JPH09288489A - Vehicle indoor noise reducing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vehicle indoor noise reducing device which is small in size and is of a low cost and has a good followup ability with respect to the muffling of a vehicle indoor noise and is capable of obtaing a sufficient noise reduction effect in accordance with the noise of a wide frequency area in the space of a vehicle. SOLUTION: This vehicle indoor noise reducing device is disposed at a position being more upstream side than a control point 6 where a frequency becoming correlative relation with the noise characteristic of the nieghborhood of a noise control point 6 in the cabinet of a vehicle and is provided with a vehicle noise detecting means 2 detecting the frequency, a sound wave outputting means 7 generating a sound having a phase reverse to that of noise in the vicinity of the control point 6 toward the cabin of the vehicle 1, an error signal detecting means 4 which is installed in the vicinity of the sound wave outputting means 7 and which generates an error signal Error by detecting the residual noise in the cabin of the vehicle 1 and a control means 14 having an adaptive filter 22 controlling the reverse phase sound so that the error signal Error becomes minimum based on a reference signal Ref from the vehicle indoor noise detecting means 2 and the error signal Error from the error signal detecting means 4.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a noise reduction device in a specific space such as a vehicle interior.

[0002]

2. Description of the Related Art Various types of noise reduction devices in a space called an active noise control system have been proposed.
In the invention described in Japanese Patent Application Laid-Open No. 6-228280, in which this device is applied to a vehicle or the like, the noise source is an input system such as an engine or a suspension, and the vibration sound thereof is detected by a microphone or the like, or an ignition pulse signal or It is detected by the crank angle signal and input to the control means as a reference signal. A noise microphone is provided in the vehicle compartment to detect the noise state near the control point and input it to the control means as an error signal. The control means outputs the input reference signal and error signal to the vehicle interior as a secondary sound from the secondary sound source output means (speaker) via an adaptive algorithm such as LMS and an adaptive filter to control the vehicle interior. The specific frequency at the point is converged to control the indoor noise to be small. In other words, the vehicle interior noise reduction device reduces the vehicle interior noise by generating sounds with the same amplitude and opposite phase (anti-phase sound) with respect to the noise (periodic sound component) of the specific frequency band in the vehicle interior from the speaker. It is supposed to do.

Further, Japanese Patent Laid-Open No. 5-61485 discloses that
A noise reduction device that changes the convergence coefficient of an adaptive filter based on a reference signal input to a control unit is disclosed.
Japanese Patent No. 95794 discloses a noise reduction circuit that changes the update rate of the filter coefficient of the adaptive filter according to a non-steady voice signal or a steady noise signal input to the control means.

[0004]

In the vehicle interior noise reduction device and circuit described above, the reference signal is detected from the noise input system, so the noise measured there is noise (frequency) at the control point in the vehicle interior. Not necessarily a characteristic. The position where the occupant hears the noise is in the vicinity of the seat in the passenger compartment, and the noise at that position is the fixed portion of the vehicle body or the like that is the noise transmission system, air propagation or air column resonance, or the vehicle interior that is the radiation system. It is what is heard at the control point through the surrounding panel etc. In other words, there is a temporal function between the noise generated in the noise source and the noise heard in the passenger compartment, and the correlation between the two is low, so the characteristics of the passenger compartment noise and the noise in the input system differ. It is difficult to reduce noise. If the vehicle interior noise is a periodic sound caused by engine vibration, the source is relatively easy to identify, so passive noise reduction measures can be taken and the noise reduction device described above can also be expected to reduce noise to some extent. It is difficult to obtain a sufficient silencing effect if the frequency characteristic is an aperiodic sound that greatly changes depending on the environment, such as wind noise or wind noise.

If the control time is not limited, the coefficient of the adaptive filter is sequentially updated to an appropriate coefficient by the LMS algorithm, so that it is possible to generate an antiphase sound with respect to the noise. The adaptive filter does not follow well with respect to, and it takes time to sufficiently reduce noise. Further, when the noise of the input system, which is a periodic sound, is used as a reference signal to generate an antiphase sound that is a secondary sound to the noise, a large amount of information is detected from the input system in order to improve reliability, Complex control processing is required by providing an adaptive filter corresponding to the noise component to be silenced (periodic sound or non-periodic sound), and the size of the control means becomes large and the control means with high arithmetic processing capability becomes necessary, resulting in high cost. I will end up.

On the other hand, in the invention disclosed in Japanese Patent Laid-Open No. 5-61485, the filter coefficient of the adaptive filter is changed according to the change of the reference signal from the input system input to the control means, so that the followability of the filter is improved. A secondary sound that corresponds to a change in noise can be output from the speaker well, but in this case as well, since the periodic signal having a low correlation with the vehicle interior noise and a signal similar thereto are used as the reference signal, the accuracy of the reference signal is rough and the control means It is considered that the number of basic calculations in (1) increases and a control delay is included, and only noise caused by periodic sounds can be reduced. Further, the configuration described in this publication requires hardware means for detecting the amount of change in noise of the input system.

On the other hand, in the invention described in Japanese Patent Laid-Open No. 6-95694, the update rate of the filter coefficient of the adaptive filter is determined by determining whether it is a non-steady voice signal or a steady noise signal input to the control means. Since it changes the adaptive filter, the secondary sound corresponding to the change in the vehicle interior noise can be output from the speaker, and the update rate of the filter coefficient can also be changed based on the reference signal detected from the input system. Since it is changed, it is considered that the accuracy of the reference signal is rough, the number of calculations in the basic control means is increased, a control delay is included, and only the noise caused by the periodic sound can be reduced. That is, since the reference signal based on the information from the input system that is in the vicinity of the noise source in the conventional technology has a low correlation with the vehicle interior noise,
The frequency range of noise reduction tends to be narrowed.

It is an object of the present invention to provide a vehicle interior noise that is small in size and low in cost, has a good followability with respect to the noise elimination of the vehicle interior noise, and has a sufficient noise reduction effect in response to noise in a wide frequency range in the space. To provide a reduction device.

[0009]

In view of the above, according to the first aspect of the present invention, the control point in the sound transmission path at which a frequency having a correlation with the noise characteristic in the vicinity of the noise control point in the vehicle is generated is higher than that in the control point. A vehicle interior noise detection means arranged on the upstream side to detect the frequency, a sound wave output means for generating the noise and an antiphase sound toward the vehicle compartment near the control point, and a sound wave output means in the vicinity of the sound wave output means. Based on the error signal detecting means installed and detecting the residual noise in the vehicle interior and outputting an error signal, the reference signal from the vehicle interior noise detecting means and the error signal from the error signal detecting means, And a control means having an adaptive filter for controlling the antiphase sound from the sound source output means so that the error signal is minimized.

According to another aspect of the present invention, the control means includes update pitch changing means for changing the coefficient update pitch of the adaptive filter according to the magnitude of the error signal from the error signal detecting means. In the invention according to claim 3, the sound wave output means is a speaker built in the headrest of the vehicle. In the invention according to claim 4, the vehicle interior noise detection means is arranged in the rear pillar space or the rear deck panel, which is a sound transmission path. In the invention according to claim 5,
The reference signal is generated based on aperiodic sounds centered on road noise.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION A vehicle interior noise reduction device according to the present invention is provided on the upstream side of a control point in a sound transmission path where a frequency having a correlation with a noise characteristic in the vicinity of a noise control point in a vehicle interior is generated. The vehicle interior noise detection means is arranged to detect the frequency in the sound transmission path and output the reference signal.
The reference signal is installed in the vicinity of a sound wave output unit that generates a sound having the same amplitude and opposite phase as the noise toward the vehicle interior as an antiphase sound (secondary sound) near the control point, and detects the residual noise in the vehicle interior. Then, the error signal from the error signal detecting means for outputting an error signal is input to the control means having an adaptive filter. The control means controls the antiphase sound from the sound source output means by updating the coefficient of the adaptive filter using an adaptive algorithm so that the error signal is minimized based on the reference signal and the error signal. In this way, the vehicle interior noise detection means,
A frequency that correlates with the noise characteristics near the control point of the noise inside the vehicle is generated.If it is placed upstream of the control point in the sound transmission path, the frequency that correlates with the noise characteristic near the control point will be detected. Therefore, the reference signal can be made closer to the noise characteristic near the control point, and the accuracy of the reference signal can be improved.

If the control means is provided with an update pitch changing means for changing the coefficient update pitch of the adaptive filter in accordance with the magnitude of the error signal from the error signal detecting means, the coefficient update pitch is changed in accordance with the magnitude of the error signal. Can be set. For example, when the noise is large, the coefficient update pitch is updated greatly,
If the coefficient update pitch is set to be small when the noise is small, the difference in the noise reduction time (convergence time) due to the size of the noise becomes small.

If the sound wave output means is a speaker built in the headrest, the antiphase sound can be output from the vicinity of the occupant who hears the noise, so that it is possible to avoid making a loud sound and to reach the occupant of the antiphase sound. Can be shortened, and only the noise heard by passengers can be reduced.

When the vehicle interior noise detection means is arranged in the rear pillar space or the rear deck panel which is a sound transmission path, a reference signal based on an aperiodic sound such as road noise which is noise that can be observed in the rear pillar or the rear deck panel. Can be output.

Further, if the noise in the vehicle compartment is mainly a non-periodic sound such as road noise, this non-periodic sound changes variously depending on the running state of the vehicle. The noise range (frequency range) that can be dealt with per hour becomes wider, and the noise reduction range becomes wider.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS In a vehicle interior noise reduction device as an embodiment of the present invention, a first microphone 2 serving as a vehicle interior noise detection means for outputting a reference signal Ref correlated with the noise in the vehicle interior 1 shown in FIG.
Are arranged in the rear pillar 3 that constitutes the radiation system of the noise D that enters the vehicle interior 1. As shown in FIG. 4, the first microphone 2 is inserted into the inside of the pillar through an opening 3A provided in the lower portion of the rear pillar 3, and is fixed inside the rear pillar 3 by an appropriate means such as a highly adhesive tape. The first microphone 2 may be attached using a bracket made of sheet metal or resin, but in this case, it is important to attach so that the bracket does not resonate.

The noise D entering the vehicle interior 1 shown in FIG.
The main source of noise is road noise that enters from the trunk room 9 through the rear pillar 3. Road noise is a non-periodic sound that greatly depends on the running condition of the vehicle and the road surface condition.
It occurs in the range of 0 Hz. In addition to the above, the aperiodic sound includes wind noise starting from about 500 Hz.
As described above, the basic configuration is that the first microphone 2 arranged in the radiation system detects the aperiodic sound and uses it as the reference signal Ref for reducing the noise in the vehicle interior 1. That is, the first microphone 2 takes in noise (frequency) in the rear pillar 3 and outputs the reference signal Ref.

Further, the noise that enters the vehicle interior 1 includes a periodic sound X transmitted from an input system such as an engine. The periodic sound X becomes a noise in the vehicle interior 1 mainly through a vehicle body fixing portion that serves as a transmission system, air propagation, air column resonance, or a panel that surrounds the vehicle interior 1 that serves as a radiation system. The noise in the vehicle interior 1 in the present embodiment refers to a sound in which the periodic sound and the aperiodic sound near the rear seat 5 arranged in the vehicle interior 1 are combined.

The headrest 6 of the rear seat 5 incorporates a second microphone 4 as an error signal detecting means for detecting the residual noise in the passenger compartment 1 and outputting an error signal Error. Inside the headrest 6 located behind the second microphone 4, a speaker 7 that serves as a sound wave output unit that outputs a secondary sound that is an antiphase sound toward the vehicle interior 1 is built in. The secondary sound is composed of a sound wave signal having a frequency characteristic of basically the same amplitude and opposite phase as the residual noise in the vehicle interior 1 detected by the second microphone 4, and is designed to cancel the noise in the vicinity of the control point. There is. The residual noise is the noise in the vehicle interior 1 that cannot be canceled out by the secondary sound.

The control point is an arbitrary point in the vehicle interior 1 space apart from the vehicle body panel 1A that surrounds the vehicle interior 1 that serves as a radiation system, and is the installation position of the second microphone 4 here. The vicinity of the control point means a space around the headrest 6, and in this embodiment, the space K around the head M of the occupant shown in FIG.

That is, in the present embodiment, when a tire (not shown) is used as a noise source, a frequency having a large correlation with the noise (frequency characteristic) in the vicinity of the control point is generated in the rear pillar 3 from the trunk room 9 in FIG. It is used as a sound transmission path. The position of the second microphone 4 is regarded as a control point in the sound transmission path, and the first microphone 1 serving as vehicle interior noise detection means is arranged in the rear pillar 3 serving as an area on the upstream side of the sound transmission path. A configuration is adopted in which a speaker 7 serving as a sound wave output unit is installed near the two microphones 4.

The first microphone 2 has a low-pass filter (LPF) 10, A / A as shown in the system block diagram of FIG.
The D converter 11 constitutes a reference signal input unit 12.
In the reference signal input unit 12, the noise D inside the rear pillar 3 detected by the first microphone 2 is D
C component and unnecessary low frequency range are cut, and A / D converter 1
1 to digital signal, power level calculation means 1
It is inputted to the control means 14 as a reference signal Ref via the signal No. 3.

The second microphone 4 is a low-pass filter (LP
F) 15 and A / D converter 16 constitute an error signal input section 17. In the error signal input section 17, the second microphone 4
The analog residual noise in the vicinity of the loudspeaker 7 detected by is cut by the low-pass filter 15 in the DC component and unnecessary low frequency range, converted into a digital signal by the A / D converter 16, and input to the control means 14 as an error signal Error. The loudspeaker 7 constitutes a control signal output unit 20 together with the D / A converter 18 and the low pass filter (LPF) 19. In the control signal output section 20, the digital signal output from the control means 14 is converted into an analog signal by the D / A converter 18, filtered by the low pass filter 19, and then transmitted as the control signal Y to the speaker 7. The speaker 7 receives the control signal Y and outputs a secondary sound to the vehicle interior 1. A control switch 21 for selecting transmission of the control signal Y to the speaker 7 is arranged between the D / A converter 18 and the low-pass filter 19. Control switch 21
Is connected to a circuit connecting the control means 14 and the speaker 7 when turned on, and cuts off this circuit when turned off.

The power level calculating means 13 integrates the vector PWR (N-1) of the reference signal Ref detected by the first microphone 2 to obtain the power level PWR (N) per unit time as the instantaneous value of the reference signal Ref. The values are averaged and calculated by the following formula.

PWR (N) = PWR (N-1) · α + SIG (N-1) · SIG (N) β (13) SIG (N); current reference signal value SIG (N-1); Previous reference signal value α; set value β; (1-α) The control signal output unit 20 is set with an upper threshold value and a lower threshold value in consideration of the voice of a passenger in the passenger compartment 1, audio sound, and the like. When the calculated power level PWR (N) is out of this threshold range, the control signal Y is not output to the speaker 7. The lower threshold in this embodiment is 30d
B, and the upper limit threshold is set to 70 dB. In other words, it is set not to intentionally reduce the noise when the noise in the passenger compartment 1 becomes large due to the voices and audio sounds of the occupants, which are noises other than the noise generated when the vehicle travels, or when the passenger compartment 1 is quiet. ing.

Although two thresholds are provided here, only the lower limit threshold and the upper limit threshold are set, or the upper limit and the lower limit threshold are set based on the calculated power level PWR (N), and the power level PWR is set. You may make it control by always a fixed width according to the state of (N).

The control means 14 is a controller composed of a well-known microcomputer, and receives the reference signal Ref from the first microphone 2 and the error signal Error from the second microphone 4 to minimize the error signal Error (preferably the error signal Error. 0), the control signal calculation unit 24 mainly includes an adaptive filter 22 having a filter coefficient W for controlling the secondary sound output from the speaker 7 and a well-known adaptive algorithm 23.
And the adaptive filter 22 according to the level of the error signal Error.
Update pitch changing means 25 for changing the update pitch Mu, which is the update speed of the filter coefficient W. As the adaptive algorithm 23 in this embodiment, the least square error estimation method (LMS method) is used because the arithmetic processing is easy, but other adaptive algorithms may be used. The adaptive filter 22 is adapted to update the filter coefficient W thereof in a timely manner by the adaptive algorithm 23.

In this embodiment, as shown in FIG. 3, the transfer function C from the speaker 7 to the control point (the second microphone position) in the vehicle compartment and the speaker 7 to the first microphone are previously set before the noise reduction control is performed. The transfer function H, which is a characteristic of the muffling transfer system including the transfer systems up to 2, is measured to identify the transfer function H.
Specifically, the M-sequence random sound is output from the speaker 7, detected by the first microphone 2 and the second microphone 4, and the transfer function C and the transfer function H of the muffling transfer system are set, respectively.

The control signal calculator 24 of FIG.
The secondary sound output from the second microphone 4 is input to the second microphone 4 and the FIR filter 26 having a predetermined filter coefficient HR to prevent howling from occurring in the second microphone 4. Only the propagation time required for the first microphone 2
And a filter 27 having a predetermined transfer function C ′ that delays the reference signal Ref of 1 in advance by a predetermined time. The filter coefficient HR of the FIR filter 26 is set so that the convolution operation of the control output Y and the transfer function H is Y · H = 0. Although the FIR filter 26 whose signal output is stable is used here, a well-known IIR filter with a small signal output time delay may be applied.

The update pitch changing means 25 changes the update pitch Mu of the filter coefficient W of the adaptive filter 22 according to the change of the error signal Error output from the error signal input section 17 to adjust the noise convergence speed. Is. Specifically, the vector E_PWR (N-1) of the error signal Error is integrated to obtain the power level E_PWR (N) per unit time,
The instantaneous values of the error signal Error are averaged, and the value is used to determine the angle change gradient ΔE of the error signal Error per unit time, that is, the update pitch μ of the adaptive filter 22 to be updated. The power level E_PWR (N) of the error signal Error is obtained by Equation 1, and the angle change gradient (update pitch μ of the latest adaptive filter) ΔE is obtained by Equation 2. Equation 1,
In Equation 2, convolution operation processing is performed respectively.

E_PWR (N) = E_PWR (N-1) · α + err (N-1) · err (N) · β Equation 1 err (N); current error signal err (N-1); previous time Error signal α; set value β; (1-α) ΔE = Mu · E_PWR · G equation 2 G; constant err; error signal level Mu; update pitch Note that the constant G and update pitch Mu are initially set. I'll do it.

Next, the noise signal D detected by the second microphone 4
Is obtained by convolving the periodic sound X from the input system and the transfer function P of the radiation system and adding the reference signal Ref detected by the first microphone 2. D = X · P + Ref (3) The signal level R of the reference signal Ref is obtained by convolving the control signal Y and the transfer function H from the speaker 7 to the first microphone 2. R = Y · H (Equation 4) The control signal Y is obtained by performing the convolution operation by the equation 5, and the signal level E of the error signal Error is obtained by performing the convolution operation by the equation 6. Y = (X + Ref) * W ... Equation 5 E = D-Y * C ... Equation 6 And the signal level E of this calculated error signal Erro
And the signal level R of the reference signal Ref are subjected to a convolution operation in Expression 7 to calculate the latest filter coefficient W _{N + 1} . W _{N + 1} = W _N + 2 · μ · E · R (Equation 7) W _{N + 1} ; Filter coefficient after updating adaptive filter (latest) W _N ; Latest filter coefficient before updating adaptive filter μ; Update pitch of adaptive filter E; error signal level R; reference signal level As a result, the adaptive filter 22 is sequentially updated to the optimum filter coefficient W that minimizes the signal level E of the error signal Error, and is output to the speaker 7. The control signal Y for controlling is controlled to an optimum signal.

The operation of the vehicle interior noise reduction system will be described with reference to the control flow of the vehicle interior noise reduction system shown in FIG. When an ignition key (not shown) is turned on, the reference signal input unit 12, the control unit 14, the control signal output unit 20, and the error signal input unit 17 are provided.
Is ready for operation. Then, when noise (non-periodic sound) that is correlated with the noise in the vicinity of the rear seat 5 is detected from the first microphone 2 in step S1, the detected noise is used as a reference signal Ref, and the process proceeds to step S2 where the A / D converter 11 performs A / D convert to digital signal.

Next, in step S3, the power level calculating means 13 calculates the power level R_PWR of the reference signal Ref.
Is calculated, averaged, and output to the control means 14. In step S4, howling correction is performed by the FIR filter 26 and feedback is performed, and the process proceeds to step S5. S5
In the step, the filter coefficient W of the optimum filter 22 is set by the filter 27 and the equation 7 of the adaptive algorithm 23, and the control signal Y is calculated. In step S6, the D / A converter 18 of the control signal output unit 18 is set. Is converted to an analog signal.

That is, the reference signal Ref is generated based on the aperiodic sound whose main component is the road noise in the rear pillar 3 that correlates with the noise in the vicinity of the rear seat 5 near the control point. Therefore, the noise transmission time to the control point (the place where the sound is heard) becomes shorter than that in the case where the reference signal is generated based on the periodic sound such as the engine rotation pulse, and the reference signal Ref and the noise in the vehicle interior 1 to be reduced The correlation with
Impulse response in the initial stage of calculation in is improved. This means that the number of filters 22, 26, 27 and the number of taps of the control means 14 are suppressed, which reduces the size of the device and the number of calculations.

A value PWR _{(N) obtained} by averaging the reference signal Ref.
Is compared with the threshold set in step S7. Here, when PWR _(N) is larger than the lower limit threshold of 30 dB and smaller than the upper limit threshold of 70 dB, it is assumed that the noise in the passenger compartment 1 is at the noise level of the control target region, and the state of the control switch 21 is determined in step S8. Is judged. When the control switch 21 is in the ON state, it is considered that there is an intention to reduce noise, and in step S9, the control signal Y is output from the control means 14 to the speaker 7 to output the secondary sound to the vehicle interior 1 (radiation).
I do.

Averaged value PWR _{(N) of the} reference signal Ref
Is smaller than 30 dB set as the lower limit threshold value and larger than 70 dB set as the upper limit threshold value, the control means 14 does not output the control signal Y. That is, the noise of the occupant in the passenger compartment 1 is louder than the noise composed of the aperiodic sound centered on the road noise observed in the passenger compartment 1 and the periodic sound from the engine (not shown). When the voice of a person is heard loud or when the noise level is lower than the sound field environment of the vehicle itself, the control signal Y
Is not output and noise reduction is not executed.

On the other hand, when the secondary sound is output from the speaker 7, the second microphone 4 of the error signal input section 17 detects this secondary sound in step S11 and outputs the error signal Error. The error signal Error at this time is converted into a digital signal by the A / D converter 16 and input to the update pitch changing means 25 and the adaptive algorithm 23.

In the update pitch changing means 25, the power level E_PWR of the error signal Error is calculated and averaged by the equation 1 in step S3, and the update pitch (convergence pitch) Mu of the filter coefficient W input in advance in step S12.
And a fixed value G are read out, and the angle change gradient ΔE per unit time according to the power level of the error signal Error, that is, the latest update pitch μ is obtained.

Here, if the angle change gradient ΔE is large,
When the change pitch Mu is updated quickly and the power level E_PWRef (N) of the error signal E is small, the angle change gradient Δ
E is small and the changed pitch Mu is updated later. Therefore, when the error signal level E is large, the update pitch of the filter coefficient W of the adaptive filter 22 is greatly changed and the convergence time is short. Therefore, even if the error signal level changes greatly, the muffling ratio can be the same, and noise can be reduced. There is no sense of discomfort with respect to the degree of convergence of. In particular, when the noise is large, the coefficient update pitch Mu is greatly updated, so the response of the adaptive filter 22 is improved, the noise reduction speed is increased, and control delay can be prevented.

That is, when the changing pitch Mu of the filter coefficient W of the adaptive filter 22 is changed according to the level of the error signal Error from the second microphone 4 serving as the error signal detecting means provided in the headrest 6, the noise of the engine or the like is changed. An extremely fine and quick update pitch control is possible as compared with changing the update pitch Mu of the filter coefficient W according to the level of the reference signal generated based on the periodic sound from the input system. Further, the latest update pitch μ represented by the angle change gradient ΔE can be automatically updated based on the conventional update pitch by convolving the update pitch (convergence pitch) Mu obtained so far with an arbitrary constant G.

Next, the relationship between the characteristics of the speaker 7 and noise reduction will be described. The speaker 7 used in this embodiment is
As shown in FIG. 7, a thickness of 20
It is provided inside the headrest 6 provided with the urethane material 6B of not less than mm. The speaker 7 is disposed on the front surface 70a of the speaker box 70 made of wood, toward the front of the vehicle indicated by the arrow. The noise measurement area is shown in Fig. 6.
Area K of 800 mm square centered on the head M shown in
Then, the data of the specific frequency (400 Hz) at which the secondary sound was output from the speaker 7 was measured.

In FIG. 8, a speaker 7 is attached to the headrest 6.
FIG. 9 shows changes in phase and amplitude in the area K when three speakers are provided, and FIG. 9 shows changes in phase and amplitude when two speakers are provided in the headrest 6. White areas indicate noise reduction areas, and hatched areas indicate sound increase areas. It is apparent from FIGS. 8 and 9 that it is possible to reduce the phase and amplitude changes as the number of speakers 7 increases, and the noise reduction area due to the secondary sound increases. That is, since the noise reduction area near the control point is stable, a certain range is ensured even when the occupant moves and the position of the head M changes, and quietness is enhanced.

10 (a), 10 (b) and 11 (a),
(b) shows a phase change and an amplitude change in the area K depending on the mounting angle of the speaker 7. 10 and 11
Since there is little change in the phase and the amplitude depending on the mounting angle of the speaker 7, it is possible to provide an allowable range for the mounting angle of the speaker 7. Further, since the speaker 7 of this embodiment is built in the headrest 6, the speaker 7 does not come into direct contact with the occupant, the injury of the occupant due to the speaker 7 is extremely small, and the headrest 6 is miniaturized.

FIG. 12 is a diagram showing the relationship between the diameter of the speaker 7 and the lowest resonance frequency, and FIG. 13 is a diagram showing the relationship between the box capacity of the speaker 7 and the lowest resonance frequency. It is clear from FIGS. 12 and 13 that the minimum resonance frequency decreases as the speaker diameter and the box volume increase. However, if the speaker diameter is too large or the volume is too large, the headrest 6 cannot or cannot be built into the headrest 6. The headrest 6 becomes larger, which causes problems in appearance and rear visibility.

FIG. 14A shows the output sound pressure characteristic of the speaker 7, and FIG. 14B shows the power consumption and the frequency characteristic of the speaker 7. In FIG. 14 (a), a thick solid line indicates a noise level equivalent to the vehicle interior noise when the third speed is fully open, and a thin solid line indicates an output level of the secondary sound from the speaker 7 when 0.8 W of electric power is supplied. Is. In this embodiment, when the speaker diameter is set to 80 mm and the box capacity is set to 2000 cc, the minimum resonance frequency is 200H.
Although it is near z, it can be seen that the speaker performance is improved by increasing the power consumption. In particular, as shown in FIG. 14A, even when the power consumption is 0.8 W, it is possible to cover the noise level when the third speed is fully opened and the frequency range at that time.

In addition to the above-mentioned characteristics, the speaker 7 also has a requirement of a response which is the time from when the control signal Y is received to when the secondary sound is actually produced. If this response is not good, no matter how much the control system appropriately selects the filter coefficient W of the adaptive filter 22 or the coefficient update pitch Mu is rapidly updated to shorten the processing time and output the control signal Y. However, it is important to select a speaker with a good response because the response will be delayed and the noise reduction action will be delayed.

The effect of the vehicle interior noise reduction device having such a configuration will be described with reference to FIG. FIG. 15 shows a measurement of the noise level in the passenger compartment 1 when the vehicle travels on a pitch road at 40 km / h. In the figure, the solid line shows the off state of the noise reduction device, and the broken line shows the on state of the noise reduction device. According to the measurement, the noise reduction is particularly observed at around 100 Hz and 200 Hz, but the noise reduction effect is generally seen in the frequency range up to 500 Hz.

It is considered that the reason why the noise around 100 Hz and 200 Hz is particularly reduced is that the road noise is used as the reference signal Ref. If the reference signal is made using a periodic sound from an input system such as an engine, the change in the reference signal per unit time is small, so noise in the vicinity of the frequency of the periodic sound is periodically reduced, which reduces noise. The area is limited. In the present embodiment, the reference signal Ref is output based on the irregular aperiodic sound represented by road noise, so that the change width of the reference signal Ref per unit time is large and the noise reduction region is wide.

Since the rear pillar 3 is close to the exhaust muffler (not shown) of the vehicle, it also serves as a transmission path for the periodic sound of the engine. Therefore, by disposing the first microphone 2 inside the rear pillar 3, the interior of the vehicle from the engine side can be improved. It is possible to detect a periodic sound having a higher correlation than the periodic sound that invades 1. For example, when a four-cylinder engine is rotating at 3000 rpm, its periodic sound is around 100 Hz.
It is possible to reduce the periodic sound component that constitutes the noise around 100 Hz in the noise.

In the embodiment described above, the first microphone 2 is shown in FIG.
As shown in FIG. 3, the noise inside the rear pillar 3 is used as the reference signal Ref. However, the vibration of the rear deck panel 28 is detected by means of an exciter or the like instead of the first microphone 2 and used as the reference signal. Alternatively, as shown in FIG. 16, a noise reduction device may be configured by providing a microphone 30 serving as vehicle interior noise detection means in a high mount stop lamp 29 provided on the rear deck panel 28. Since the rear deck panel 28 is arranged between the trunk room 9 and the vehicle interior 1, the road noise, which is an aperiodic sound, serves as a transmission path of sound that enters the vehicle interior 1 from the trunk room 9. The reason why the microphone 30 is provided in the high mount stop lamp 29 is that sounds other than road noise, such as voices of passengers and audio sounds, are not detected as much as possible.

As described above, when the microphone 30 having the same function as that of the first microphone 2 is provided in the stop lamp 29 on the rear deck panel 28, the road noise which is an aperiodic sound and the exhaust sound which becomes a periodic sound from the exhaust muffler 31 are provided. It is possible to detect a noise, such as a sound, which has a high correlation with the noise in the vicinity of the headrest 6 which is the control point in the vehicle interior 1 and output it as the reference signal Ref.

Therefore, as in the above-mentioned embodiment, the reference signal R
Since the accuracy of ef is improved, the accuracy of the control signal Y is also improved, and the noise region to be controlled is widened. In addition, since the signal accuracy is good, the update of the filter coefficient W of the adaptive filter 22 is reduced, the arithmetic processing load on the control unit 14 can be reduced, the control unit 14 can be prevented from increasing in size and cost, and noise can be reduced. It can be carried out.

In addition, even in this case, the changing pitch Mu of the filter coefficient W of the adaptive filter 22 is changed according to the level of the error signal Error from the second microphone 4 serving as the error signal detecting means provided in the headrest 6. Finer update pitch control is possible than when the update pitch Mu of the filter coefficient W is changed according to the level of the reference signal generated based on the periodic sound from the noise input system such as the engine. That is, the accuracy of the reference signal Ref is improved by
This is because the accuracy of the error signal Error is also improved.

In this embodiment, the noise reduction control near the headrest 6 of the rear seat 5 on one side of the vehicle shown in FIG. 1 is performed, but the headrest 6'of the left rear seat 5'is performed.
Alternatively, the second microphone 4'and the speaker 7'having the same structure as the right rear seat 5 may be built in, and the first microphone 2'may be provided in parallel in the left rear pillar 3 '. In this case, a control means is separately provided for each of the left and right seats 5 and 5'and each speaker 7,
The secondary sound output from 7'may be controlled, or both of the speakers 7,
7'may be subjected to well-known multipoint control.

Further, in the noise reduction device of this embodiment, the noise reduction on the rear seat side has been mainly described, but it is needless to say that the noise reduction device may be used for noise reduction on the front seat side not shown. In this case, the location of the noise (non-periodic sound) that correlates with the noise near the front seat side is actually measured and found, and a microphone or a vibration exciter that serves as vehicle interior noise detection means is placed at that location. It is advisable to control the secondary sound output from the speaker that is the sound wave output unit that outputs the reference signal Ref that is correlated with the noise of 1. In this case as well, of course, a microphone serving as an error signal output means is provided in the headrest on the front seat side near the control point, and the error signal Error in the vehicle interior 1 is fed back to the control means and used for calculating the filter coefficient W of the adaptive filter 22. At the same time, the update pitch Mu of the adaptive filter 22 can be changed according to the error signal Error by calculating the filter update pitch using Expressions 1 and 2 described above, and the noise reduction response can be improved.

[0057]

According to the first aspect of the present invention, the vehicle interior noise detecting means is provided with a control point in a sound transmission path at which a frequency having a correlation with a noise characteristic near a noise control point in the vehicle interior is generated. By arranging it on the upstream side, the frequency that correlates with the noise characteristic near the control point is detected, so the reference signal can be made closer to the noise characteristic near the control point, and the accuracy of the reference signal is good. Become. As a result, the precision of the reference signal becomes finer, the reliability increases, the number of control calculations decreases, and the coefficient of the adaptive filter can be updated with a small device at low cost without using expensive control means, and noise in the vehicle interior is stabilized. Can be reduced.

According to the second aspect of the present invention, since the update pitch of the coefficient of the adaptive filter can be set in correspondence with the magnitude of the error signal from the error signal detecting means, it can be set short regardless of the level of the error signal. The error signal can be attenuated to the minimum level in time, and the response delay of noise attenuation control can be extremely reduced.

According to the third aspect of the present invention, the sound wave output means is the speaker built in the headrest, so that the antiphase sound can be output from the vicinity of the occupant who hears the noise, and a loud sound is not generated. The time until the antiphase sound reaches the occupant is shortened, and only the noise heard by the occupant can be reduced.

According to the fourth aspect of the present invention, the vehicle interior noise detecting means is arranged in the rear pillar space or the rear deck panel, which is a sound transmission path, respectively, so that the road noise that can be detected in the rear pillar or the rear deck panel is detected. Since the reference signal based on the non-periodic sound such as noise is output, the frequency region controlled per unit time can be widened as compared with the periodic sound, and the noise reduction region can be stabilized and expanded.

According to the fifth aspect of the present invention, the noise in the passenger compartment is reduced mainly by using non-periodic sounds such as road noises that change variously depending on the running state of the vehicle. The frequency range that can be handled per unit time is wider than that of the above, and the noise reduction region can be wide and stable.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a vehicle interior noise reduction device showing an embodiment of the present invention.

FIG. 2 is a system block diagram of a vehicle interior noise reduction device showing an embodiment of the present invention.

FIG. 3 is a block diagram illustrating a control algorithm of the present invention.

FIG. 4 is an enlarged perspective view of the vicinity of a mounting portion of the first microphone.

FIG. 5 is a flowchart showing an example of a control procedure.

FIG. 6 is a plan view showing an example of a noise reduction area according to the present invention.

FIG. 7 is a partially cutaway perspective view showing a configuration of a headrest and a mounting state of a speaker.

FIG. 8 is a diagram showing a change in a noise reduction region when one speaker is used.

FIG. 9 is a diagram showing a change in a noise reduction area when two speakers are used.

FIG. 10 is a diagram showing a phase change of a noise reduction region depending on a mounting angle of a speaker.

FIG. 11 is a diagram showing a change in amplitude of a noise reduction region depending on a mounting angle of a speaker.

FIG. 12 is a diagram showing a relationship between a speaker aperture and a speaker minimum resonance frequency.

FIG. 13 is a diagram showing a relationship between a speaker box and a speaker minimum resonance frequency.

14A is a diagram showing a relationship between a noise frequency and a noise level, and FIG. 14B is a diagram showing a relationship between a frequency generated from a speaker and power consumption.

FIG. 15 is a measurement diagram showing the noise reduction effect of the present invention.

FIG. 16 is a perspective view showing a modified example of the mounting position of the first microphone.

[Explanation of symbols]

 1 vehicle interior 2, 2'30 vehicle interior noise detection means 3, 3'rear pillar 4, 4'error signal detection means 6 vicinity of control point (headrest) 7,7 'sound wave output means (speaker) 14 control means 22 adaptive filter 25 Updating pitch changing means 28 Rear deck panel Ref Reference signal Error Error signal W Adaptive filter Mu Coefficient updating pitch

Claims (5)

[Claims] 1. A vehicle interior noise that is arranged upstream of the control point in a sound transmission path in which a frequency having a correlation with a noise characteristic in the vicinity of a noise control point in the vehicle interior is detected and detects the frequency. A detection means, a sound wave output means for generating the noise and the anti-phase sound toward the vehicle compartment near the control point, and a sound wave output means installed near the sound wave output means for detecting the residual noise in the vehicle compartment and making an error. Based on the error signal detecting means for outputting a signal, the reference signal from the vehicle interior noise detecting means, and the error signal from the error signal detecting means, the error signal from the sound source output means is minimized so as to minimize the error signal. A vehicle interior noise reduction device comprising: a control unit having an adaptive filter for controlling an antiphase sound. 2. The control means comprises update pitch changing means for changing the coefficient update pitch of the adaptive filter in accordance with the output state of the error signal from the error signal detecting means. The vehicle interior noise reduction device described. 3. The vehicle interior noise reduction device according to claim 1, wherein the sound wave output means is a speaker built in a headrest of the vehicle. 4. The vehicle interior noise reduction device according to claim 1, wherein the vehicle interior noise detection means is arranged in a rear pillar space or a rear deck panel which serves as a sound transmission path. 5. The reference signal is generated based on an aperiodic sound centered on road noise.
The vehicle interior noise reduction device described.