Patent Translate Powered by EPO and Google Notice This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate, complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or financial decisions, should not be based on machine-translation output. DESCRIPTION JP2002055684 [0001] BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention generates a secondary sound having the same amplitude and opposite phase as noise, and causes this to interfere with the noise to cancel it, and feeds back the signal after the interference to generate a secondary sound. The present invention relates to a feedback-type active noise control device used to generate [0002] 2. Description of the Related Art As a technique for reducing noise, a technique of generating a secondary sound having the same amplitude and opposite phase as noise and making it interfere with noise, thereby canceling noise with secondary sound and reducing noise Is being studied. As an application example of such a technique, for example, a case applied to an ear protector shown in JP-A-63-503186, a case applied to an audio headphone shown in JP-A-6-343195, etc. There is. [0003] FIG. 9 is a schematic view showing an example of a headphone using a conventional feedback type active noise control device. In the figure, 11 is a headphone, 12 is a diaphragm, and 13 is a microphone. In the example shown in FIG. 9, a local sealed space is formed between the 10-04-2019 1 diaphragm 12 of the headphone 11 (or ear protector) and the ear canal, and the small microphone 13 is disposed in the closed space in proximity to the diaphragm 12. There is. An electrical processing C is added to the signal of the microphone 13 to form a so-called feedback circuit that combines with the desired electrical signal to be reproduced, thereby obtaining an output acoustic signal with significantly reduced external noise and distortion. be able to. [0004] In such a configuration, a disturbance input such as noise is collected by the microphone 13, and a signal having the same amplitude and the opposite phase as the disturbance input is superimposed on the input signal and emitted from the diaphragm 12 as a secondary sound. As a result, the disturbance input and the sound emitted from the diaphragm 12 interfere with each other to cancel the disturbance input. Therefore, the listener can hear only the sound of the input signal. [0005] FIG. 10 is a block diagram showing an example of a general feedback system. In general, the feedback system is known to have an effect of suppressing the influence of the disturbance input d on the output y and an effect of suppressing the influence of the characteristic variation of the system on the output y. These effects are obtained by multiplying P (s), C (s) and H (s) in FIG. 10, so-called cycle transfer characteristic G0 (s) = P (s) × C (s) in this feedback system. It is known that the higher the gain characteristic of H (s), the higher it becomes. However, if the gain characteristic of the open loop transfer characteristic G0 (s) is increased indiscriminately, the stability of the entire system is impaired and the risk of oscillation increases, so the disturbance suppression effect, parameter variation suppression effect, and stability It is necessary to determine the characteristics of the compensation elements C (s) and H (s) while taking tradeoffs. [0006] In the case of forming a feedback circuit as an active noise control device, the time delay required for the sound wave to reach the error detection means such as the microphone from the speaker emitting the secondary sound becomes a major issue. The time delay element is generally represented by the time delay element = Exp (-jτω) (1) in the frequency domain. Here, 10-04-2019 2 τ represents a delay time, and ω represents an angular frequency. The equation (1) represents a characteristic of rotating on a circle having a size of a radius of 1 on a real-imaginary plane. This is considered from the Nyquist stability discriminant which specifies the stability of the feedback system, and at a frequency at which the time delay element passes (−1, 0) on the realimaging plane, P (s) × C ( s) It means that the gain of × H (s) can not be made 1.0 or more. That is, the larger the time delay included in the system, the more the loop transfer characteristic passes (-1, 0) on the real-imaging plane within a certain frequency range, and the design of the compensator becomes more difficult. Become. This is the reason why the feedback type active noise control device is limited to local noise control, that is, noise control in a space where the time delay is very short. [0007] In a very narrow enclosed space such as the headphone shown in FIG. 9 described above, the distance from the diaphragm 12 to the microphone 13 can be made extremely short, and the time delay can be made very short. it can. However, application to applications other than such applications has been difficult. [0008] In recent years, attempts have been made to achieve both stability against time delay and control performance. For example, Journal of the Japan Society of Mechanical Engineers (C edition), Vol. 62, no. 597, Nomin et al., "Active Noise Control of One-Dimensional Exhaust Duct System Applying H.INF. As shown in 157-162, there is a design case to which H 制 御 control theory is applied. Also, Proceedings of the Japan Society of Mechanical Engineers (C edition), Vol. 65, no. 637, Sano et al., "Study on modeling of acoustic transfer system for active noise control and design method of PID feedback control system," pp. As shown in 121-127, a case where the Smith method effective for a control object having a large time delay such as a chemical plant is applied to the design of a feedback compensation circuit is also reported. [0009] However, in any case, the effect is limited to the resonance mode inside the narrow-band duct, and there is a portion where the sound is increased in a part of the frequency range. That is, it is necessary to trade off the gain in order to maintain the stability of the feedback system by the 10-04-2019 3 delay element as described above. If the gain is uniformly lowered in a wide frequency range, the target gain can not be obtained at the target frequency, and a secondary sound that cancels out the noise can not be generated. In addition, when the gain is adjusted for each frequency domain, a phase difference occurs. Therefore, in order to obtain the noise reduction effect as the entire noise frequency range, it is necessary to experimentally obtain the adjustment of the gain and the phase difference for each frequency domain. Therefore, it has been very difficult to obtain uniform noise reduction effects over a wide range of continuous frequencies. [0010] Also, the Journal of the Japan Society of Mechanical Engineers (C edition), Vol. 61, no. No. 588, et al., "Active noise control of one-dimensional exhaust duct system by feedback control", pp. 588. It is also pointed out at 118-124 that the noise reduction effect of the feedback type active noise control device is mainly governed by the dynamic characteristics of the secondary sound emitting speaker that emits the secondary sound. [0011] SUMMARY OF THE INVENTION The present invention has been made in view of the abovedescribed circumstances, and achieves both stability of the system and uniform noise reduction in a wide range of continuous frequencies, and a system It is an object of the present invention to provide a feedback type active noise control device capable of designing a typical controller. [0012] SUMMARY OF THE INVENTION In the feedback type active noise control system of the present invention, the noise source is surrounded by a one-dimensional duct whose cross-sectional dimension is sufficiently smaller than the wavelength of the noise to be silenced. In such ducts, the sound waves of noise can be treated as plane waves. And while providing the secondary sound radiation speaker which radiates | emits a secondary sound in the duct, the secondary sound and noise radiated from a secondary sound radiation speaker are more distant from the said noise source than a secondary sound radiation speaker An error detection means is provided for detecting the sound pressure of the synthetic wave generated by interference. Furthermore, the secondary sound generation filter performs electrical signal processing on the sound pressure detected by the error detection means to 10-04-2019 4 generate a signal for secondary sound and outputs the signal to the secondary sound radiation speaker. Thus, a feedback circuit is configured. [0013] In such a feedback circuit, in order to obtain a uniform noise reduction effect in a predetermined frequency range, if the amplitude characteristic of the loop transfer characteristic G0 (s) of the feedback circuit is constant in the predetermined frequency range Good. However, since the frequency response characteristic P (s) from the input signal of the secondary sound radiation speaker to the error detection means is included in G0 (s), its amplitude characteristic is not constant. In general, the amplitude characteristic of G0 (s) can be made constant by multiplying the inverse characteristic of P (s) (the inverse of the amplitude characteristic and the sign conversion of the phase characteristic). However, P (s) includes a time delay until the sound wave emitted from the speaker reaches the error detection means. That is, in order to obtain P (s), when using the signal obtained by the error detection means, the sound signal including noise on the cone surface of the secondary sound radiation speaker at the time when the sound reaches the error detection means It will be necessary. The noise signal on the surface of the cone of this secondary sound radiation speaker is a signal of the sound to be detected in the future by the error detection means, so detection by the error detection means to obtain the inverse characteristic of P (s) It will require the future value of the signal and can not be realized. [0014] Therefore, first, the dynamic characteristic of P (s) is separated into a minimum phase portion Pmin (s) containing no delay component in a predetermined frequency range and the other portion, and the inverse characteristic of this Pmin (s) is obtained. Design the secondary sound generation filter to have. As a result, since the loop transfer characteristic G0 (s) of the feedback circuit has a constant gain in a predetermined frequency range, a uniform noise reduction effect can be obtained. Moreover, such a filter can be easily designed from the characteristics of the secondary sound radiation speaker and the like, and can be realized without repeating experimental trial and error. [0015] However, with this as it is, in the high frequency component, the phase difference becomes large 10-04-2019 5 due to the influence of the delay, and not only the noise can not be eliminated but also the feedback system may become unstable. Therefore, the gain of the high frequency component is reduced, and the influence of the high frequency component having a large phase difference is suppressed. As a result, it is possible to avoid the risk of oscillation of the feedback system and to perform stable noise control. The high frequency component of noise may be separately dealt with by a sound absorbing material or the like. [0016] The secondary sound generation filter as described above includes, for example, constant multiplication signal processing means for multiplying the electric signal of the sound pressure detected by the error detection means by a constant, and inverting the output signal from the constant multiplication signal processing means Inverting means, inverse characteristic signal processing means for performing signal processing on the output signal from the inverting means, low frequency pass signal processing means for reducing high frequency components of the output signal of the inverse characteristic signal processing means, and It can consist of The inverse characteristic signal processing means is, as described above, an amplitude characteristic that has an inverse relationship with the amplitude characteristic of the electro-acoustic frequency response from the input signal of the secondary sound radiation speaker to the output signal of the error detection means within a predetermined frequency range. Can be configured with an IIR digital filter. The numerator polynomial of the IIR digital filter used at this time matches the denominator polynomial in the case where the electro-acoustic frequency response characteristic is modeled as the IIR digital filter. Also, among the characteristic roots of the numerator polynomial in the case where the electro-acoustic frequency response characteristic is modeled as an IIR digital filter, the denominator polynomial is a characteristic root that is the inverse of the characteristic root whose size exceeds 1.0. And match a polynomial that has the same thing for characteristic roots whose size does not exceed 1.0. [0017] In the IIR digital filter of such a characteristic, since all characteristic roots in the denominator polynomial do not exceed 1.0, it is stable if there is no delay component. Also, in this IIR digital filter, among the characteristic roots of the molecular polynomial when the electro-acoustic frequency response characteristic is modeled as an IIR digital filter, the inverse of the characteristic roots with a magnitude of the characteristic root of more than 1.0 Give a characteristic root. The gain is compensated when the inverse characteristic root is given in this way. Further, the phase has a minimum phase characteristic excluding the delay component. 10-04-2019 6 [0018] Such an IIR digital filter has an inverse characteristic of the minimum phase characteristic in a predetermined frequency range of the electro-acoustic frequency response from the input signal of the secondary sound radiation speaker to the output signal of the error detection means. , Second phase signal with minimum phase characteristics can be generated. Under this condition, the phase difference increases as the frequency increases due to the delay component. Therefore, the high frequency component of the output signal of the inverse characteristic signal processing unit is reduced by the low frequency pass signal processing unit, and feedback is performed by the high frequency component. It prevents the circuit from becoming unstable. [0019] DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram showing an embodiment of a feedback type active noise control system according to the present invention. In the figure, 1 is a noise source, 2 is a duct, 3 is a secondary sound emission speaker, 4 is an error detection unit, 5 is a secondary sound generation filter, 6 is a fixed signal processing unit, 7 is a reverse unit, 8 is a reverse A characteristic signal processing unit 9 is a low frequency pass signal processing unit. [0020] The duct 2 is formed so as to surround the noise source 1, and the longest length of its cross section is made sufficiently shorter than the wavelength of the generated noise. In the duct of such a configuration, the noise emitted from the noise source 1 can be approximated as a plane wave. [0021] A secondary sound radiation speaker 3 for emitting secondary sound is disposed at a position on the outlet side of the duct 2 as viewed from the noise source 1. The distance from the secondary sound radiation speaker 3 to the outlet of the duct 2 may have a length (about several 10-04-2019 7 centimeters) such that a sound wave generated from the secondary sound radiation speaker 3 can be approximated as a plane wave. As described above, by making the longest length of the cross section of the duct 2 sufficiently smaller than the noise wavelength, it is possible to treat all the sound waves inside the duct 2 as plane waves, so the secondary sound radiation speaker 3 The silencing effect realized in the vicinity can be obtained uniformly in the cross section of the duct 2. [0022] An error detection unit 4 is provided in the vicinity of the secondary sound radiation speaker 3. The error detection unit 4 detects the sound pressure of the synthetic sound in which the noise from the noise source 1 and the secondary sound from the secondary sound radiation speaker 3 interfere with each other. For example, it is comprised including a microphone etc. [0023] A secondary sound generation filter 5 is disposed downstream of the error detection unit 4. The secondary sound generation filter 5 is composed of a constant-magnification signal processing unit 6, an inverting unit 7, an inverse characteristic signal processing unit 8, and a low frequency band pass signal processing unit 9. The fixed magnification signal processing unit 6 multiplies the signal of the error detection means by a constant. The inverting unit 7 inverts the output signal from the constant-magnification signal processing unit 6. The inverse characteristic signal processing unit 8 has a relationship between the amplitude characteristic of the electro-acoustic frequency response from the input signal of the secondary sound emission speaker 3 to the output signal of the error detection unit 4 and the inverse of the minimum phase characteristic within a predetermined frequency range. Signal processing is performed on the output signal inverted by the inverting unit 7. The inverse characteristic signal processing unit 8 can be configured by, for example, an IIR digital filter as described later. The low frequency range pass signal processing unit 9 reduces high frequency components of the output signal of the inverse characteristic signal processing unit 8. [0024] The signal detected by the error detection unit 4 is multiplied by a constant by the constant multiplication signal processing unit 6 by the secondary sound generation filter 5 having such a 10-04-2019 8 configuration, and is further multiplied by -1 by the inversion unit 7 to be inverted. And sent to the inverse characteristic signal processing unit 8. The inverse characteristic signal processing unit 8 performs the signal processing according to the above-mentioned characteristic, and the low frequency range pass signal processing unit 9 reduces high frequency components. Then, the secondary sound radiation speaker 3 is driven by the output signal of the low frequency range pass signal processing unit 9 to radiate the secondary sound into the duct 2. [0025] Here, a method of realizing the inverse characteristic signal processing unit 8 will be described in detail below. First, the electro-acoustic frequency response from the input signal of the secondary sound emission speaker 3 to the output signal of the error detection unit 4 is measured using an FFT analyzer or the like. FIG. 2 is a graph showing an example of the electro-acoustic frequency response from the input signal of the secondary sound radiation speaker 3 to the output signal of the error detection unit 4. In the frequency response characteristic shown in FIG. 2, for example, there is a peak of gain at a frequency of 100 tens of Hz, and the phase is inverted by 180 degrees. Such a large peak of gain often appears as a characteristic of the secondary sound emission speaker 3. If a gain characteristic as shown in FIG. 2A is obtained, basically a signal using this inverse characteristic is generated, and noise can be eliminated if interference occurs. However, as described above, in order to ensure the stability of the feedback system, the gain can not be 1 or more at a frequency at which the phase is reversed. That is, if the phase has a 180degree inverted characteristic, when the secondary sound of that frequency is emitted, it acts to intensify the noise, which may cause oscillation. However, if the peak of such a large gain is suppressed to 1 or less and the gain is uniformly reduced in all frequency ranges, there is a possibility that the desired gain can not be obtained even at the target noise frequency. [0026] In the electro-acoustic frequency response from the input signal of the secondary sound emission speaker 3 to the output signal of the error detection unit 4 as shown in FIG. 2, the sound radiated from the secondary sound emission speaker 3 is A delay component until reaching the error detection unit 4 is included. Therefore, a large change is particularly shown in the phase characteristics shown in FIG. In the present invention, the inverse characteristic signal processing unit 8 is configured to perform signal processing in the minimum phase characteristic except for the delay component which has a great influence on such phase characteristic. 10-04-2019 9 [0027] As shown in FIG. 2, when the electro-acoustic frequency response from the input signal of the secondary sound radiation speaker 3 to the output signal of the error detection unit 4 is obtained, the frequency response characteristic is numerically expressed in the form of an IIR digital filter Model. The frequency response characteristic shown in FIG. 2 can be expressed as a 12th-order model when expressed in an IIR digital model. [0028] In general, the IIR digital model is represented by the ratio of the numerator polynomial to the denominator polynomial (fractional equation) as follows. In this equation, the characteristic root of the numerator polynomial is called "zero point", and the characteristic root of the denominator polynomial is called "pole", and it is known as an important factor that determines the characteristic of the digital filter. In addition, when the size of these "zeros" and "poles" exceeds 1.0, they are referred to as "unstable zeros" and "unstable poles", respectively. In particular, it is known that the output of the digital filter diverges when there is even one "unstable pole". [0029] FIG. 3 is a distribution diagram of poles and zeros when the electro-acoustic frequency response characteristic from the input signal of the secondary sound radiation speaker 3 to the output signal of the error detection unit 4 is represented as an IIR digital model. The poles and zeros in the IIR digital model obtained from the electro-acoustic frequency response from the input signal of the secondary sound radiation speaker 3 shown in FIG. 2 to the output signal of the error detection unit 4 are determined and plotted on the real-imaging plane It becomes like FIG. In FIG. 3, points indicated by black rhombus indicate poles, and points indicated by black triangles indicate zeros. As can be seen with reference to FIG. 3, it can be seen that the poles are all within a unit circle centered at the origin of radius 1.0. However, it can be seen that unstable zeros exist at coordinates (1.01, 0) and coordinates (5.98, 0) in the case of zeros. [0030] In this IIR digital model, it is stable because there is no unstable pole. However, in order to 10-04-2019 10 generate a secondary sound that actually cancels out noise, an inverse characteristic of the electro-acoustic frequency response from the input signal of the secondary sound radiation speaker 3 to the output signal of the error detection unit 4 is required. That is, an inverse IIR digital model in which the denominator polynomial and the numerator polynomial in the abovedescribed IIR digital model are interchanged is required. At this time, in the inverse IIR digital model in which the denominator polynomial and the numerator polynomial are simply interchanged, the unstable zero point in the original IIR digital model becomes an unstable pole, which results in divergence. [0031] Therefore, in the IIR digital model described above, among the characteristic roots of the numerator polynomial, for those whose size exceeds 1.0, the characteristic roots to be the inverse number are given, and the size is 1.0 A new IIR digital model is designed to match a polynomial that has the same one for characteristic roots that do not exceed. That is, the numerator polynomial with the unstable zero point (1.01, 0) shown in FIG. 3 as (0.99, 0) and the unstable zero point (5. 98, 0) as (0.167, 0) is shown. Ask. FIG. 4 is a distribution diagram of poles and zeros in the new IIR digital model. The representation of the points in the figure is the same as in FIG. By giving a characteristic root which is the reciprocal of the unstable zero as described above, the zeros of the new IIR digital model are all within the unit circle as shown in FIG. [0032] As described above, the amplitude characteristic with respect to the frequency among the frequency response characteristics of the original IIR digital model is stored as it is in the new IIR digital model by the conversion operation giving the characteristic root which is the reciprocal of the unstable zero point. Ru. The phase characteristic is realized as a so-called "minimum phase characteristic" obtained by removing the time delay from the phase characteristic of the original IIR digital model. Thus, in the original IIR digital model obtained from the electro-acoustic frequency response from the input signal of the secondary sound emission speaker 3 to the output signal of the error detection unit 4, the unstable zero point of the molecular polynomial is its inverse The simple operation of giving characteristic roots gives the minimum phase characteristic excluding the time delay component. [0033] 10-04-2019 11 The inverse IIR digital model created by replacing the numerator polynomial and the denominator polynomial of the new IIR digital model obtained as described above has no unstable pole and is stable. Also, in this inverse IIR digital model, signals can be processed without the need for future values. The inverse characteristic signal processing unit 8 can be configured by an IIR digital filter that realizes such an inverse IIR digital model. [0034] As the frequency response of the portion combining the inverse characteristic signal processing unit 8 and the secondary sound radiation speaker 3 realized as described above, the amplitude characteristic is 1.0 in a predetermined frequency region and the phase becomes higher as the frequency becomes higher. There remains a time delay component in which the phase angle is delayed. This is because the inverse characteristic signal processing unit 8 separates the minimum phase characteristic and the time delay component and performs signal processing on the minimum phase characteristic, so the time delay component remains as it is. [0035] For example, if the distance from the secondary sound radiation speaker 3 to the error detection unit 4 is about 3 cm, the phase delay at 1 kHz can be suppressed to 36 degrees, but at 5000 Hz the phase is near 180 degrees. And become unstable as a system. However, in the high frequency region of 5000 Hz, a passive noise reduction method using a sound insulating material or a sound absorbing material is more effective than such active noise control. Therefore, in the present invention, a low frequency band pass signal processing unit 9 for suppressing the gain in the high frequency band is provided, and the noise is reduced by the passive noise reduction method without performing the active noise control in the high frequency band. There is. [0036] FIG. 5 is a graph showing an example of the frequency response characteristic of the low frequency pass signal processor. FIG. 5A shows gain characteristics, and FIG. 5B shows phase characteristics. As shown in FIG. 5A, the low frequency range pass signal processing unit 9 reduces the gain in the high frequency range. This prevents oscillation or the like due to the secondary sound having a large phase difference. In addition, since the phase adjustment is not 10-04-2019 12 performed in the low frequency range pass signal processing unit 9, as shown in FIG. 5B, the phase is gradually delayed in the high frequency range. Of course, processing may be performed to advance the phase as the frequency becomes higher, and it may be configured to correct the phase delay even by a small amount. [0037] FIG. 6 is a Nyquist diagram showing the loop transfer characteristic of the system depending on the presence or absence of the low frequency range signal processing unit. The example shown in FIG. 6 shows an example in which the constant-magnification signal processing unit 6 is given an amplitude of 1.2 times. In this case, when the low frequency pass signal processing unit 9 is not provided, the loop transfer characteristic passes through the left side of the point of (-1.0, 0) when passing the real axis. However, by providing the low frequency range pass signal processing unit 9, the gain in the high frequency range is suppressed, and as shown in FIG. It can be increased. This makes it possible to avoid the risk of oscillation. [0038] By configuring the secondary sound generation filter 5 as described above, stability of the system and noise reduction in a continuous frequency range can be achieved. Further, each of the parts constituting such a secondary sound generation filter 5 can be determined almost uniquely if the characteristics of the secondary sound radiation speaker 3 are determined, and it is It can be designed without relying on intuition or experience. In this example, the secondary sound generation filter 5 is shown by four components, but it is also possible to configure some of them as long as the above-mentioned function is achieved. [0039] FIG. 7 is a graph showing an example of the noise reduction effect by the conventional feedback system, and FIG. 8 is a graph showing an example of the noise reduction effect according to the present invention. In the figure, the thin solid line indicates the gain before silencing, and the thick solid line indicates the gain after silencing. Furthermore, the broken line indicates the feedback gain. In the conventional feedback system, as shown in FIG. 7, the feedback gain (broken line) provided varies with frequency due to the influence of the characteristics of the secondary sound radiation speaker, and the reduction effect is increased to a high frequency. 10-04-2019 13 Frequency is mixed. As a result, assuming that the predetermined frequency range is 1 kHz, for example, when viewed as the noise reduction amount of the entire 1 kHz, only a noise reduction effect of less than 2 dB can be obtained. [0040] On the other hand, in the noise reduction according to the present invention, as shown by a broken line in FIG. 8, uniform feedback gain can be added over all frequencies. By this, it was possible to realize noise reduction in a continuous frequency range. For example, when the predetermined frequency range is 1 kHz, a noise reduction effect of 6 dB can be obtained as the noise reduction amount of the entire 1 kHz. [0041] As is apparent from the above description, according to the feedback type active noise control device of the present invention, stable noise reduction in a continuous frequency range can be made possible. Furthermore, since there is no trial and error part in the controller design, it is possible to greatly reduce development costs and mass production costs. As a result, there is an effect that it is possible to provide a low-cost, very small active noise control device utilizing the advantage of the feedback type. [0042] Brief description of the drawings [0043] FIG. 1 is a block diagram showing an embodiment of a feedback type active noise control system according to the present invention. [0044] FIG. 2 is a graph showing an example of an electro-acoustic frequency response from an input signal of a secondary sound emission speaker to an output signal of an error detection unit. [0045] 10-04-2019 14 FIG. 3 is a distribution diagram of poles and zeros when an electro-acoustic frequency response characteristic from an input signal of the secondary sound emission speaker to an output signal of the error detection unit is represented as an IIR digital model. [0046] FIG. 4 is a distribution diagram of poles and zeros in the new IIR digital model. [0047] FIG. 5 is a graph showing an example of the frequency response characteristic of the low frequency pass signal processing unit. [0048] FIG. 6 is a Nyquist diagram showing the round trip transfer characteristic of the system depending on the presence or absence of the low frequency pass signal processing unit. [0049] FIG. 7 is a graph showing an example of the noise reduction effect by the conventional feedback system. [0050] FIG. 8 is a graph showing an example of the noise reduction effect according to the present invention. [0051] FIG. 9 is a schematic configuration view showing an example of a headphone using a conventional feedback type active noise control device. [0052] 10 is a block diagram showing an example of a general feedback system. [0053] Explanation of sign 10-04-2019 15 [0054] DESCRIPTION OF SYMBOLS 1 ... noise source, 2 ... duct, 3 ... secondary sound radiation speaker, 4 ... error detection part, 5 ... secondary sound generation filter, 6 ... fixed size signal processing part, 7 ... inversion part, 8 ... reverse characteristic signal processing 9, low frequency pass signal processing unit, 11 headphones, 12 diaphragms, 13 microphones. 10-04-2019 16

1/--страниц