JPH0758223B2 - Method of measuring sound deadening / damping effect, measuring device, and signal source search device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for measuring a sound deadening / damping effect, a measuring device and a signal source exploration device, and more particularly to a method for measuring a sound deadening / damping effect by an electronic sound deadening system or an electronic control system, and a measuring device. The present invention relates to a signal source search device.

[0002]

2. Description of the Related Art In recent years, a sound wave having a phase opposite to that of a sound wave from a noise source and having the same sound pressure is generated in a region where the sound wave can propagate in a three-dimensional direction, and a predetermined region in the propagation region is generated. An electronic muffling system has been proposed that muffles sound by the sound wave interference. In this type of electronic muffling system, an additional sound having a phase opposite to that of the noise and having the same sound pressure is generated from a speaker in a predetermined region to be muffled, and the drive signal for driving this speaker detects the noise. It is created by an adaptive digital filter based on an input from a sensor microphone or the like and an error output of an error sensor that detects an interference sound of noise and an additional sound in a predetermined area to be muted.

FIG. 10 shows the basic configuration of a conventional electronic silencing system. A noise signal n (n) is detected by a sensor microphone or the like, and its output signal x (n) is generated by an adaptive digital filter 1 and a reference signal generator. Added to Part 2. Incidentally, T (z) is a transmission characteristic of a sensor microphone or the like. By the way, the adaptive digital filter 1 can be realized by an FIR filter having a variable tap weight (filter coefficient) and an adaptive algorithm for controlling the FIR filter, and the adaptive algorithm has an input x (n) and an error output e (n). From the information, the filter coefficient of the adaptive digital filter is adjusted so that the energy of the error output e (n) is minimized under some evaluation criteria.

Next, a method for setting the filter coefficient to the optimum value will be described. The output y (n) of the adaptive digital filter 1 is given by the convolution of the input x (n) and the filter coefficient w _i . The error output e (n) can be expressed as Can be expressed as In the equation (2), r (n) is a filtered reference signal, Is.

[0005] The following vector representation for simplicity, R = [r (n) , r (n-1), ... r (n-I + 1)] T W = [w O, w 1, ... w I- ₁ ] ^T , the above formula (2) can be represented by the following formula: e (n) = d (n) + ^RT · W (4)

Here, the root mean square error (MSE: mean-squ)
are error) E [e (n) ² ] is obtained from the equation (4), J = E [e (n) ² ] = E [d (n) ² ] + 2W ^T E [R ^T d (n) ] + W ^T E [R ^T R] W (5), and MSE becomes a quadratic function of the filter coefficient. The second derivative is the first, and if the derivative is set to 0, the minimum value J
A solution with _min is obtained.

In the FX algorithm (Filtered-x LSM algorithm) which is the steepest descent method algorithm, the instantaneous squared error e (n) ² itself is used as the estimator of MSE J, and the derivative of J (gradient ∇ ), The estimator ∇ _n of Then, the filter coefficient of the adaptive digital filter is recursively updated by the following equation using ∇ _n . W _{n + 1} = W _n + μ (−∇ _n ) = W _n −2 μR _n ^T e (n) (7) where μ is a positive scalar and is a parameter for controlling the magnitude of the correction amount in each iteration. Is. The above equation (7) means that the filter coefficient is sequentially updated in the direction opposite to the gradient vector (∇ _n ) (in the direction of the steepest descent of the error surface), and if this is continued, the MSE will finally reach the minimum value J. It reaches _min , and the filter coefficient has an optimum value.

In FIG. 10, the adaptive digital filter 1 performs a convolution operation of the signal x (n) and the given filter coefficient W (z), and outputs a speaker drive signal y (n). In the figure, d (n) is a desired response of the noise signal n (n) in the error sensor 3, and G (z) is a transmission characteristic from the noise source to the error sensor 3. Further, d '(n) is a desired response of the drive signal y (n) in the error sensor 3, and C (z) represents a transmission characteristic from the speaker to the error sensor 3.

The error sensor 3 detects the error signal e (n) from the d (n) and d '(n), and outputs the error signal e (n).
It outputs to the control unit 4 which updates the filter coefficient based on the algorithm. The reference signal r (n) is applied to the other input of the control unit 4. The reference signal r
(n) is obtained by the convolution operation of the signal x (n) and the filter coefficient C ′ (z) indicating the transmission characteristic from the speaker to the error sensor 3.

It should be noted that the above-mentioned algorithm uses the error output e
Although (n) is only one, a plurality of error sensors are provided in order to widen the predetermined area to be muted, and the same algorithm is applied to the case where the output e (n) is plural. There is one described in Japanese Patent Laid-Open No. 1-501344. In addition, an electronic vibration damping system has also been proposed, in which vibration propagating from a vibration source is damped in a predetermined region similarly to the electronic silencing system.

[0011]

By the way, when the above-mentioned electronic silencing system or electronic damping system is applied to a sound field or a vibration field, there is a strong correlation with the signal observed in the target sound (vibration) field. How to detect the noise source (or vibration source) information in the primary sound (vibration) field and how to extract accurate information are extremely important in order to improve the silencing and damping effect.

Further, if it is possible to estimate in advance how much the sound deadening amount or the vibration damping amount can be obtained when the electronic muffling system or the electronic vibration damping system is applied to the sound field or the vibration field, the electronic muffling system or the electronic vibration damping It is useful for deciding whether or not to introduce a vibration system, and also promotes the construction of an electronic silencing system or an electronic damping system. Conventionally, in a sound field or a vibration field having stationary and irregular noise source (or vibration source) information, multicoherence (m
It is known that if the ultimate coherence function) is known, the upper limit of the sound deadening amount or the vibration attenuation amount (the upper limit of the performance of the electronic silencing system or the electronic damping system) can be estimated.

As a means for obtaining this multicoherence, a sequential solution method in the frequency domain is known. However, in this method, since the Fast Fourier Transform (FFT) is used, it is necessary to assume the stationarity of noise source (or vibration source) information. Also, since it is calculated in block units, 1
In the case of multiple channels such as 6 channels and 32 channels, the amount of calculation becomes enormous and it becomes impossible to deal with it at the personal computer level, and it is virtually impossible to measure multi-coherence in the field. Furthermore, although this is an argument on the assumption that the input information is stationary, the noise source (or vibration source) information is often non-stationary.

The present invention has been made in view of the above circumstances, and has a sound deadening / vibrating effect (a sound deadening amount or a vibration attenuation amount, (multi)) by an electronic sound deadening system or an electronic vibration damping system.
(Physical quantity equivalent to coherence) can be measured in real time in a calculation amount that can be processed at the personal computer level, and the number of input sensors necessary for performing suitable silencing and the position of the signal source to determine the position etc. An object of the present invention is to provide a method of measuring a sound deadening / damping effect that can be searched, a measuring device, and a signal source search device.

[0015]

In order to achieve the above object, the present invention detects a signal of a signal source in a region in which sound waves and vibration waves can propagate in a three-dimensional direction, and outputs a first detection signal. One or a plurality of first detecting means for detecting the signal of the signal source at an evaluation point where noise is suppressed / vibrated by interference in a predetermined area within the propagation area, and a second detection signal is output. One or a plurality of second detection means, an adaptive digital filter that processes the first detection signal based on a predetermined filter coefficient, an output signal of the adaptive digital filter, and the second digital filter. An error signal is obtained from the difference from the detection signal, a filter coefficient for minimizing the error signal is calculated based on the error signal according to a predetermined algorithm, and the filter coefficient of the adaptive digital filter is calculated using the filter coefficient. And a control means for updating the number,
The first sound detecting means is arranged at an appropriate position for detecting the signal of the signal source, and the second detecting means is arranged at the evaluation point so that the filter coefficient of the adaptive digital filter is sufficiently converged. After that, the second detection signal,
Sampling the error signal for a predetermined period of time, calculating the auto power spectra of the sampled second detection signal and error signal, and calculating the amount of silence or vibration attenuation based on the ratio of each auto power spectrum. Is characterized by.

Further, instead of the auto power spectrum of the error signal, the auto power spectrum of the output signal of the adaptive digital filter is calculated, and the ratio of each auto power spectrum is calculated, which corresponds to (multi) coherence. It is characterized by finding adaptive (multi) coherence. Further, calculation means for calculating a physical quantity (silence volume (or vibration attenuation) or adaptive (multi) coherence) showing a silencing / damping effect based on the ratio of the auto power spectrum, and after the physical quantity is calculated. It is characterized in that it is provided with means for changing the sensitivity direction of the first detection means by a predetermined angle and means for displaying at least a physical quantity of each of the first detection means in each sensitivity direction.

[0017]

According to the present invention, the first detecting means is arranged at an appropriate position for detecting the signal of the signal source (noise source or vibration source), and the second detecting means is arranged at the evaluation point. To do. The adaptive digital filter processes the first detection signal output from the first detecting means based on a filter coefficient given in advance. Here, the filter coefficient is an error signal obtained from the difference between the output signal of the adaptive digital filter and the second detection signal output from the second detection means, and the error signal is calculated according to a predetermined algorithm based on the error signal. It is calculated so that the error signal is minimized.

After the filter coefficients of the adaptive digital filter have sufficiently converged, the second detection signal and the error signal are sampled for a predetermined period, and the sampled second detection signal and error signal are sampled. Each auto power spectrum is calculated, and the amount of silence or vibration attenuation is calculated based on the ratio of each auto power spectrum.

Further, instead of the auto power spectrum of the error signal, the auto power spectrum of the output signal of the adaptive digital filter is calculated, and the ratio of each auto power spectrum is calculated, which corresponds to (multi) coherence. I try to find adaptive (multi) coherence. If the system is close to steady and the filter coefficients are sufficiently converged, the adaptive (multi) coherence obtained as described above matches the (multi) coherence. Further, this adaptive (multi) coherence does not require signal acquisition for a relatively long time, unlike the conventional analysis / analysis in the frequency domain, and the dynamic characteristics of the system can be grasped in real time (instantaneous).

According to another aspect of the present invention, a physical quantity exhibiting a sound deadening / damping effect such as the volume of sound (or the amount of vibration attenuation) or adaptive (multi) coherence is calculated, and after calculating the physical quantity, the physical quantity is calculated. The sensitivity direction of the first detecting means is changed by a predetermined angle, and by repeating this, a physical quantity exhibiting a sound deadening / damping effect is obtained for each sensitivity direction. This makes it possible to know in which direction and position the first detection means should be installed in order to improve the noise reduction / vibration suppression effect.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A preferred embodiment of a method for measuring a sound deadening / damping effect according to the present invention, a measuring device and a signal source searching device will be described in detail below with reference to the accompanying drawings. First, the definition of normal coherence and multicoherence will be described. FIG. 8 shows a model with one input and one output. In the figure, G _XX (z): Input auto power spectrum G _YY (z): Output auto power spectrum G _ZZ (z): System H output auto power spectrum G _nn (z): No correlation with input Noise auto power spectrum G _XY (z): Input / output cross power spectrum.

The coherence is given by Is defined by

Further, the coherence gamma ² represents the percentage that contains G in _{G YY (z) ZZ (z} ) (G YY (z) values excluding the G _nn (z) from). That is, Can also be expressed as

Next, the formula (8) = the formula (9) is proved. G _ZZ (z) = | H | ² · G _XX (z)… (10) G _XZ (z) = H · G _XX (z)… (11) Substituting equation (12) into equation (10), Here, since there is no correlation between the input and noise, G _XZ (z) = G _XY (z) (14) Then, substituting Eqs. (13) and (14) into Eq. (9), 8) Matches the formula.

In the equation (9), if there is no noise, that is, G _nn (z) = 0, then γ ² = 1. This is a case of an ideal constant coefficient linear system having one input and one output. Also, if system H does not exist, that is, if the input and output counterparts are completely uncorrelated, the coherence is zero. Next, multicoherence will be described.

As shown in FIG. 9, consider a model of a multi-input / single-output system in which inputs are stationary irregular processes and inputs are correlated. Multicoherence in this model
The signal source can be obtained by the ratio of the G _VV (z) minus the output G _YY (z) uncorrelated noise G _nn (z), and the output G _YY (z). Here, G _VV (z) is obtained by summing the product of the transfer function H _i (z) in a certain transfer path and the cross spectrum G _iV of its input and output for all inputs.

Now, consider the case where the output of the primary sound field (vibration field) as shown in FIG. 1 is sensed. In the figure, for simplicity, the plant noise (observation system noise) uncorrelated to the noise source (vibration source) is ignored. The detection sensor 10 is a sensor arranged at a position estimated to be a source of noise and vibration.
The signal x (n) detected by 0 is applied to the adaptive digital filter 12. Adaptive digital filter 1
2 is the input signal x (n) and the given filter coefficient W
The signal y (n) is obtained by the convolution operation with (z), and this signal y (n) is output to the addition point 16.

On the other hand, the error sensor 16 is a sensor arranged at an evaluation point where noise reduction or vibration suppression is performed, and the signal d (n) detected by the error sensor 16 is added to the addition point 16
Added to other inputs of. Adaptive digital filter 1
For 2, the filter coefficient is updated according to the LMS algorithm. That is, the error signal e obtained at the addition point 14
The filter coefficient is optimized so that the root mean square value E [e (n) ² ] of (n) is minimized.

FIG. 2 shows the model of FIG. 1 in comparison with the electronic silencing system or electronic damping system shown in FIG. 10. In this model, the transmission characteristic C from the speaker or the damping actuator to the error sensor is shown. Since (z) is an ideal system that does not exist, d '(n) = y (n), r (n) = x (n)
Is. Therefore, the control unit 15 updates the filter coefficient W of the adaptive digital filter 12 according to the following equation: W _{n + 1} = W _n -2 μe (n) x (n) (16) to optimize the filter coefficient W. _{Set to the} value W _opt (z).

Optimal solution of filter coefficient after adaptation W _opt (z)
(Winner filter solution) is derived as follows from the relationship between the power spectral density function and the transfer function of each signal. However, S _xd (z): Cross power vector between signal x (n) and signal d (n) S _xx (z): Auto power spectrum of signal x (n) Therefore, the adaptive digital filter is It can be seen that the inverse modeling of the transfer function T from the sensor to the sensor and the forward modeling of the transfer function G from the signal source to the evaluation point are performed at the same time. Also, the auto power spectrum S _yy (z) of the output of the optimum filter W _opt (z)
Is S _yy (z) = | W _opt (z) | ² S _xx (z) = | G (z) | ² S _nn (z) = S _dd (z)… (18) where S _nn (z ): Auto power spectrum of signal source n (n) S _dd (z): Auto power spectrum of signal d (n) In power spectra S _yy (z) and S _dd (z) that can be obtained as described above Strictly speaking, although it is different from the (multi) coherence based on the Fourier reference in the normal frequency domain, the (multi) coherence γ is similar to the equation (9).
² can be defined by the following equation: γ ² = S _yy (z) / S _dd (z) (19) Hereinafter, the (multi) coherence thus defined is referred to as adaptive (multi) coherence.

Since the auto power spectrum S _ee (z) of the error signal e (n) is S _ee (z) = S _dd (z) -S _yy (z) (20), S on both sides _When normalized by _dd (z), it becomes S _ee (z) / S _dd (z) = 1-γ ² (21). From the definition on the left side, the reciprocal of this equation is an ideal value using an adaptive digital filter. It can be considered as a sound deadening amount or vibration attenuation amount achieved by realization of a simple electronic silencing system or electronic vibration damping system. That is, consumption volume or vibration attenuation Att, the table by the following equation, Att = 10log (1-γ 2) -1 = 10log (S dd (z) / S ee (z)) (dB) ... (22) To be done.

FIG. 3 is a block diagram showing an embodiment of a sound deadening / vibration damping effect measuring apparatus according to the present invention. In the figure, the part surrounded by the alternate long and short dash line has the same configuration as that of FIG. 1, and therefore detailed description thereof is omitted here. The sampling unit 20 samples the error signal e (n) and the output signal d (n) of the error sensor 16 at, for example, a sampling frequency of 1 KHz after the filter coefficient W of the adaptive digital filter 12 has sufficiently converged, and the sampling is performed. The signal e (n) and the signal d (n) are converted into a fast Fourier transform unit (FFT) 2
Output to 2.

The FFT 22 inputs the input signal e (n) and signal d
(n) is divided by a time window of a predetermined period, and the fast Fourier transform is performed on the divided signals e (n) and d (n) to obtain signals e (n) and d (n). Auto power spectra S _ee (z) and S _dd (z) are obtained, and these auto power spectra S _ee (z) and S _dd (z) are output to the sound volume or vibration attenuation amount calculation unit 24.

The calculation unit 24 calculates the above-mentioned expression (22) based on the input auto power spectra S _ee (z) and S _dd (z), and obtains the sound deadening amount or the vibration attenuation amount Att (dB). . Volume reduction or vibration attenuation Att calculated in this way
With (dB), the silencing / damping effect (upper limit of performance) when an electronic silencing system or an electronic control system is implemented in this system can be estimated on the spot in advance.

In the above embodiment, the attenuation amount Att (dB) of the sound volume or vibration is obtained, but the adaptive (multi) coherence defined by the above-mentioned equation (19) may be obtained. In this case, the output signal y (n) of the adaptive digital filter 12 is sampled instead of the signal e (n), the auto power spectrum S _yy (z) of the signal y (n) is obtained, and the auto power spectrum S _dd ( The adaptive (multi) coherence γ ² is obtained from the ratio of z) and S _yy (z).

Further, since the technique of adaptive signal processing is used, even if the system forming the field is time-varying linear, measurement can be performed by following the behavior. That is, the signal e (n), the signal d (n) or the signal y (n) and the signal d (n), which are subjected to the fast Fourier transform processing, are delimited by shifting the time window little by little and the sequential auto power spectrum S _ee (z ) And S
_{By obtaining dd} (z) or the signals S _yy (z) and S _dd (z), the dynamic characteristics of the system can be grasped in real time (instantaneous).

Next, consider a model in which the number of signal sources and the number of adaptive digital filters (the number of detection sensors) are increased as shown in FIG. At this time, if homogeneity of those numbers is guaranteed, the optimum solutions W _1opt and W _2opt can be independently obtained, and can be written as follows using a matrix expression. [W _OPT (z)] = [S _XX (z)] ^-1 [S _XD (z)] = [[T (z)] ^H [S _NN (z)] [T (z)]] ^-1 × [T (z)] ^H [S _NN (z)] [G (z)] = [T (z)] ^-1 [G (z)] (23) where the subscript H represents Hermitian transformation, The meanings of the symbols in the formula are as follows. [W _OPT (z)] = [W _1opt (z), W _2opt (z)] ^T [S _XX (z)] = [[s _x1 (z)] ^T , [s _x2 (z)] ^T ] ^T ([S _x1 (z)] ^T = [S _x1x1 (z), S _x1x2 (z)], [s _x2 (z)] ^T = [S _x2x1 (z), S _x2x2 (z)]) [S _XD (z)] = [S _x1d (z), S _x2d (z)] ^T [T (z)] = [[t ₁ (z)] ^T , [t ₂ (z)] ^T ] ^T ([t ₁ (z)] ^T = [T ₁₁ (z), T ₁₂ (z)], [t ₂ (z)] ^T = [T ₂₁ (z), T ₂₂ (z)]) [S _NN (z)] = _Diag [S _n1n1 (z), S _n2n2 (z)] [G (z)] = [G ₁ (z), G ₂ (z)] ^T Here, the subscript T represents transposition transformation. From this equation, the adaptive digital filter group inputs the detection sensor group,
It can be seen that a virtual transmission system (multi-input 1-output system) can be expressed when the evaluation point (error sensor) is used as the output. That is, if an adaptive digital filter using, for example, the LMS algorithm is adopted in the electronic silencing system or the electronic vibration control system, the optimum approximation of the virtual transmission system is required in terms of the root mean square error.
Also, the auto power spectrum of the output of the optimum filter group S _yy (z) = [S _XD (z)] ^H [S _XX (z)] ^-1 [S _XD (z)]
The multicoherence η ² is Is defined as Furthermore, the error evaluation function (auto power spectrum of the error signal) is _Jopt (z) = _Sdd (z)-[ _SXD (z)] ^H [ _SXX (z)] ^-1 [ _SXD (z). ] = S _ee (z) (25) If both sides of this equation are normalized by J _d (z) = S _dd (z), Therefore, from the definition on the left side, the reciprocal of this equation can be considered to be the volume of silence (or the amount of vibration attenuation) achieved by realizing an ideal electronic silencing system (or electronic damping system) using an FIR adaptive filter. it can. That is, J _opt (z) /
By evaluating J _d (z), the effect (upper limit of performance) when the electronic silencing system (or electronic vibration damping system) is introduced can be estimated in advance by the multi-input system.

Now, by estimating the muffling volume in the electronic muffling system which controls in the actual three-dimensional acoustic space,
We tried to evaluate the performance of the system. Here, an experiment was carried out in the case where there were two uncorrelated signal sources (noise sources), two detection sensors, and one error sensor. At this time, the volume of the noise source was changed with time in order to observe the temporal fluctuations of the silence and adaptive (multi) coherence. This was done for the purpose of improving the relative S / N ratio to noises other than the noise source in the normal room, and observing how the volume reduction and adaptive (multi) coherence increase as a result. is there.

A round speaker driven by an M-series random signal was used as a noise source, and a microphone was used as a detection sensor and an error sensor. Further, the number of taps of the filter, the step size parameter, and the sampling frequency in the measurement system were set to: -The number of taps of the adaptive filter: 128 points-The step size parameter: 0.0003-The sampling frequency: 700 Hz. The measurement results are shown in FIGS.

The M-sequence noise generated from the noise source (loudspeaker) is band-limited to about 30 to 200 [Hz], and the noise reduction or adaptive (multi) herence is accompanied by the passage of time near that. It is observed that it is changing. Especially in this band, we can see how the silence level increases as the adaptive (multi) coherence approaches 1.0.

FIG. 7 shows an example of the arrangement positions of the detection sensor and the error sensor arranged in the automobile.
In the figure, S1: error sensor arranged in the passenger seat ear position S2: error sensor arranged in the rear seat ear position S3: detection sensor arranged in the driver's seat floor S4: arranged in the rear seat floor Installed detection sensor S5: Detection sensor mounted on engine mount bracket S6: Detection sensor mounted on center bearing bracket S7: Detection sensor mounted on propeller shaft S8: Rear final drive It is a detection sensor provided.

Then, the measuring device according to the present invention makes it possible to selectively selectively operate the error sensor and the detection sensor to estimate the muting volume and the like, thereby detecting each detection signal of the detection sensor and the detection sensor signal. You can see the correlation of. The detection sensor that outputs a detection signal having a large correlation contributes to muffling. That is, according to the present invention, since the degree of contribution of the muffling by each detection sensor can be known, at which position the detection sensor should be arranged, the number of detection sensors capable of effectively muffling or damping, etc. I also understand.

Further, as a detection sensor, a sensor array or an intensity sensor capable of changing the direction of noise detection sensitivity is used, and after the muting or adaptive multi-coherence is calculated, the sensitivity direction of the sensor is set to a predetermined value. By changing the angle and repeating this, it is possible to obtain the muffling volume or the adaptive multicoherence exhibiting the muffling effect for each sensitivity direction.

In other words, it is possible to provide a signal source search device that can know in which direction or position the detection sensor should be installed in order to improve the sound deadening effect. In this embodiment, the case where there is one error sensor at the evaluation point has been described. However, even when there is a plurality of error sensors, the filter coefficient of the adaptive digital filter can be converged to the optimum value, so that the error sensor is used. You may make it provide multiple pieces. According to this, it also becomes possible to measure the silencing effect by the electronic silencing system or the electronic damping system having a plurality of error sensors.

[0045]

As described above, the noise reduction according to the present invention
According to the measuring method and the measuring device of the vibration damping effect, it is possible to process the noise reduction / vibration effect (volume reduction or vibration attenuation, physical quantity equivalent to multi-coherence, etc.) by the electronic noise suppression system or electronic vibration damping system at the PC level. It is possible to perform measurement with a large amount of calculation and in real time, and it is possible to easily and accurately determine the number of input sensors and their arrangement positions, etc., which are necessary for performing suitable silencing and damping. Furthermore, according to the signal source search device of the present invention, when the signal sources are distributed,
It is possible to search for the signal source.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of essential parts of a sound deadening / vibration suppression effect measuring apparatus and a signal source exploration apparatus according to the present invention.

FIG. 2 is a diagram used for explaining an algorithm for updating the filter coefficient in FIG. 1 to an optimum value.

FIG. 3 is a block diagram showing an embodiment of a measurement / measurement apparatus for noise reduction / damping effect according to the present invention.

FIG. 4 is a diagram showing a model when the number of signal sources and the number of adaptive digital filters are increased.

FIG. 5 is a graph showing an example of the muting volume of the system in real time.

FIG. 6 is a graph showing an example of adaptive (multi) coherence of a system in real time.

FIG. 7 is a diagram showing an example of positions where a detection sensor and an error sensor are arranged in an automobile.

FIG. 8 is a diagram showing a one-input one-output model.

FIG. 9 is a diagram showing a model of a multi-input / single-output system.

FIG. 10 is a block diagram showing an example of a conventional electronic silencing system or electronic damping system.

[Explanation of symbols]

 10 ... Detection sensor 12 ... Adaptive digital filter 14 ... Addition point 15 ... Control part 16 ... Error sensor 20 ... Sampling part 22 ... Fast Fourier transform part 24 ... Arithmetic part

Claims (5)

[Claims] 1. One or a plurality of first detection means for detecting a signal of a signal source in a region where a sound wave / oscillation wave can propagate in a three-dimensional direction and outputting a first detection signal; One or a plurality of second detection means for detecting a signal of the signal source at an evaluation point where noise suppression / vibration is performed by interference in a predetermined area within the area, and outputting a second detection signal; An adaptive digital filter for processing the detection signal of No. 1 on the basis of a filter coefficient given in advance, and an error signal is obtained from the difference between the output signal of the adaptive digital filter and the second detection signal. A filter coefficient for minimizing the error signal according to a predetermined algorithm based on a predetermined algorithm and updating the filter coefficient of the adaptive digital filter with the filter coefficient. The second detecting means is arranged at an appropriate position for detecting the signal of the signal source, and the second detecting means is arranged at the evaluation point. After the filter coefficient of the adaptive digital filter is sufficiently converged, the second means is detected. Detection signal and the error signal are sampled for a predetermined period of time, the auto power spectra of the sampled second detection signal and error signal are respectively calculated, and the sound volume or vibration is attenuated based on the ratio of each auto power spectrum. A method for measuring the sound deadening / damping effect, which is characterized by calculating the amount. 2. One or a plurality of first detection means for detecting a signal of a signal source in a region where a sound wave / oscillation wave can propagate in a three-dimensional direction and outputting a first detection signal; One or a plurality of second detection means for detecting a signal of the signal source at an evaluation point where noise suppression / vibration is performed by interference in a predetermined area within the area, and outputting a second detection signal; An adaptive digital filter for processing the detection signal of No. 1 on the basis of a filter coefficient given in advance, and an error signal is obtained from the difference between the output signal of the adaptive digital filter and the second detection signal. A filter coefficient for minimizing the error signal according to a predetermined algorithm based on a predetermined algorithm and updating the filter coefficient of the adaptive digital filter with the filter coefficient. Means is provided at an appropriate position for detecting the signal of the signal source, and the second detecting means is provided at the evaluation point, and after the filter coefficient of the adaptive digital filter has sufficiently converged, the adaptive type An output signal of the digital filter and the second detection signal are sampled for a predetermined period, auto power spectra of the sampled output signal and the second detection signal are calculated, and a ratio of each auto power spectrum is calculated. A method for measuring the sound deadening / damping effect, which is characterized by obtaining the adaptive (multi) coherence corresponding to the (multi) coherence. 3. One or a plurality of first detection means for detecting a signal of a signal source in a region where a sound wave / oscillation wave can propagate in a three-dimensional direction, and outputting a first detection signal; One or a plurality of second detection means for detecting a signal of the signal source at an evaluation point where noise suppression / vibration is performed by interference in a predetermined area within the area, and outputting a second detection signal; An adaptive digital filter for processing the detection signal of No. 1 on the basis of a filter coefficient given in advance, and an error signal is obtained from the difference between the output signal of the adaptive digital filter and the second detection signal. Based on a predetermined algorithm based on a predetermined algorithm, a filter coefficient for minimizing the error signal is calculated, and a control means for updating the filter coefficient of the adaptive digital filter with the filter coefficient; and the adaptive digital filter. The second detection signal and the after filter coefficients of the filter are sufficiently converged, the predetermined period of time sampling the error signal, first and said sampling 2
Of the detection signal and the error signal, and calculating means for calculating the volume of mute / damping based on the ratio of each auto power spectrum, and measuring device. 4. One or a plurality of first detection means for detecting a signal of a signal source in a region where a sound wave / oscillation wave can propagate in a three-dimensional direction and outputting a first detection signal; One or a plurality of second detection means for detecting a signal of the signal source at an evaluation point where noise suppression / vibration is performed by interference in a predetermined area within the area, and outputting a second detection signal; An adaptive digital filter for processing the detection signal of No. 1 on the basis of a filter coefficient given in advance, and an error signal is obtained from the difference between the output signal of the adaptive digital filter and the second detection signal. Based on a predetermined algorithm based on a predetermined algorithm, a filter coefficient for minimizing the error signal is calculated, and a control means for updating the filter coefficient of the adaptive digital filter with the filter coefficient; and the adaptive digital filter. Filter factor of the filter and an output signal of said adaptive digital filter, and said second detection signal by a predetermined period sampled After sufficiently converged,
Calculating means for calculating an adaptive power (multi) coherence corresponding to the (multi) coherence by calculating the respective autopower spectra of the sampled output signal and the second detection signal, and calculating the ratio of the respective autopower spectra; A device for measuring the sound deadening effect by an electronic sound deadening system, which is characterized by being equipped. 5. One or a plurality of first detecting means for detecting a signal of a signal source in a region in which a sound wave / vibration wave can propagate in a three-dimensional direction with a sensitivity in a predetermined direction and outputting a first detection signal. Detection means, and one or a plurality of second ones that detect a signal of the signal source at an evaluation point where noise suppression / vibration is performed by interference in a predetermined area within the propagation area and output a second detection signal. Detecting means; an adaptive digital filter for processing the first detection signal based on a predetermined filter coefficient; and an error signal from the difference between the output signal of the adaptive digital filter and the second detection signal. Control means for obtaining and calculating a filter coefficient for minimizing the error signal according to a predetermined algorithm based on the error signal and updating the filter coefficient of the adaptive digital filter with the filter coefficient. And after the filter coefficient of the adaptive digital filter has sufficiently converged, the error signal and the second detection signal or the output signal of the adaptive digital filter are sampled for a predetermined period, and the sampled error signal ,
Calculating means for respectively calculating an auto power spectrum of the second detection signal or the output signal, and calculating a physical quantity exhibiting a silencing / vibrating effect based on the ratio of each auto power spectrum; and after calculating the physical quantity, the calculating means 1. A signal source exploration device comprising: a means for changing the sensitivity direction of the first detecting means by a predetermined angle; and a means for displaying at least a physical quantity of the first detecting means in each of the sensitivity directions. .