JP2001142469A - Active muffler - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve such problem that noise has not been always minimized at the ears of a listener because a conventional active muffler is controlled so that the detected output of an error microphone is minimized, but the error microphone is difficultly arranged close to the ears of the listener, then is arranged away from the ears. SOLUTION: An adaptive filter C1' is provided between a speaker 4 that outputs signal sound and an error microphone 2 that is used for error detection. A filter VC' converts speaker output sound detected by the microphone 2 into the speaker output sound at the ears of the listener. The active muffler is incorporated with an adaptive filter C2' to obtain the sound field characteristic between the speaker 4 and the ears of the listener by using the filters C1' and VC' to identify the adaptive filter C1' expressing the sound field characteristic between the speaker 4 and the microphone 2.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an active silencer using active noise control, and more particularly to an active silencer having an improved silencing effect at the listener's ear.

[0002]

2. Description of the Related Art In recent years, there has been proposed an active silencer using active noise control that silences vehicle interior noise by outputting a control sound from a speaker. For example, as shown in FIG. Reference sensor 1 which is a signal detector such as a microphone or an engine speed sensor, error microphone 2 which is an error detector, speaker 4 which outputs control sound, and speaker 4 expressed by a transfer function and error microphone 2 A filter C # representing a sound field characteristic, an adaptive filter W # representing a sound field characteristic between the reference sensor 1 and the error microphone 2 represented by a transfer function, and an LMS as a coefficient calculator
It is composed of a computing unit 8.

[0003] In the active silencer configured as described above, noise is detected by the reference sensor 1,
The detection signal is input to filter C # and adaptive filter W #. The noise signal input to the adaptive filter W ＾ is
The signal is processed by the adaptive filter W # and output from the speaker 4. Then, in the error microphone 2 placed near the listener, the interference between the reproduced sound from the speaker 4 and the noise from the noise source is detected, whereby the coefficient of the adaptive filter W ＾ is changed and the feedback is performed. The speaker 4 is set so that the detection signal of the error microphone 2 as a signal is minimized.
The output is adjusted to attenuate the noise.
In this case, the detection signal from the error microphone 2 is
LM so that the detection signal of the error microphone 2 is minimized based on the detection signal and the output of the filter C #.
S operation (least squares method) is performed, whereby the adaptive filter W ＾ is adapted to adapt to the sound field characteristics at each time.
Have been changed and identified.

[0004]

In the active silencer configured as described above, control is performed so that the detection signal of the error microphone 2 which is a feedback signal is minimized. Is difficult to place near the ear of the listener located in the space to be silenced,
The noise was not always minimized at the listener's ear because it was located away from the ear. In view of the above, the present invention provides an active silencer capable of performing control such that noise at the listener's ear is actually minimized.

[0005]

The problem to be solved by the present invention is as described above. Next, means for solving the problem will be described. That is, in claim 1, the sound field characteristics between the speaker that outputs the signal sound for canceling the noise and the ear of the listener, and the sound field characteristics between the speaker and the error microphone that performs the error detection are described. A filter configured to obtain a speaker output sound detected by the error microphone and a filter that converts the speaker output sound at the ear position of the listener into a speaker output sound is incorporated.

According to a second aspect of the present invention, a filter representing a sound field characteristic between a speaker and an error microphone is an adaptive filter capable of online identification.

Further, in claim 3, a filter for converting the speaker output sound detected by the error microphone into the speaker output sound at the ear position of the listener is identified in advance.

According to a fourth aspect of the present invention, when identifying a filter for converting a speaker output sound detected by an error microphone into a speaker output sound at a listener's ear position, a signal detected at the listener's ear position is identified. It was configured to delay.

According to a fifth aspect of the present invention, when the detection of the signal output from the speaker at the ear position of the listener is performed earlier than the detection of the signal output from the speaker by the error microphone, the speaker and the listener A filter representing the sound field characteristics between the speaker and the error microphone, and a filter representing the sound field characteristics between the speaker and the listener's ear by delaying the detection signal between the speaker and the ear in time. And a filter for converting the speaker output sound detected by the error microphone into the speaker output sound at the ear position of the listener.

[0010] In claim 6, the error microphone is arranged at a position where the distance between the listener's ear and the speaker is equal to or greater than the distance between the speaker and the error microphone.

According to a seventh aspect of the present invention, a sound field characteristic between a reference sensor for detecting noise and an ear of a listener and a sound field characteristic between the reference sensor and an error microphone for detecting an error are represented. A filter configured to obtain the noise detected by the error microphone using a filter that converts the noise detected at the ear position of the listener into a noise is incorporated.

Further, in claim 8, a filter for converting noise detected by the error microphone into noise at the ear position of the listener is identified in advance.

According to a ninth aspect of the present invention, when identifying a filter that converts noise detected by the error microphone into noise at the ear position of the listener, a signal detected at the ear position of the listener is delayed. did.

According to a tenth aspect of the present invention, when the detection of the signal from the reference sensor position at the ear position of the listener is performed earlier than the detection of the signal from the reference sensor position by the error microphone, the detection from the reference sensor position is performed. The filter that represents the sound field characteristic between the reference sensor and the error microphone by delaying the detection signal at the ear position of the listener in time, and the sound field characteristic between the reference sensor and the ear of the listener Depending on the filter
A filter for converting noise detected by the error microphone into noise at the ear position of the listener is identified.

In the eleventh aspect, the original sound to be silenced is used as a signal for identifying in advance a filter for converting noise detected by the error microphone into noise at the ear position of the listener.

According to a twelfth aspect of the present invention, a filter that outputs a random signal from near the noise source and converts noise detected by the error microphone into noise at the ear position of the listener is identified in advance.

Further, the error microphone is arranged at a position where the distance between the ear of the listener and the reference sensor is equal to or greater than the distance between the reference sensor and the error microphone.

[0018]

Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an active silencer of the present invention, FIG. 2 is a block diagram showing a configuration for identifying an adaptive filter between a speaker and an error microphone, and FIG. 3 is a diagram showing a speaker output sound detected by an error microphone. FIG. 4 is a block diagram illustrating a configuration for identifying a filter that converts a sound output from a speaker at the ear position of a listener; FIG. 4 is a diagram illustrating a microphone in which the distance between the speaker and the error microphone is arranged at the speaker and the ear position of the listener; FIG. 5 is a diagram showing an arrangement state in which the distance is larger than the distance between the microphone and the detection signal by the microphone and the detection signal by the error microphone arranged at the ear position of the listener of the output signal of the speaker in this state. In the state of 4, the detection signal of the microphone arranged at the ear position of the listener is delayed, and the output signal of the speaker in that state FIG. 6 is a diagram showing a detection signal by a microphone and a detection signal by an error microphone arranged at the ear position of the listener. FIG. 6 shows that the distance between the ear of the listener and the speaker is equal to the distance between the speaker and the error microphone. FIG. 7 is a block diagram showing a configuration for identifying a filter that converts noise detected by the error microphone into noise at the ear position of the listener, and FIG. Indicates the arrangement state in which the distance between the reference sensor and the error microphone is larger than the distance between the reference sensor and the microphone arranged at the ear position of the listener, and the listener's ear of the noise signal in this state. FIG. 9 is a diagram showing a detection signal by a microphone arranged at a position and a detection signal by an error microphone. FIG. 9 shows the ear of the listener in the state shown in FIG. FIG. 10 is a diagram illustrating a state in which the detection signal of the microphone arranged at the position is delayed, and a detection signal of the microphone arranged at the ear position of the listener and a detection signal of the error microphone of the noise signal in that state; FIG. 11 shows a state in which an error microphone is arranged at a position where the distance between the ear and the reference sensor is equal to or larger than the distance between the reference sensor and the error microphone. FIG. 1 is a diagram showing a state in which it is possible to identify a filter for converting noise output and detected by an error microphone into noise at the ear position of a listener;
2 is a block diagram showing a conventional active silencer.

First, the basic configuration of the active silencer of the present invention will be described. As shown in FIG. 1, the active silencer includes a reference sensor 1, which is a noise detector,
An error microphone 2 serving as an error detector, a speaker 4 for outputting a control sound, an adaptive filter C2 # representing a sound field characteristic between the speaker 4 represented by a transfer function and the ear of a listener located in a sound-reduction target space; An adaptive filter W2 # representing a sound field characteristic between the reference sensor 1 represented by the transfer function and the ear of the listener located in the sound-reduction target space, and the error microphone 2 represented by the transfer function and located in the sound-reduction target space. It is composed of an adaptive filter VW # representing a sound field characteristic between the listener and the ear, an LMS calculator 8 as a coefficient calculator, and the like. Further, the adaptive filter C2 # is located at an adaptive filter C1 # representing a sound field characteristic between the error microphone 2 and the speaker 4 represented by the transfer function and the error filter 2 represented by the transfer function. And an adaptive filter VC # representing the sound field characteristics between the listener and the ear of the listener.

The operation of the thus configured active silencer will be described. First, noise from a noise source is detected by the reference sensor 1, and a reference signal, which is a detection signal, is input to the adaptive filters C2 # and W2 #. In this example, the noise from the noise source is, for example, an engine sound. The reference sensor 1 is configured as a microphone, for example, but when the noise source is an engine sound as in this example, an engine speed pickup or the like may be used.

The noise signal input to the adaptive filter W2 # is signal-processed by the adaptive filter W2 #, then output from the speaker 4, and also input to the adaptive filters C2 # and C1 #. The output signal from the speaker 4 interferes with the noise from the noise source in the noise reduction target space where the listener's ear is located, and thereby the noise in the noise reduction target space is reduced. Further, in the error microphone 2 installed at a position distant from the listener's ear, an interference between the output sound from the speaker 4 and the noise from the noise source is detected.

The output signal of the adaptive filter C1 # is subtracted from the error signal, which is the detection signal from the error microphone 2, by the subtracter 10. Here, the adaptive filter C1 # is
The transfer function between the speaker 4 and the error microphone 2 is identified in advance as a coefficient. In this case, a signal output from the subtractor 10 is a noise signal detected by the error microphone 2 through a sound field between the reference sensor 1 and the error microphone 2 and a sound signal between the speaker 4 and the error microphone 2. From the output signal from the speaker 4 detected through the field, the output signal of the speaker 4 represented by the transfer function is multiplied by the coefficient of the adaptive filter C1 # representing the sound field characteristic between the speaker 4 and the error microphone 2. This is the signal from which the signal has been removed.

Accordingly, the error microphone 2 is output from the subtractor 10.
, Only the noise signal detected by the adaptive filter V is output to the output signal from the subtractor 10.
W ＾ is multiplied. The adaptive filter VW ＾ is
The output signal from the subtractor 10 multiplied by the adaptive filter VW # represents the sound field characteristics between the error microphone 2 and the sound-reduction target space where the ear of the listener is located. , Ie, a noise signal detected by a microphone actually installed at the ear of the listener.

As described above, the sound field characteristics between the reference sensor 1 and the ear of the listener are determined by the reference sensor 1
An adaptive filter W1 # representing a sound field characteristic between the error microphone 1 and the error microphone 2 and an adaptive filter VW # representing a sound field characteristic between the error microphone 1 and the ear of the listener. In this active noise reduction device, the adaptive filter W2 expressing the sound field characteristic between the reference sensor 1 and the listener's ear determined as described above as a transfer function.
＾ is incorporated.

The output signal from the subtracter 10 is input to the adder 9 through the adaptive filter VW #, and the output signal from the speaker 4 is input to the adder 9 through the adaptive filter C2 #. The adaptive filter C2 # represents an adaptive filter C1 # representing a sound field characteristic between the speaker 4 and the error microphone 2 represented by the transfer function, and the adaptive filter C2 # represents the sound field characteristic between the error microphone 2 represented by the transfer function and the ear of the listener. And an adaptive filter VC # representing the sound field characteristic of
The output signal from the speaker 4 passing through 2 ＾ is a noise signal detected through a sound field between the speaker 4 and the ear of the listener, that is, a speaker detected by a microphone actually installed at the ear of the listener. 4 output signal. In this manner, the sound field characteristics between the speaker 4 and the listener's ear, the adaptive filter C1 # representing the sound field characteristics between the speaker 4 and the error microphone 2, the error filter 2 and the listener's ear are shown. Adaptive filter VC フ ィ ル タ representing the sound field characteristic between
It is configured to obtain using

In the adder 9, the input subtractor 1
The output signal from 0 and the output signal from the loudspeaker 4 are added, but the two signals cancel each other because they have waveforms of opposite phases. Then, the output signal from the adaptive filter C2 # together with the output signal from the adder 9 is input to the LMS calculator 8, and the LMS calculator 8 controls the adaptive filter so that the output signal from the adder 9 is minimized. Adjust the coefficient of W2 ＾. That is, the adaptive filter W2 # is identified such that noise at the listener's ear is minimized.

As described above, by considering the sound field characteristics between the speaker 4 and the listener's ear and the sound field characteristics between the reference sensor 1 and the listener's ear, the sound field at the listener's ear is considered. The active silencer can be configured so that the noise is minimized, and the listener's bodily silencing effect can be improved.

Next, an adaptive filter C1 # constituting an adaptive filter C2 # whose output is input to the LMS calculator 8
The following describes a method of identifying the adaptive filter C1 #, which adjusts the coefficient of the adaptive filter to adapt to the sound field characteristics at each time.
FIG. 2 shows an active silencer capable of identifying the adaptive filter C2 #. A signal processor 11 is provided downstream of the error microphone 2, and an error signal which is a detection signal of the error microphone 2 is output from the signal processor 11. It is configured to be input to the processor 11. The speaker 4 outputs an M-sequence signal that is randomly output over the entire frequency band together with the noise signal that has been signal-processed by the adaptive filter W2 #.

In the error microphone 2, both the noise source and the engine sound from the speaker 4 and the M-sequence signal from the speaker 4 are detected, and this detection signal is input to the signal processor 11 as an error signal. You. Signal processor 1
1 is, for example, a delay 21 for delaying a signal;
It is composed of an S operator 22 and an adaptive filter 23, and the detection signal input to the signal processor 11 is delayed by the delay 21 for a certain time.

Here, since the signal correlated with the engine sound included in the error signal is a periodic signal having a constant cycle based on the engine speed, a signal correlated with the engine sound in the error signal is used. By superimposing the signal correlated with this engine sound with a signal delayed for a certain time,
They will be offset by each other. Since the M-sequence signal output from the speaker 4 is a random signal having no periodicity, the M-sequence signal in the detection signal and a signal obtained by delaying the M-sequence signal for a predetermined time are superimposed. Are not offset by each other.

In the signal processor 11, the error signal is converted from the input error signal into a delay signal 21.
, A signal delayed for a certain time is subtracted, and a signal obtained by the subtraction is output. in this way,
By subtracting the error signal delayed by the delay 21 from the input error signal, signals correlated with the engine sound, which is a periodic signal, are canceled out and deleted, and only the signal correlated with the M-sequence signal is deleted. It can be separated and output as a separated signal.

In this case, the error signal delayed by the delay 21 is subtracted from the error signal input to the signal processor 11 after the delay time and the amplitude are adjusted by the adaptive filter 23. Also, delay 2
1. The error signal delayed by 1 and the signal processor 11
The separated signal obtained by subtracting the error signal delayed by the delay 21 from the error signal input to the LMS calculator 22 is input to the LMS computing unit 22, and based on these signals, the LMS
The LMS calculation is performed by the calculator 22, and the coefficient of the adaptive filter 23 is adjusted so that the signal remaining in the separated signal and correlated with the engine sound is minimized.

The separated signal obtained by separating the signal correlated with the engine sound by the signal processor 11 is applied to the adaptive filter C1 representing the sound field characteristic between the speaker 4 and the error microphone 2 by the subtracter 7. After the M-sequence signal passing through ＾ is subtracted, the signal is input to the LMS calculator 13. In addition, the LMS
The M-sequence signal emitted from the active silencer is directly input to the arithmetic unit 13. An LMS operation is performed by the LMS operation unit 13 based on these signals, and the adaptive filter C1 # is set so that the difference between the M-sequence signal emitted from the active silencer and the output signal from the subtractor 7 is minimized. Is adjusted.

As described above, in the signal processor 11, a signal obtained by performing a filtering process by the adaptive filter 23 on the error signal delayed by the delay 21 from the input error signal which is the detection signal of the error microphone 2. By performing signal processing such as subtraction,
It is possible to delete a signal correlated with the engine sound from the error signal and leave only a separated signal correlated with the M-sequence signal. By performing the calculation by the LMS calculator 13 so that the separated signal and the M-sequence signal emitted from the active silencer match, the sound field characteristics between the speaker 4 and the error microphone 2 can be identified. It is possible to do.

That is, by adjusting the coefficient of the adaptive filter C1 # as described above, the adaptive filter C1 # is identified so as to adapt to the sound field characteristic between the loudspeaker 4 and the error microphone 2 that changes every moment. You can do it.
As a result, the sound field characteristics between the speaker 4 and the error microphone 2 can be accurately reproduced, and the noise reduction characteristics of the active noise reduction device can be improved. As described above, in the signal processor 11 in this example, the engine sound and the M
An error signal detected by the error microphone 2 is used for signal processing for removing a signal correlated with the engine sound from a signal in which a series signal is mixed.

Next, a description will be given of an active silencer in which the adaptive filter VC # between the error microphone 2 and the ear of the listener can be identified in advance. As shown in FIG.
First, the microphone 5 is installed near the ear of the listener, and the signal processor 16 is provided downstream of the microphone 5. The signal processor 16 includes the delay 21 and the LMS operation unit 2.
2 and an adaptive filter VC #. An output signal from the speaker 4 is detected by the microphone 5 through a sound field between the speaker 4 and the ear of the listener, and the detection signal is input to the signal processor 16. Further, the output signal from the speaker 4 is detected by the error microphone 2 through a sound field between the speaker 4 and the error microphone 2, and the detection signal is input to the signal processor 16.

In the signal processor 16, the detection signal from the microphone 5 is input to the subtractor 6 after being delayed for a predetermined time by the delay 21, and the detection signal from the error microphone 2 is subtracted through the adaptive filter VC #. The detection signal of the error microphone 2 is subtracted from the detection signal of the microphone 5 after that. Then, the output signal from the subtractor 6 and the detection signal of the error microphone 2 are input to the LMS computing unit 22, and the LMS computing unit 2
2, and the coefficient of the adaptive filter VC # is adjusted so that the output signal from the subtractor 6 is minimized.
That is, a signal obtained by multiplying a detection signal from the error microphone 2, which is an output signal of the speaker 4 detected through a sound field between the speaker 4 and the error microphone 2, by the adaptive filter VC #, is output from the speaker 4 and the microphone 5. The adaptive filter VC # is identified so as to be equal to the detection signal from the microphone 5 which is the output signal of the speaker 4 detected through the sound field between the two.

Here, in the case of this example, as shown in the left diagram of FIG. 4, the distance L between the speaker 4 and the error microphone 2 is determined.
_C1 is the distance L _C2 between the speaker 4 and the microphone 5
The speaker 4, the error microphone 2, and the microphone 5 are installed so as to be larger. Therefore,
The output signal of the speaker 4 is detected by the microphone 5 earlier by ΔT1 than by the error microphone 2. As described above, when the detection of the output signal of the speaker 4 by the microphone 5 is performed earlier than the detection by the error microphone 2, the conversion that converts the detection signal of the error microphone 2 into the detection signal of the microphone 5 on the time axis. Since the coefficient V0 does not exist, the subtractor 6 cannot subtract both signals.

[0039] Therefore, as shown in FIG. 5, by delaying the detection signal of the microphone 5 by the delay 21 by .DELTA.Td, 'as the distance L _C2 from virtually speaker 4' microphone 5 wherein the distance L _C1 In this case, the output signal of the speaker 4 is detected by the microphone 5 with a delay of ΔT2 from the detection by the error microphone 2. As described above, by detecting the output signal of the speaker 4 by the error microphone 2 earlier than the detection by the microphone 5, the detection signal of the error microphone 2 is converted to the detection signal of the microphone 5 on the time axis. The conversion coefficient V1 to be converted can be obtained. This is equivalent to converting the detection signal of the error microphone 2 into the detection signal of the microphone 5 in the past, so that the subtractor 6 can subtract both signals.

As described above, when identifying the adaptive filter VC # between the error microphone 2 and the listener's ear, the microphone 5 is installed at the listener's ear position, and the speaker 4 detected by the microphone 5 is used. Is configured to delay the output signal from the speaker 4, identification in consideration of the sound field characteristics between the speaker 4 and the ear of the listener can be performed in advance, for example, at the time of assembling or shipping the vehicle. Thus, it is possible to improve the bodily sensation silencing effect of the active silencer.

Further, when identifying the adaptive filter VC #, an arrangement may be made wherein the output signal from the speaker 4 detected by the microphone 5 is not delayed without using the delay 21. That is, as shown in FIG.
The distance L _C2 between the speaker 4 and the microphone 5 is smaller than the distance L _C1 between the speaker 4 and the error microphone 2.
The speaker 4, the error microphone 2, and the microphone 5 are set to be equal or larger.

As described above, by making the distance L _C2 greater than the distance L _C1 , the detection of the output signal of the speaker 4 by the microphone 5 is delayed from the detection by the error microphone 2 without using the delay 21. Can be done,
It is possible to perform subtraction of both signals by the subtractor 6.
This simplifies the configuration of the active silencer,
Cost can be reduced.

Next, an active noise reduction device will be described in which the adaptive filter VW # between the reference sensor 1 and the ear of the listener can be identified in advance. As shown in FIG. 7, similarly to the case where the adaptive filter VC # is identified, first, the microphone 5 is installed near the ear of the listener, and the signal processor 17 is provided downstream of the microphone 5. The signal processor 17 includes the delay 21, the LMS calculator 22, and an adaptive filter VW #. The microphone 5 detects a noise signal from the noise source through a sound field between the reference sensor 1 and the ear of the listener, and the detection signal is input to the signal processor 17. In addition, the noise signal from the noise source is detected by the error microphone 2 through the sound field between the reference microphone 1 and the error microphone 2, and the detected signal is output to the signal processor 17.
Is input to

In the signal processor 17, the detection signal from the microphone 5 is input to the subtractor 6 after being delayed by a predetermined time by the delay 21, and the detection signal from the error microphone 2 is subtracted through the adaptive filter VW #. The detection signal of the error microphone 2 is subtracted from the detection signal of the microphone 5 after that. Then, the output signal from the subtractor 6 and the detection signal of the error microphone 2 are input to the LMS computing unit 22, and the LMS computing unit 2
2 and the coefficient of the adaptive filter VW # is adjusted so that the output signal from the subtractor 6 is minimized.
That is, the error microphone 2 which is a noise signal detected through the sound field between the reference sensor 1 and the error microphone 2
Is multiplied by the adaptive filter VW # so that the signal obtained by multiplying the adaptive filter VW # by the adaptive filter VW # becomes equal to the signal detected by the microphone 5 which is a noise signal detected through the sound field between the reference sensor 1 and the microphone 5.同 定 is identified.

In this case, as shown in the left diagram of FIG. 8, the distance L _w1 between the reference sensor 1 and the error microphone 2 is determined by the distance between the reference sensor 1 and the microphone 5.
The reference sensor 1, the error microphone 2, and the microphone 5 are provided so as to be larger than the distance _Lw2 between the reference sensor 1 and the error microphone 2. Therefore, detection of noise is performed earlier by the microphone 5 by ΔT′1 than by the error microphone 2. As described above, when the noise detection by the microphone 5 is performed earlier than the detection by the error microphone 2, the conversion coefficient V′0 for converting the detection signal of the error microphone 2 into the detection signal of the microphone 5 on the time axis. Does not exist, the subtractor 6 cannot perform the subtraction of both signals.

Therefore, as shown in FIG. 9, the detection signal of the microphone 5 is delayed by ΔT′d by the delay 21 so that the distance L _w2 ′ from the reference sensor 1 is virtually reduced like the microphone 5 ′. Distance L
_{It is set} to be larger than _w1 so that the noise signal is detected by the microphone 5 with a delay of ΔT′2 from the detection by the error microphone 2. In this way, by configuring the detection of the noise signal by the error microphone 2 earlier than the detection by the microphone 5, the detection signal of the error microphone 2 on the time axis is detected.
It is possible to obtain the conversion coefficient V′1 to be converted into the detection signal of This is equivalent to converting the detection signal of the error microphone 2 into the detection signal of the microphone 5 in the past, so that the subtractor 6 can subtract both signals.

As described above, when identifying the adaptive filter VW # between the error microphone 2 and the listener's ear, the microphone 5 is installed at the listener's ear position, and the noise signal detected by the microphone 5 is detected. , The identification considering the sound field characteristics between the reference sensor 1 and the ear of the listener can be performed in advance, for example, at the time of assembling or shipping the vehicle. It is possible to improve the bodily sensation silencing effect of the device.

Further, when identifying the adaptive filter VW #, the noise signal detected by the microphone 5 may not be delayed without using the delay 21. That is, as shown in FIG. 10, the distance L _w2 between the reference sensor 1 and the microphone 5 is equal to or greater than the distance L _w1 between the reference sensor 1 and the error microphone 2. The reference sensor 1, the error microphone 2, and the microphone 5 are installed.

As described above, by making the distance L _w2 larger than the distance L _w1 , the noise signal can be detected by the microphone 5 later than the error microphone 2 without using the delay 21. It is possible to perform the subtraction of both signals by the subtractor 6. This makes it possible to simplify the configuration of the active silencer and reduce costs.

In this example, the adaptive filter VW ＾
The engine sound is used as the noise from the noise source used for the identification of the engine, and in particular, the engine sound and the sweep sound during operation at a single rotation speed are used. Further, as shown in FIG. 11, a speaker 31 is installed near a noise source, the M-sequence signal that is a random signal over a wide band is output from the speaker 31, and the above-described case using noise such as engine sound is used. Similarly, the adaptive filter VW ＾
Can also be identified.

[0051]

As described above, the present invention has the following advantages. That is, as in claim 1,
A filter representing a sound field characteristic between a speaker outputting a signal sound for canceling noise and the ear of a listener, a sound field characteristic between the speaker and an error microphone performing error detection,
Since a filter configured to obtain a speaker output sound detected by the error microphone using a filter that converts the output sound into a speaker output sound at the ear position of the listener is incorporated, the sound field between the speaker and the listener's ear is incorporated. In consideration of the characteristics, it is possible to configure the active silencer so that the noise at the listener's ear is minimized, and the listener's bodily sensation silencing effect can be improved.

Further, since the filter representing the sound field characteristic between the speaker and the error microphone is an adaptive filter that can be identified online, the sound field characteristic between the speaker and the ear of the listener can be reduced. The sound field characteristics between the loudspeaker and the error microphone can be accurately reproduced while taking into account the above, and the noise reduction characteristics of the active noise reduction device can be improved.

Further, since the filter for converting the speaker output sound detected by the error microphone into the speaker output sound at the ear position of the listener is identified in advance, the filter between the speaker and the ear of the listener can be identified. Identification considering sound field characteristics,
For example, this can be performed in advance at the time of assembling or shipping a vehicle, and the bodily sensation silencing effect of the active silencer can be improved.

Further, when identifying a filter for converting the speaker output sound detected by the error microphone into the speaker output sound at the ear position of the listener, the signal detected at the ear position of the listener may be identified. Since the delay is configured, the speaker, the error microphone, and the microphone are installed so that the distance between the speaker and the error microphone is larger than the distance between the speaker and the microphone installed at the ear of the listener. In this case, the identification can be performed in consideration of the sound field characteristics between the speaker and the ear of the listener, and the bodily sensation silencing effect of the active silencer can be improved.

Further, when the detection of the signal output from the speaker at the ear position of the listener is performed earlier than the detection of the signal output from the speaker by the error microphone, the speaker and the listener may be used. A filter representing the sound field characteristics between the speaker and the error microphone, and a filter representing the sound field characteristics between the speaker and the listener's ear by delaying the detection signal between the speaker and the ear in time. Since the filter that converts the speaker output sound detected by the error microphone into the speaker output sound at the ear position of the listener is identified, the distance between the speaker and the error microphone is determined by the microphone installed at the speaker and the ear of the listener. Even if speakers, error microphones, and microphones are installed so that they are larger than the distance between Can perform identification in consideration of the sound field characteristic between, it is possible to improve the experience silencing effect of the active noise cancellation device.

Further, the error microphone is arranged at a position where the distance between the ear of the listener and the speaker is equal to or greater than the distance between the speaker and the error microphone. The adaptive filter between the speaker and the error microphone can be identified without delaying the adaptive filter between the speaker and the ear of the listener in time. This makes it possible to simplify the configuration of the active silencer and reduce costs.

Further, the sound field characteristic between the reference sensor for detecting noise and the ear of the listener is defined as:
The filter is configured to obtain a sound field characteristic between the reference sensor and an error microphone that performs error detection, and a filter that converts noise detected by the error microphone into noise at the ear position of the listener. Since the filter is incorporated, the active silencer can be configured so that the noise at the listener's ear is minimized in consideration of the sound field characteristics between the reference sensor and the listener's ear. It is possible to improve the bodily sensation silencing effect of the listener.

Furthermore, since a filter for converting noise detected by the error microphone into noise at the ear position of the listener is identified in advance, the sound field characteristics between the reference sensor and the ear of the listener are identified. Can be performed in advance, for example, at the time of assembling or shipping a vehicle, and the bodily sensation silencing effect of the active silencer can be improved.

Further, when identifying a filter for converting noise detected by the error microphone into noise at the ear position of the listener, a signal detected at the ear position of the listener is delayed. Therefore, the reference sensor, the error microphone, and the microphone are installed so that the distance between the reference sensor and the error microphone is larger than the distance between the reference sensor and the microphone installed near the listener's ear. In this case, the identification can be performed in consideration of the sound field characteristics between the reference sensor and the ear of the listener, and it is possible to improve the bodily sensation silencing effect of the active silencer.

Further, in the case where the detection of the signal from the reference sensor position at the ear position of the listener is performed earlier than the detection of the signal from the reference sensor position by the error microphone, according to the present invention, The filter that represents the sound field characteristic between the reference sensor and the error microphone by delaying the detection signal at the ear position of the listener in time, and the sound field characteristic between the reference sensor and the ear of the listener The filter that expresses the noise detected by the error microphone and converts it into noise at the ear position of the listener is identified, so the distance between the reference sensor and the error microphone is set at the reference sensor and the ear of the listener. The reference sensor, the error microphone, and the microphone Even if such as O emissions is installed,
Identification can be reliably performed in consideration of the sound field characteristics between the reference sensor and the ear of the listener, and the bodily sensation silencing effect of the active silencer can be improved.

Further, the original sound to be silenced is used as a signal for identifying in advance a filter for converting noise detected by the error microphone into noise at the ear position of the listener. The active silencer can be configured to minimize the noise at the listener's ear in consideration of the sound field characteristics between the reference sensor and the listener's ear without specially providing It is possible to improve the bodily sensation silencing effect of the listener.

Further, a filter for outputting a random signal from the vicinity of the noise source and converting the noise detected by the error microphone into the noise at the ear position of the listener is identified in advance. The active noise reduction device is configured so that the noise at the listener's ear can be minimized in consideration of the sound field characteristics between the reference sensor and the listener's ear in a stable and highly accurate manner. It is possible to improve the effect of silencing the listener.

Furthermore, the error microphone is arranged at a position where the distance between the ear of the listener and the reference sensor is equal to or larger than the distance between the reference sensor and the error microphone. Thus, the adaptive filter between the reference sensor and the error microphone can be identified without delaying the adaptive filter between the reference sensor and the listener's ear in time.
This simplifies the configuration of the active silencer,
Cost can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an active silencer of the present invention.

FIG. 2 is a block diagram showing a configuration for identifying an adaptive filter between a speaker and an error microphone.

FIG. 3 is a block diagram showing a configuration for identifying a filter that converts a speaker output sound detected by an error microphone into a speaker output sound at the ear position of a listener.

FIG. 4 illustrates an arrangement state in which the distance between the speaker and the error microphone is larger than the distance between the speaker and the microphone arranged at the ear position of the listener, and the output signal of the speaker in this state. FIG. 3 is a diagram illustrating a detection signal from a microphone and a detection signal from an error microphone.

FIG. 5 shows a state in which the detection signal of the microphone arranged at the ear position of the listener is delayed in the state of FIG. 4, and a detection signal of the output signal of the speaker in that state detected by the microphone arranged at the ear position of the listener. FIG. 4 is a diagram showing a detection signal from an error microphone.

FIG. 6 is a diagram showing a state in which an error microphone is arranged at a position where the distance between the listener's ear and the speaker is equal to or greater than the distance between the speaker and the error microphone.

FIG. 7 is a block diagram illustrating a configuration for identifying a filter that converts noise detected by the error microphone into noise at the ear position of the listener.

FIG. 8 shows an arrangement state in which the distance between the reference sensor and the error microphone is larger than the distance between the reference sensor and the microphone arranged at the ear position of the listener, and reception of a noise signal in this state. It is a figure which shows the detection signal by the microphone arrange | positioned at the ear position of a listener, and the detection signal by an error microphone.

FIG. 9 shows a state in which the detection signal of the microphone arranged at the ear position of the listener is delayed in the state of FIG. 8, and a detection signal and an error of the noise signal by the microphone arranged at the ear position of the listener in that state. It is a figure showing a detection signal by a microphone.

FIG. 10 is a diagram showing a state where an error microphone is arranged at a position where a distance between a listener's ear and a reference sensor is equal to or larger than a distance between the reference sensor and the error microphone.

FIG. 11 is a diagram showing a state in which a random signal is output from the vicinity of the noise source and a filter that converts noise detected by the error microphone into noise at the ear position of the listener has been identified.

FIG. 12 is a block diagram showing a conventional active silencer.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Reference sensor 2 Error microphone 4 Speaker 8.13 LMS arithmetic unit 9 Adder 10 Subtractor 21 Delay C1 ＾ Adaptive filter (representing sound field characteristic between speaker and error microphone) C2 ＾ (Speaker and listener's ear Adaptive filter VC ＾ (which converts the speaker output sound detected by the error microphone into the speaker output sound at the ear position of the listener) W2 ＾ (the reference sensor and the ear area of the listener) Adaptive filter VW ＾ (converts the noise detected by the error microphone into noise at the ear position of the listener)

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Shigeki Tada 1-32 Chayacho, Kita-ku, Osaka-shi, Osaka F-term in Yanmar Diesel Co., Ltd. (Reference) 3D020 BA02 BC01 BD05 BE04 5D061 FF02 5J023 DA05 DB03 DC08 DD07

Claims (13)

[Claims] A filter that represents a sound field characteristic between a speaker that outputs a signal sound for canceling noise and a listener's ear, and a sound field characteristic between the speaker and an error microphone that performs error detection. An active silencer characterized by incorporating a filter configured to obtain a speaker output sound detected by an error microphone using a filter that converts the speaker output sound into a speaker output sound at the ear position of a listener. 2. The active silencer according to claim 1, wherein a filter representing a sound field characteristic between the speaker and the error microphone is an adaptive filter capable of online identification. 3. The active silencer according to claim 1, wherein a filter for converting a speaker output sound detected by the error microphone into a speaker output sound at the ear position of the listener is identified in advance. 4. A configuration for delaying a signal detected at the ear position of the listener when identifying a filter for converting a speaker output sound detected by the error microphone into a speaker output sound at the ear position of the listener. The active silencer according to claim 3, wherein: 5. When the detection of the signal output from the speaker at the ear position of the listener is performed earlier than the detection of the signal output from the speaker by the error microphone, the distance between the speaker and the ear of the listener is determined. The detection signal is detected by the error microphone by delaying the detection signal in time and using a filter representing the sound field characteristic between the speaker and the error microphone and a filter representing the sound field characteristic between the speaker and the ear of the listener. 4. The active silencer according to claim 1, wherein a filter that converts the speaker output sound into the speaker output sound at the ear position of the listener is identified. 6. The error microphone according to claim 1, wherein the distance between the listener's ear and the speaker is equal to or greater than the distance between the speaker and the error microphone. The active silencer according to claim 3. 7. A filter representing a sound field characteristic between a reference sensor for detecting noise and an ear of a listener, a filter representing a sound field characteristic between the reference sensor and an error microphone for performing error detection, and an error microphone. An active silencer incorporating a filter configured to obtain the detected noise using a filter that converts the detected noise into noise at the ear position of the listener. 8. The active silencer according to claim 7, wherein a filter for converting noise detected by the error microphone into noise at the ear position of the listener has been identified in advance. 9. When identifying a filter that converts noise detected by an error microphone into noise at the ear position of a listener,
8. The active silencer according to claim 7, wherein a signal detected at the ear position of the listener is delayed. 10. When the detection of the signal from the reference sensor position at the ear position of the listener is performed earlier than the detection of the signal from the reference sensor position by the error microphone, the ear position of the listener from the reference sensor position. The detection signal at time is delayed in time, and a filter representing a sound field characteristic between the reference sensor and the error microphone, and a filter representing a sound field characteristic between the reference sensor and the ear of the listener receive an error. 10. The active silencer according to claim 7, wherein a filter that converts noise detected by the microphone into noise at the ear position of the listener is identified. 11. The original sound to be silenced is used as a signal for identifying in advance a filter for converting noise detected by an error microphone into noise at the ear position of a listener. The active silencer according to any one of claims 10 to 13. 12. A filter for outputting a random signal from the vicinity of a noise source and identifying in advance a filter for converting noise detected by an error microphone into noise at the ear position of a listener. The active silencer according to claim 10. 13. The error microphone is arranged at a position where the distance between the listener's ear and the reference sensor is equal to or greater than the distance between the reference sensor and the error microphone. An active silencer according to claim 7 or claim 8.