output voltage.

Next, the second initialization is performed. This 2
In the initial setting for the second time, the operator (resident) is
By adjusting the gains of the amplifiers 13 ₁ and 13 ₂ while observing the effective value ratios of the output voltages of the amplifiers (13 ₁ and 13 ₂ ) displayed in, the spatial position (that is, the error microphone 5 ₁ , Set the center of gravity of 5 ₂ ) S2. For example, when it is desired to set the spatial position S2 (FIG. 1), which internally divides the error microphones 5 ₁ and 5 ₂ into x: Y, as the muffling point, the effective value ratio a of the output voltage
/ B is equal to X / Y so that each amplifier 13 ₁ , 1
Adjust the gain of 3 ₂ . Note that b is the effective value of the output voltage of the amplifier 13 ₁ , and a is the effective value of the output voltage of the amplifier 13 ₂ . In other respects, it is substantially the same as the above-described first-time initial setting. When the filter coefficient h1 of the correction filter 83 and the filter coefficient h2 of the correction filter 84 for feedback are obtained by the second initial setting, the operator (resident) operates the operation means (not shown) to obtain them. The stored filter coefficients h1 and h2 are input as the filter coefficients h1 and h2 of the spatial region S2 to be silenced, and the storage device 122 is instructed to store them. As a result, the obtained filter coefficients h1 and h2 are the effective value ratio a / b of the amplifier output voltage.
Together with (= X / Y), they are stored as the filter coefficients h1 and h2 of the spatial area S2 to be silenced.

When the initial setting described above is completed,
First, the operator (resident) operates the operating means (not shown).
Then, if S1 is desired as the spatial area to be silenced, the signal
The processing unit 12 determines the spatial area S to be silenced from the storage device 122.
Read the effective value ratio of the amplifier output voltage corresponding to 1.
It is displayed on the indicator 21. The operator (resident) uses each amplifier 1
Three ₁, 13₂Of the amplifier output voltage by adjusting the gain of
Match the value ratio a / b with the value Y / X displayed on the display 21.
It On the other hand, the signal processing unit 12 silences the storage device 122.
Read out the filter coefficients h1 and h2 of the spatial region S1 to be
Correction filter 83 and a correction filter for feedback.
To the router 84. From this, the mute operation is started.
It In Fig. 1, noise entering the room through the window opening 7
Sensor microphone 2 detects its level and
After being converted into the sound signal SM, in the amplifier 9 (FIG. 2)
Amplified and passed through LPF 10 to A / D converter 11
And converted into a digital noise signal X (i). Soshi
The digital noise output from the A / D converter 11.
The sound signal X (i) passes through the signal processing unit 12 and the differential amplifier 8
5 (FIG. 5) non-inverting input terminal.

On the other hand, in the CNF 82 shown in FIG. 5, the control sound signal subjected to the convolution operation with the filter coefficient j1 (initial value is 0) is the correction filter 84 for feedback.
In FIG. 11, after the convolution operation is performed with the filter coefficient h2, it is input to the inverting input terminal of the differential amplifier 85. As a result, the differential amplifier 85 subtracts the output signal H (i) of the feedback correction filter 84 from the noise signal to perform feedback correction.

Next, the output signal of the differential amplifier 85 is CN
It is input to the F82 and the correction filter 83. Thereby, in the CNF 82, the convolution operation is performed with the filter coefficient j1 to generate the control sound signal, which is then converted into the analog signal in the D / A converter 17 (FIG. 2),
It is input to the amplifier 19 via the LPF 18, amplified, and then input to the loudspeaker 1 shown in FIG. However, since the initial value of the filter coefficient j1 is 0, no sound wave is emitted from the loudspeaker 1 immediately after the start of the muffling operation. Further, in the correction filter 83 shown in FIG. 5, convolution calculation is performed on the output signal of the differential amplifier 85 with the filter coefficient h1 set at the time of initialization, and the calculation result XHC (i) is input to the ADF 81. It

On the other hand, the level of residual noise is detected by the error microphones 5 ₁ and 5 ₂ (FIG. 1), and the residual noise signal EM is detected.
1, converted to EM2. Two error microphones 5 ₁ , 5 ₂
The noise signals of the two systems respectively output from the amplifier 1
3 ₁ , 13 ₂ (Fig. 2) is amplified with a predetermined gain,
After being mixed in the mixer 14, it is passed through the LPF 15 and A
It is converted into a digital mixed residual noise signal ε (i) in the / D converter 16 (the effective value ratio a / b of the output voltages of the amplifiers 13 ₁ and 13 ₂ is equal to Y / X). This allows
In the ADF 81 shown in FIG. 5, the feedback correction filter 84 output from the correction filter 83 and the noise signal XHC corrected by the correction filter 83
Based on the value of (i), the adaptive processing by the Filtered-x LMS algorithm shown in Expression (3) is performed by using the mixed residual noise signal ε (i), and the error microphones 5 ₁ , 5 _{2 are} always used.
The inverse characteristic F21 of the transfer function G21 between the sensor microphone 2 and the center of gravity of the error microphones 5 ₁ and 5 ₂ is calculated so that the residual noise at the center of gravity of the error microphone is minimized. The filter coefficient value of the ADF 81 is the inverse characteristic F21 of the obtained transfer function G21.
Are all updated to the filter coefficient j1 corresponding to. As a result, all the filter coefficients of the CNF 82 are replaced with this filter coefficient j1. j1 (i + 1) (n) = j1 (i) (n) -2 μXHC (i−n) ε (i) (3) In equation (3), i is the sampling time and n is the ADF.
81 is a tap number (filter order), and μ is a convergence coefficient.

Therefore, in the CNF 82, the convolution operation is performed on the newly input noise signal with the newly set filter coefficient j1 to generate the new control sound signal SP, and then the D / A converter is generated. In 17 (FIG. 2), it is converted into an analog signal, is input to the amplifier 19 via the LPF 18, is amplified, and is input to the loudspeaker 1 (FIG. 1). As a result, a new control sound corresponding to the new control sound signal SP is radiated as a sound wave from the loudspeaker 1 and interferes with the external noise having the same amplitude and opposite phase in the spatial region S1 desired to be silenced to weaken the external noise.

The level of the sound pressure of the residual noise after interference is detected by the error microphones 5 ₁ and 5 ₂ and converted into residual noise signals EM1 and EM2, and then the amplifiers 13 ₁ and 13 are used.
₂ (Fig. 2) is amplified with a predetermined gain, and mixer 1
Mixed in 4. Then, the output signal of the mixer 14 is converted into a digital mixed residual noise signal ε (i) in the A / D converter 16 via the LPF 15, and is input again to the ADF 81 (FIG. 5) of the signal processing unit 12.
The adaptive active control described above is repeated in the ADF 81 for each sampling cycle (one noise signal output from the A / D converter 11), and finally, external noise and control sound are generated in the spatial region S1 where silencing is desired. With the same amplitude and opposite phase, they cancel each other sufficiently and the external noise is actively silenced in the spatial region S1 where the noise is desired.

Next, the spatial area to be silenced is changed from S1 to S2.
When it is desired to change to, the operator (resident) operates the operating means to set S2 as the space area to be muted.
In response to this setting operation, the signal processing unit 12 causes the storage device 1 to
The effective value ratio of the amplifier output voltage corresponding to the spatial region S2 to be silenced is read from 22 and displayed on the display 21. The operator (resident) adjusts the gains of the amplifiers 13 ₁ and 13 ₂ to display the effective value ratio a / b of the amplifier output voltage on the display 21.
Set to the value X / Y displayed on. On the other hand, the signal processing unit 1
2 reads the filter coefficients h1 and h2 of the spatial region S2 to be silenced from the storage device 122 and transfers them to the correction filter 83 and the correction filter 84 for feedback. As a result, the above-described silencing operation is repeated, and finally, in the space region S2 where silencing is desired, the external noise and the control sound have the same amplitude and opposite phase, and the space is sufficiently canceled to cancel each other. External noise is actively silenced in the area S2.

Although the embodiment of the present invention has been described in detail above with reference to the drawings, the specific structure is not limited to this embodiment, and the design change and the like without departing from the scope of the present invention. Even this is included in this invention. For example, the present invention is not limited to window openings, but may be applied to noise entering from a duct. The number of error microphones is not limited to two and may be three or more. When the number of error microphones is three or more, the spatial region to be silenced can be appropriately changed within the range surrounded by the error microphones. Further, the residual noise detecting means is not limited to the error microphone, but an acceleration sensor may be used. Further, for example, in the above-described embodiments, the example in which the signal processing unit is configured by software has been shown, but the present invention is not limited to this, and part or all of the signal processing unit may be configured by hardware. Further, in the above-mentioned embodiment, the noise of 500 Hz or less is to be silenced, but the present invention is, of course, 500 H.
It is also applicable to muffling noise of z or higher. Further, in the above-described embodiment, the case where two spatial regions to be muted is set has been described, but three or more spatial regions can be set.

[0038]

As described above, according to the muffling device of the present invention, the position to mute can be arbitrarily moved, although there are certain restrictions. Therefore, it can contribute to making the sound environment of the house comfortable.

[0039]

[Brief description of drawings]

FIG. 1 is a schematic diagram showing an overall configuration of a silencer according to an embodiment of the present invention.

FIG. 2 is a block diagram showing an electrical configuration of a controller used in the silencer of the same embodiment.

FIG. 3 is a block diagram showing an electrical configuration of a signal processing unit incorporated in the controller.

FIG. 4 is a block diagram showing an electrical configuration at the time of initial setting (correction filter identification) of the signal processing unit realized by software (Filtered-x LMS algorithm).

FIG. 5 is a block diagram showing an electrical configuration of the signal processing unit realized by software (Filtered-x LMS algorithm) during a muffling operation.

FIG. 6 is an explanatory diagram for explaining a method of setting a spatial region to be silenced by the silencer of the embodiment.

FIG. 7 is a schematic diagram showing an overall configuration of a conventional silencer applied to a passenger compartment such as an aircraft or an automobile.

FIG. 8 is a block diagram showing an electrical configuration of a controller used in a conventional silencer.

[Explanation of symbols]

1 Loudspeaker (control sound emitting means) 2 Sensor microphone (noise detecting means) 5 ₁ , 5 ₂ Error microphone (residual noise detecting means) 8 Controller (control means) 81 Adaptive digital filter (ADF) 82 Controlled filter (CNF) ) 83 correction filter 84 correction filter for feedback 12 signal processing unit (main part of control means) 13 ₁ , 13 ₂ amplifier (residual noise signal amplification means) 14 mixer (residual noise signal mixing means) 20 effective value ratio calculation circuit 21 Display S1, S2 Space area to be silenced

Claims (1)

[Claims] 1. A noise detecting means for detecting an incoming noise and outputting a noise signal, and a control sound emitting means for emitting a control sound to a predetermined spatial region to be silenced based on an input control sound signal. The residual noise is provided such that it surrounds or surrounds a predetermined space area to be silenced, and the residual noise is canceled by interference with the control sound emitted from the control sound emitting means and remains. A plurality of residual noise detecting means for detecting and outputting a residual noise signal can be independently set and adjusted to a predetermined amplification degree, and the residual noise signals output from the corresponding residual noise detecting means can be independently set and adjusted. A plurality of residual noise signal amplifying means for amplifying the residual noise signal and a residual noise signal for amplifying the plurality of residual noise signals outputted from the plurality of residual noise signal amplifying means. No. mixing means, based on the noise signal and the residual noise signal after amplification and mixing, in the predetermined spatial region to be silenced determined from the ratio of the effective value of the plurality of residual noise signals after amplification and before mixing On the other hand, in order to always minimize the residual noise, a control means for generating a control sound signal for emitting a control sound having a phase opposite to the waveform component of the noise and supplying the control sound signal to the control sound emitting means is provided. A silencer characterized by the above.