JP5295086B2 - Active noise control device and active noise control method - Google Patents

Description

  The present invention relates to an active noise control device and an active noise control method.

  As a conventional noise control device and noise control method, a sound source that outputs control sound is prepared, and active noise control (ANC: Active Noise) that reduces the noise at the control point by causing the output control sound to interfere with the noise Controls and methods are known (see, for example, Patent Document 1). The apparatus described in Patent Document 1 includes a reference microphone that detects noise, a control speaker that outputs control sound, and a first error microphone that is disposed near the control speaker on the control point side and detects noise and control sound. The forward type active noise control is performed. Before the active noise control is performed, the second error microphone is arranged at a position farther than the first error microphone as viewed from the control speaker, and the transfer characteristics between the first error microphone and the second error microphone are acquired in advance. Then, the transfer characteristic for control is calculated using the acquired transfer characteristic between the first error microphone and the second error microphone. By using the transfer characteristic obtained by the calculation, active noise control is performed as if the second error microphone existed at the control point.

JP 2004-177419 A

  By the way, in the active noise control device, when the noise source, the control speaker, and the error microphones arranged at the control point are arranged in a straight line, the propagation directions of the noise and the control sound coincide. In this case, the area where the silencing effect is exerted is a wide area extending downstream of the control point. On the other hand, when the noise source moves to a position deviating from the straight line connecting the control speaker and the error microphone, the propagation directions of the noise and the control sound do not match and intersect. In this case, the region where the silencing effect is produced is limited to the vicinity of the control point, and the sound is increased in other regions. That is, in the conventional active noise control device, when the noise source moves, there is a possibility that the area where the silencing effect is exerted may be reduced.

  Furthermore, it is desired that the active noise control device be miniaturized while maintaining a certain silencing effect. However, if an error microphone is arranged in the vicinity of the control speaker as in the device described in Patent Document 1, there is a possibility that the area where the silencing effect is exerted may be reduced for reasons other than the above. In general active noise control, the ratio of the characteristic of the path from the noise source to the error microphone and the characteristic of the path from the control speaker to the error microphone is used by using an adaptive filter (Filtered-X LMS) or transfer function method. Control filter. Here, the characteristics of the path from the control speaker to the error microphone are strongly influenced by the structure of the sound pressure driving surface and the support structure of the speaker as the distance between the control speaker and the error microphone is shortened. For this reason, if an error microphone is placed in the vicinity of the control speaker, there is a difference between the characteristics of the path from the control speaker to the error microphone and the characteristics of the path from the control speaker to a point located downstream of the position of the error microphone. Occurs. Therefore, when the noise source, the control speaker, and the error microphone are arranged in a straight line, the error microphone is arranged in the vicinity of the control speaker, and active noise control is performed using the characteristics of the path from the control speaker to the error microphone, the error microphone Although the silencing effect can be expected only in the vicinity of the arrangement position of the error microphone, the silencing effect cannot be expected on the downstream side of the arrangement position of the error microphone.

  Accordingly, the present invention has been made to solve such a technical problem, and can maintain the area of the area that provides a silencing effect even when the noise source moves and can be downsized. It is an object to provide an active noise control device and an active noise control method.

  Here, as a result of intensive research, the present inventor has focused on the fact that the waveform becomes similar to the temporal and spatial progression of the sound wave at a position away from the control speaker. The structure of the sound pressure driving surface and the speaker support structure are derived by deriving the characteristics of the error path of the microphone placed near the control speaker using the measurement results obtained by placing the microphone at a position far from the control speaker. We found that a microphone can be placed near the control speaker without considering the effects of objects, and the closer the control speaker is to the microphone that detects the control sound and noise, the more effective it is even if the noise source moves. It was found that the area of the playing area can be maintained.

Therefore, an active noise control device according to the present invention is an active noise control device that outputs control sound to a control region and controls noise in the control region, and is disposed between a noise source and the control region. A control sound source that outputs a control sound to the control region side, a reference microphone that is disposed between the noise source and the control sound source and detects noise output from the noise source , the control sound source, and the control juxtaposed prior Symbol control sound source while being arranged between the areas, to identify a near characteristic detecting microphone for detecting noise and the control sound, the characteristics of the transmission path from the noise source to the vicinity characteristic detecting microphone identification Means, a transmission path characteristic from the noise source to the proximity characteristic detection microphone, a transmission path characteristic from the noise source identified in advance by measurement to a predetermined control point in the control region, and a measurement in advance. same An error path calculating means for calculating an error path characteristic from the control sound source to the proximity characteristic detecting microphone based on a characteristic of the error path from the control sound source to the control point, and a proximity characteristic from the noise source Based on the characteristics of the transmission path to the detection microphone and the characteristics of the error path from the control sound source to the proximity characteristic detection microphone, the control sound and noise are phase-interfered at the control point to reduce noise in the control area. Control filter constituting means constituting the control filter, and feedforward control means for feedforward controlling the control sound of the control sound source based on the detection result of the reference microphone and the control filter.

  In the active noise control device according to the present invention, the characteristics of the transmission path from the noise source to the proximity characteristic detection microphone, the characteristics of the transmission path from the noise source identified in advance by measurement to a predetermined control point in the control region, and Based on the characteristic of the error path from the control sound source to the control point identified in advance by measurement, the characteristic of the error path from the control sound source to the neighborhood characteristic detection microphone is calculated. In general, the characteristics of an acoustic propagation path can be expressed as a simple phase delay and an amplitude attenuation with distance. For this reason, for example, the characteristics of the transmission path from the noise source to the neighborhood characteristic detection microphone are compared with the characteristics of the transmission path from the noise source to the control point, and the error path from the control sound source to the control point is based on the result. The characteristics of the error path that are not affected by the structure of the sound pressure drive surface and the speaker support structure are calculated by linearly transforming the characteristics of the sound and determining the characteristics of the error path from the control sound source to the nearby characteristic detection microphone. be able to. This makes it possible to achieve a silencing effect also on the downstream side of the control point, and to arrange noise in the vicinity of the control sound source and to reduce the noise at the control point without arranging a microphone at the control point. Can be muted. By arranging the proximity characteristic detection microphone close to the control sound source, the distance from the noise source to the control sound source and the distance from the noise source to the proximity characteristic detection microphone can be handled as being substantially equal. That is, in discussing the positional relationship between the control sound source and the proximity characteristic detection microphone and the noise source, a straight line connecting the control sound source and the proximity characteristic detection microphone can be handled as a point. For this reason, even when the noise source moves, the noise source, the control sound source, and the proximity characteristic detection microphone can always be handled as being arranged in a straight line. Therefore, the region where the silencing effect is exerted is only the region where the control sound source and the proximity characteristic detection microphone are symmetrically moved in a direction that is point-symmetric with the position of the noise source as the noise source moves. It is possible to avoid the area from being reduced.

  Here, the error path calculation means calculates a phase and an amplitude related to a frequency band to be controlled based on a characteristic of a transmission path from the noise source to the proximity characteristic detection microphone, and is previously identified by measurement. Based on the characteristics of the transmission path from the noise source to the control point, the phase and amplitude related to the frequency band to be controlled are calculated, and the control sound source identified in advance by measurement based on both the calculated phase and amplitude It is preferable to calculate the characteristic of the error path from the control sound source to the neighboring characteristic detection microphone by correcting the phase and amplitude of the characteristic of the error path from the control point to the control point.

  With this configuration, the characteristics of the error path from the control sound source to the control point based on the characteristics of the transmission path from the noise source to the proximity characteristic detection microphone and the characteristics of the transmission path from the noise source to the control point. Can be linearly transformed. As described above, the characteristic of the error path from the control sound source to the neighborhood characteristic detecting microphone can be calculated from the characteristic not affected by the structure of the sound pressure driving surface and the speaker support structure.

  Further, the error path calculating means divides the characteristic of the transmission path from the noise source to the proximity characteristic detection microphone by the characteristic of the transmission path from the noise source identified in advance by measurement to the control point. The error path characteristic from the control sound source to the neighboring characteristic detection microphone is calculated by adding the obtained value to the error path characteristic from the control sound source identified in advance by measurement to the control point. Is preferred.

  With this configuration, the characteristics of the error path from the control sound source to the control point, the characteristics of the transmission path from the noise source to the proximity characteristic detection microphone, and the transmission from the noise source identified in advance by measurement to the control point The ratio with the characteristics of the route can be integrated. Therefore, while maintaining the sound waveform of the error path from the control sound source to the control point, the characteristics of the transmission path from the noise source to the nearby characteristic detection microphone and the control point from the noise source identified in advance by measurement It is possible to shift only the phase difference from the characteristics of the transmission path up to, that is, the position of the proximity characteristic detection microphone. Therefore, the characteristics of the error path that is not affected by the structure of the sound pressure driving surface and the speaker support structure can be calculated by a simple calculation.

The active noise control method according to the present invention is disposed between a noise source and a control region, and is disposed between a control sound source that outputs a control sound to the control region side, and the noise source and the control sound source. And a reference microphone that detects noise output from the noise source , and a proximity characteristic detection that is arranged between the control sound source and the control region and is arranged in parallel with the control sound source to detect noise and control sound. An active noise control method for an active noise control apparatus having a microphone, the identification step for identifying a characteristic of a transmission path from the noise source to the neighborhood characteristic detection microphone, and the noise source identified by the identification step Characteristics of the transmission path to the proximity characteristic detection microphone, characteristics of the transmission path from the noise source identified in advance by measurement to a predetermined control point in the control region, and identification in advance by measurement An error path calculating step of calculating an error path characteristic from the control sound source to the proximity characteristic detecting microphone based on a characteristic of an error path from the control sound source to the control point, and a proximity characteristic from the noise source Based on the characteristics of the transmission path to the detection microphone and the characteristics of the error path from the control sound source to the proximity characteristic detection microphone, the control sound and noise are phase-interfered at the control point to reduce noise in the control area. And a feedforward control step for feedforward controlling the control sound of the control sound source based on the detection result of the reference microphone and the control filter.

  According to the active noise control method of the present invention, the same effect as that of the above-described active noise control device can be obtained.

  According to the present invention, it is possible to maintain the area of the area that provides a silencing effect even when the noise source moves, and to reduce the size.

It is a schematic diagram of a silence system which has an active noise control device concerning an embodiment of the present invention. It is a block diagram which shows the control circuit of the 1st error path | route identification part with which the active noise control apparatus in FIG. 1 is provided. It is a block diagram which shows the control circuit of the 2nd error path | route identification part with which the active noise control apparatus in FIG. 1 is provided. It is a block diagram which shows the control circuit of the 1st transmission path identification part with which the active noise control apparatus in FIG. 1 is provided. It is a block diagram which shows the control circuit of the 2nd transmission path identification part with which the active noise control apparatus in FIG. 1 is provided. It is a block diagram which shows the control circuit of the noise control part with which the active noise control apparatus in FIG. 1 is provided. It is a schematic diagram for demonstrating the calculation which calculates | requires an error path | route. It is a schematic diagram for demonstrating the frequency dependence of a phase. It is a schematic diagram explaining the effect of the active noise control apparatus in FIG. It is a schematic diagram explaining the effect of the active noise control apparatus in FIG. It is a block diagram which shows the structure of the active noise control apparatus used for the Example. It is a block diagram which shows the structure of the active noise control apparatus used for the Example. It is an experimental result which shows the silencing effect in an Example visually. It is an experimental result which shows the silencing effect in an Example visually. It is an experimental result which shows visually the silencing effect in a comparative example. It is an experimental result which shows visually the silencing effect in a comparative example.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In addition, in each figure, the same code | symbol is attached | subjected to the same or an equivalent part, and the overlapping description is abbreviate | omitted.

  The active noise control apparatus according to the present embodiment is an active noise control apparatus that performs so-called active noise control in which a control sound is output with respect to noise and canceled by causing phase interference.

  First, the configuration of the active noise control device according to the present embodiment will be described. FIG. 1 is a schematic diagram showing a configuration of a silencing system having an active noise control device according to the present embodiment. As shown in FIG. 1, the silencing system 6 silences the noise output from the noise source S in the area A in the area B, for example, in an indoor office partitioned by the partition 21 into the areas A and B. It is preferably used for the above. Here, a point targeted by the active noise control device provided in the silencing system 6 for silencing is defined as a control point P. It is assumed that this control point P is set on the region B side. That is, the area B becomes a control area. In the following, it is assumed that the sound field in the indoor office is time-variant. For example, it is assumed that the noise source S moves in the office, and the sound speed changes in the office.

  The silencer system 6 is, for example, arranged in parallel in the vicinity of a reference microphone 11 that detects noise output from the noise source S, a control speaker (control sound source) 13 that outputs control sound, and the control speaker 13, and transmits control sound and noise. A proximity characteristic detecting microphone 12 for detecting, a DSP (Digital Signal Processor) device 14 for controlling the control speaker 13, and a personal computer (not shown) connected to the DSP device 14 are provided. Here, the DSP is, for example, an application processor that can perform digital signal processing at high speed, such as ROM (Read Only Memory), RAM (Random Access Memory), accelerator, A / D (Analog / Digital). It includes a conversion circuit, a D / A (Digital / Analog) conversion circuit, a signal amplification circuit, a howling suppression circuit, and the like. The reference microphone 11, the proximity characteristic detection microphone 12, the control speaker 13, and the DSP device 14 constitute an active noise control device.

  The reference microphone 11 is disposed, for example, in the vicinity of the control speaker 13 and has a function of outputting an analog signal corresponding to the noise output from the noise source S. The reference microphone 11 may be disposed in the vicinity of the noise source S. In addition, the proximity characteristic detection microphone 12 is provided, for example, in the vicinity of the control speaker 13 and has a function of outputting an analog signal corresponding to the control sound and noise. The reference microphone 11 and the proximity characteristic detection microphone 12 are configured to be able to output analog signals of detected sounds to the DSP device 14 respectively.

  The control speaker 13 is a sound source that outputs a control sound for canceling noise. The control speaker 13 is arranged above the partition 21 so as to output a control sound to the control point P side. The control speaker 13 is arranged so that the direct component of the output control sound has the same direction of arrival as the direct component of the noise coming from the noise source S at the control point P. That is, the control speaker 13 is arranged so that the direct sound of the noise and the direct sound of the control sound follow the same path, and the sound surface of the direct sound of the noise and the sound surface of the direct sound of the control sound overlap at the control point P. Is done. Here, the same direction of arrival means that it includes substantially the same direction of arrival, and includes a normally allowed installation error. The control speaker 13 may be configured to be drivable so as to automatically follow the direct sound of the noise. The control speaker 13 is connected to the DSP device 14 and has a function of outputting a control sound according to a control signal of the DSP device 14.

  The DSP device 14 is connected to the reference microphone 11, the proximity characteristic detection microphone 12, the control speaker 13, and the personal computer. And it has the function to predict the noise signal in the control point P. The DSP device 14 has a function of generating a signal having a phase opposite to that of the predicted noise signal and inputting the signal to the control speaker 13.

  In order to realize these functions, the DSP device 14 includes a first transmission path identification unit that identifies a sound field characteristic of the first transmission path, which is a path from the noise source S to the proximity characteristic detection microphone 12, and the noise source S. A second transmission path identification unit that identifies the sound field characteristic of the second transmission path that is the path to the control point P, and the sound field characteristic of the first error path that is the path from the control speaker 13 to the neighborhood characteristic detection microphone 12 are identified. A first error path identification unit that performs identification, a second error path identification unit that identifies a sound field characteristic of a second error path that is a path from the control speaker 13 to the control point P, and a noise control unit (identification that controls noise and control sound) Means, error path calculating means, control filter constituting means, feedforward control means).

First, details of the first error path identification unit that identifies the sound field characteristic C _n of the first error path from the control speaker 13 to the proximity characteristic detection microphone 12 will be described with reference to FIG. FIG. 2 is a block diagram illustrating a control circuit of the first error path identification unit included in the DSP device 14. As shown in FIG. 2, the DSP device 14 includes an error filter C _n for identifying the sound field characteristic C _n of the first error path. Here, the filter is a signal that performs signal processing on an input signal based on a set filter coefficient to generate an output signal. For example, a FIR (Finite Impulse Response) filter is used. The same applies to the filters described below. The DSP device 14 has a function of outputting a signal y (t) for outputting, for example, M-sequence noise from the control speaker 13 for identification. The same signal y (t) is input to the error filter C _n and the output signal c _n (t) is calculated. Then, the M-sequence picked up in the vicinity characteristic detection microphone 12 follows the first error path noise signal e _{n (t),} the calculated output signal c _{n (t)} electrically subtracted signal u _{1 (t} ). The error filter C _n has a function to minimize the value of the signal u ₁ (t). Specifically, the filter coefficient of the error filter C _n is determined so that the value of the signal u ₁ (t) is minimized, and the sound field characteristic C _n of the first error path is identified.

Next, details of the second error path identification unit for identifying the sound field characteristic C _f of the second error path from the control speaker 13 to the control point P will be described with reference to FIG. FIG. 3 is a block diagram illustrating a control circuit of the second error path identification unit included in the DSP device 14. As shown in FIG. 3, the DSP device 14 has an error filter C _f for identifying the sound field characteristic C _f of the second error path. At the control point P, a remote characteristic detection microphone 15 having the same characteristics as the proximity characteristic detection microphone 12 is disposed. The DSP device 14 has a function of outputting a signal y (t) for outputting, for example, M-sequence noise from the control speaker 13 for identification. The same signal y (t) is input to the error filter C _f and the output signal c _f (t) is calculated. Then, the signal u ₂ (t) obtained by electrically subtracting the calculated output signal c _f (t) from the M-sequence noise signal e _f (t) collected by the proximity characteristic detection microphone 12 along the second error path. ). The error filter C _f is configured to minimize the value of the signal u ₂ (t). Specifically, the filter coefficient of the error filter C _f is determined so that the value of the signal u ₂ (t) is minimized, and the sound field characteristic C _f of the second error path is identified.

Next, details of the first transmission path identification unit that identifies the sound field characteristic T _n of the first transmission path from the noise source to the proximity characteristic detection microphone 12 will be described with reference to FIG. 4. FIG. 4 is a block diagram illustrating a control circuit of the first transmission path identification unit included in the DSP device 14. As shown in FIG. 4, the DSP device 14 has a transfer filter T _n for identifying the sound field characteristic T _n of the first transfer path. For identification, for example, noise from an assumed noise source is used. The DSP device 14 has a function of inputting the noise signal x (t) output from the reference microphone 11 to the transfer filter T _n and calculating the output signal p _n (t). Then, a signal u ₃ (t) obtained by electrically subtracting the calculated output signal p _n (t) from the noise signal e _n (t) collected by the proximity characteristic detection microphone 12 along the first transmission path. It has a function to calculate. The transfer filter T _n has a function of minimizing the value of the signal u ₃ (t). Specifically, the filter coefficient of the transfer filter T _n is determined so that the value of the signal u ₃ (t) is minimized, and the sound field characteristic T _n of the first transfer path is identified.

Next, the details of the second transmission path identification unit for identifying the sound field characteristic _Tf of the second transmission path from the noise source to the control point P will be described with reference to FIG. FIG. 5 is a block diagram illustrating a control circuit of the second transmission path identification unit included in the DSP device 14. As shown in FIG. 5, the DSP device 14 has a transfer filter T _f for identifying the sound field characteristic T _f of the second transfer path. At the control point P, a remote characteristic detection microphone 15 having the same characteristics as the proximity characteristic detection microphone 12 is disposed. For identification, for example, noise from an assumed noise source is used. The DSP device 14 has a _function of inputting the noise signal x (t) output from the reference microphone 11 to the transfer filter T _f and calculating the output signal p _f (t). Then, the signal u ₄ (t) obtained by electrically subtracting the calculated output signal p _f (t) from the noise signal e _f (t) collected by the proximity characteristic detection microphone 12 along the second transmission path. It has a function to calculate. The transfer filter T _f is configured to minimize the value of the signal u ₄ (t). Specifically, the filter coefficient of the transfer filter T _f is determined so that the value of the signal u ₄ (t) is minimized, and the sound field characteristic T _f of the first transfer path is identified.

Next, details of the noise control unit that controls noise and control sound will be described with reference to FIG. FIG. 6 is a block diagram illustrating a control circuit of a noise control unit included in the DSP device 14. As shown in FIG. 6, the DSP device 14 includes a transfer filter T _n for identifying the sound field characteristic T _n of the first transfer path, and an error filter C identified in advance by the first error path identifying unit shown in FIG. _n , a correction filter C _m calculated by a calculation described later, and a control filter W for controlling the control sound of the control speaker 13 are provided.

In the circuit K1 shown in FIG. 6, transmit filter T _n, the filter coefficients are configured to be updated automatically adjusted by the adaptive algorithm. As the adaptive algorithm, for example, LMS (Least Mean Square algorithm) is used.

For the transfer filter T _n , for example, a filter coefficient identified by the first transfer path identification unit illustrated in FIG. 4 is set as an initial value. Then, in the circuit K1 in FIG. 6, the output signal x of the reference microphone 11 (t) is input and outputs a signal _p n (t) through a transmission filter _{T n.} Further, the signal p _n (t) output from the transfer filter T _n and the signal y (t) obtained through the control filter W and the converter G are input to the output signal x (t) of the reference microphone 11. Then, a signal u ₅ (t) obtained by subtracting the signal c _n (t) output through the error filter C _n from the output signal e (t) of the proximity characteristic detection microphone 12 is input. The transfer filter T _n is updated by adaptive processing so that the signal u ₅ (t) is minimized.

The noise control unit also has a function of calculating the error path characteristic C _m from the control speaker 13 to the proximity characteristic detection microphone 12 by calculation. The sound field characteristic C _m is calculated based on the sound field characteristic C _f identified in advance by the second error path identification unit shown in FIG. This calculation method assumes a linear phase system having a waveform that resembles the temporal and spatial progression of sound waves, and finds the method 1 by changing the phase and amplitude of the sound field characteristic C _f , the transmission path There is a method 2 in which the sound field characteristic C _f is changed using the ratio of the transfer characteristics T _n and T _f which are the sound field characteristics of the sound field.

  First, method 1 will be described with reference to FIG. FIG. 7 is a schematic diagram for explaining the calculation for obtaining the error path. FIG. 7A shows a waveform when a control sound is output from the control speaker 13 located in the vicinity of the proximity characteristic detection microphone 12. A waveform detected by the proximity characteristic detection microphone 12 is indicated by S1, and waveforms at points P1 and P2 arranged farther than the proximity characteristic detection microphone 12 are indicated by S2 and S3. As shown in FIG. 7A, when sound is output from the control speaker 13, the waveform S1 detected by the proximity characteristic detection microphone 12 depends on the structure of the sound pressure driving surface and the support structure of the control speaker 13. It becomes turbulent under the influence and becomes a non-linear phase system. On the other hand, when the waveform S1 of the non-linear phase system propagates to the point P1, the influence of the control speaker 13 is reduced, and the waveform S2 is free of disturbance. And if waveform S2 propagates to point P2, it will become waveform S3 from which only an amplitude and a phase differ compared with waveform S2. Thus, the waveform at a point away from the control speaker 13 becomes a linear phase system that resembles the sound wave in time and space.

  The characteristics of the acoustic propagation path can be expressed as a simple phase delay and an amplitude attenuation with distance. Therefore, the linear phase waveform S1 can be calculated using the linear phase waveforms S2 and S3 at the points P1 and P2. That is, by obtaining the phase difference and the amplitude difference between the waveforms S2 and S3, it is possible to calculate the linear phase system waveform S1 that would be detected by the proximity characteristic detection microphone 12 by going back in time. FIG. 7B shows a waveform at each point when output from the control speaker 13 a arranged away from the proximity characteristic detection microphone 12. As shown in FIG. 7B, the control speaker 13 is moved away from the proximity characteristic detection microphone 12 by converting the non-linear phase waveform S1 shown in FIG. 7A into the linear phase waveform S1. It will be the same. By using the waveform S1 shown in FIG. 7B, it is possible to suitably predict the waveforms at the points P1 and P2 using the waveform S1.

In method 1, assuming the above-described linear phase system, the sound field characteristic C _m (correction filter C _m ) is calculated by changing only the phase and intensity without changing the frequency spectrum of the sound field characteristic C _f . The sound field characteristic T _n identified in advance by the first transmission path identification unit shown in FIG. 4 or the sound field characteristic T _n derived by the circuit K1 shown in FIG. 6 and the second transmission path identification unit shown in FIG. based on the previously identified sound field characteristic T _f, to calculate the phase and intensity of the sound field characteristic C _m.

First, the phase change will be described. For example, the frequency dependence of the phase in the control frequency band of the sound field characteristics T _n and T _f is linearly approximated by a linear function of frequency. Then, the sound field characteristic T _n, the difference between the linear function of the frequency used for the approximation of T _f, is subtracted from the sound field characteristic C _f, to calculate the sound field characteristic C _m. When the phase of the sound field characteristics T _n and T _f can be approximated to a linear function of frequency, that is, when the phase is a linear phase, it can be expressed that the group delay of the sound field characteristics T _n and T _f is constant. Therefore, in other words the processing, the sound field characteristic T _n, the average value of the group delay of T _f, is subtracted from the sound field characteristic C _f, it can be said that calculating the sound field characteristic C _m. This phase correction will be described with reference to FIG. 8 (a) is the frequency f dependency of the phase θ of the sound field characteristic T _n, FIG. 8 (b) is a graph showing the frequency f dependency of the phase θ of the sound field characteristic T _f, the actual frequency The dependency is indicated by a solid line, and a straight line approximation is indicated by a dotted line. Further, the control frequency band is from f (low) to f (high). As shown in FIGS. 8A and 8B, the frequency dependence of the phase of the sound field characteristics T _n and T _f shown by solid lines is expressed by linear functions K _n · f and K _f · f of frequencies shown by dotted lines. Approximate each straight line. This is an approximation assuming that the differential value (group delay) of the phase θ by the frequency f is constant. Then, the sound field characteristic C _m is calculated by subtracting the difference (K _n −K _f ) · f between the linear functions K _n · f and K _f · _f from the sound field characteristic C _f .

Next, the change in strength will be described. Intensity of the sound field characteristic C _f, for example, is calculated by using the average value of the amplitude difference of the sound field characteristic T _n, T _f of the control frequency bands. By changing the sound field characteristic C _f based on the calculated phase and intensity, sound field characteristic C _m is calculated.

The sound field characteristic C _m is expressed by the following equation 1 where α (a constant depending on the frequency f) and β (a constant independent of the frequency f) are the phase corrections of the sound field characteristic C _f described above.

In Method 1, the sound field characteristic C _m is calculated by changing the phase α and the gain β in Equation 1.

Next, method 2 for changing the sound field characteristic C _f using the ratio of the transfer characteristics T _n and T _f will be described. In the case of the method 2, the sound field characteristic _Cm is expressed by the following formula 2.

Unlike Equation 1, Equation 2 is a mathematical equation that can be established without assuming a linear phase system. When expressed by the polar coordinates of Equation 2, Equation 3 below is obtained.

As shown in Equation 3, the ratio of the transfer characteristics T _n and T _f | T _n (f) | / | T _f (f) | corrects the amplitude of the sound field characteristic C _m , and exp [j · {θ · T _f (f) −θ · T _n (f)}] is a term for correcting the phase of the sound field characteristic C _m . _That, T f, _T difference of the primary function of the frequency f used for approximation of _{n {θ · T f (f} ) -θ · T n (f)} , the phase for correction for subtracting from the phase of the _{C f} Used as Thus, in Method 2, both the amplitude and phase are functions of the frequency f. For this reason, even if the sound field characteristic _Cm is deformed due to reflection or the like, it is possible to accurately correct by the above equation 3.

Here, the mathematical validity of Equation 2 is verified. The control filter W that controls the proximity characteristic detection microphone 12 disposed in the vicinity of the control speaker 13 by the direct method is represented by the following Expression 4.

In Expression 4, when the sound field characteristic C _n is replaced with the sound field characteristic C _m of Expression 1, the following Expression 5 is obtained.

Equation 5 is nothing but a control filter at a remote control point. The physical interpretation of Equation 2 will be described. Assuming a linear phase system, Equation 2 is _equivalent to the phase difference (time difference) between the sound field characteristic T _n and the sound field characteristic T _f while maintaining the waveform of the sound field characteristic C _f , that is, the proximity characteristic detection microphone 12. Is equivalent to backshifting to the position of Therefore, unlike the conventional method, the sound field characteristic T _n and the sound field characteristic T _f can be controlled even when the conditions are different.

Returning to FIG. 6, the circuit K2 will be described. The circuit K2 includes a transfer filter T _n , a correction filter C _m , and a control filter W. Transmission filter T _n is a copy of the transmission filter T _n identified in the circuit K1 is used. Correction filter C _m are those calculated by any of the methods 1 and 2 described above is used. The control filter W is configured such that the filter coefficient is automatically adjusted and updated by an adaptive algorithm. For example, LMS is used as the adaptive algorithm.

The LMS output signal x of the reference microphone 11 (t) has a function of inputting a signal r (t) which is output via the inputted correction filter C _m contained in the circuit K2. Also, from the signal x (t) signal is via the transmission filter T _n are input output q (t), the signal r (t) is input and output via the control filter W signal s (t) is It has a function of adaptively updating the control filter W so that the value of the electrically subtracted signal u ₆ (t) is minimized.

  Further, the circuit K3 includes a control filter W and a converter G having a control signal that passes through the control filter W as an opposite phase. As this control filter W, a copy of the control filter W identified by the circuit K2 is used. Then, the signal x (t) from the reference microphone 11 is input, a control signal is generated via the control filter W, the generated control signal is converted into a signal y (t) having an opposite phase by the converter G, and control is performed. A function of outputting from the speaker 13 is provided.

  Since the active noise control apparatus has the control circuit shown in FIG. 6, the noise at the control point P is canceled as in the case where the microphone is arranged at the control point P without arranging the microphone at the control point P. Control sound can be output.

  Next, the operation of the silencing system 6 having the active noise control device according to the present embodiment will be described.

As shown in FIG. 1, the control speaker 13 is installed so that the arrival directions of the noise D1 and the control sound D2 at the control point P are the same. Then, before executing the noise control, for example, M-sequence noise corresponding to the control signal of the DSP device 14 is output from the control speaker 13. As shown in FIG. 2, output M-sequence noise signals are picked up to reach the vicinity of characteristic detection microphone 12 follows the first error path, is converted into an electric signal e _{n (t).} The converted electric signal _e n (t) is input to the DSP device 14. Then, the first error path identification unit identifies the characteristic C _n of the first error path so that the subtraction result u ₁ (t) is minimized. Before executing noise control, the far-field characteristic detection microphone 15 is disposed at the control point P, and for example, M-sequence noise corresponding to the control signal of the DSP device 14 is output from the control speaker 13. As shown in FIG. 3, the output M-sequence noise signal follows the second error path, reaches the far characteristic detection microphone 15, is collected, and is converted into an electric signal e _f (t). The converted electrical signal e _f (t) is input to the DSP device 14. Then, the second error path characteristic C _f is identified by the second error path identification unit so that the subtraction result u ₂ (t) is minimized.

Next, the noise is output from the noise source S before the noise control is executed. Or you may provide the control speaker which outputs the noise close | similar to the noise assumed in pseudo | simulation, and may output a noise. As shown in FIG. 4, the output noise is picked up by the reference microphone 11 and converted into an electric signal x (t). The converted electrical signal x (t) is input to the DSP device 14. Meanwhile, the noise output from the noise source S, picked up to reach the vicinity of characteristic detection microphone 12 follows the first transmission path, is converted into an electric signal e _{n (t).} The converted electric signal _e n (t) is input to the DSP device 14. Then, the first transmission path identification unit identifies the characteristic T _n of the first transmission path so that the subtraction result u ₃ (t) is minimized. Further, the far-field characteristic detection microphone 15 is arranged at the control point P before the noise control is executed, and the noise is output from the noise source S. As shown in FIG. 5, the output noise is picked up by the reference microphone 11 and converted into an electric signal x (t). The converted electrical signal x (t) is input to the DSP device 14. On the other hand, the noise output from the noise source S reaches the far characteristic detection microphone 15 along the second transmission path, is collected, and is converted into an electric signal e _f (t). The converted electrical signal e _f (t) is input to the DSP device 14. Then, the second transmission path identification unit identifies the characteristic T _f of the second transmission path so that the subtraction result u ₄ (t) is minimized.

Thus, before the active noise control, the sound field characteristic C _n , the sound field characteristic C _f , the sound field characteristic T _n , and the sound field characteristic T _f are identified. Then, the remote characteristic detection microphone 15 arranged at the control point P is removed.

  Next, the noise to be controlled is output from the noise source S. As shown in FIG. 6, the output noise is picked up by the reference microphone 11 and converted into an electric signal x (t). The converted electrical signal x (t) is input to the DSP device 14. In the circuit K3, the electrical signal x (t) becomes an output signal via the control filter W set in the circuit K2, and further becomes an antiphase signal y (t) via the converter G. The signal y (t) is output from the control speaker 13 as a control signal. Then, at the control point P, the output control sound has a phase opposite to that of the noise and cancels out due to phase interference with each other. As described above, the DSP device 14 outputs the control sound corresponding to the noise from the control speaker 13, and feedforward control is performed so as to produce a silencing effect at the control point P.

Then, the transfer filter T _n is adjusted by the LMS of the circuit K1 so that the electric signal u ₅ (t) as a calculation result is minimized. The electric signal c _n (t) is subtracted from the electric signal e ₅ (t) in order to remove the influence of the control sound. As a result, even if the sound field characteristic T _n of the propagation path changes with time, for example, due to a temperature change or the noise source S moves, the transfer filter T _n follows the change. Can be identified.

When the sound field characteristic T _n is identified in the circuit K1, the sound field characteristic C _m is calculated using the method 1 or the method 2 described above. If method 1, the sound field characteristic C _m by adjusting the phase and amplitude of the sound field characteristic C _f using Equation 1 is generated. On the other hand, if method 2, first using Equation 2, the sound field characteristic C _m generated by the ratio of the second transmission characteristic is integrated into the sound field characteristic C _f. Then, the control filter W is adjusted by the LMS of the circuit K2 so that the electric signal u ₆ (t) as a calculation result is minimized.

As described above, in the active noise control device and the active noise control method according to the present embodiment, the characteristics T _n of the transmission path from the noise source S to the nearby characteristic detection microphone 12, the noise source S identified in advance from the measurement, within the control region Based on the characteristic T _f of the transmission path to the predetermined control point P and the characteristic C _f of the error path from the control speaker 13 to the control point P identified in advance by measurement, the proximity characteristic detecting microphone 12 from the control speaker 13 is obtained. characteristics C _m errors route to is calculated. In general, the characteristics of an acoustic propagation path can be expressed as a simple phase delay and an amplitude attenuation with distance. For this reason, for example, the transmission path characteristic T _n from the noise source S to the proximity characteristic detection microphone 12 is compared with the transmission path characteristic T _f from the noise source S to the control point P, and control is performed based on the result. The error path characteristic C _f from the speaker 13 to the control point P is linearly converted to obtain the error path characteristic C _m from the control speaker 13 to the neighborhood characteristic detection microphone 12, so that the structure of the sound pressure driving surface and the control speaker are obtained. it is possible to calculate the characteristic C _m of error path not affected by the 13 support structure. As a result, as in the case where the microphone is arranged at the control point P, it is possible to achieve a silencing effect also on the downstream side of the control point P, and the proximity characteristic detection microphone 12 can be arranged in the vicinity of the control speaker 13. it can.

  Further, according to the active noise control device and the active noise control method according to the present embodiment, the proximity characteristic detection microphone 12 can be disposed close to the control speaker 13, so that, for example, the control speaker 13 and the proximity characteristic detection microphone 12 are arranged. And a compact system can be constructed. In addition, since it is not necessary to place a microphone in the control area, the present invention can be applied even when there is no place to install the microphone, or when it becomes an obstacle to activities when installed. Further, unlike the conventional method, a position having a sufficient distance from the proximity characteristic detection microphone 12 can be set as the control point P.

  Furthermore, according to the active noise control device and the active noise control method according to the present embodiment, the proximity characteristic detection microphone 12 can be disposed close to the control speaker 13, so the distance from the noise source S to the control speaker 13. And the distance from the noise source S to the proximity characteristic detection microphone 12 can be handled as being substantially equal. This will be described in detail with reference to FIGS. FIG. 9 is a plan view for explaining the action and effect when the proximity characteristic detection microphone 12 is brought close to the control speaker 13, and FIG. 10 is for explaining the action and effect when the control speaker 13 and the vicinity characteristic detection microphone 12 are separated. It is a top view. 9 and 10, the noise source before movement is N1, and the noise source after movement is N2. As shown in FIG. 9, before the movement, the noise source N1, the control speaker 13, and the proximity characteristic detection microphone 12 are arranged on a straight line X1. For this reason, a region having a silencing effect spreads over a wide range at the control point P3 and downstream thereof. On the other hand, when the noise source N1 moves to the noise source N2, since the distance L between the control speaker 13 and the proximity characteristic detection microphone 12 is small, the distance from the noise source N2 to the control speaker 13 and the proximity characteristic detection microphone from the noise source N2 The distance up to 12 can be handled as being almost equal. For this reason, the distance L connecting the control speaker 13 and the proximity characteristic detection microphone 12 can be handled as a point. Thereby, the noise source N2, the control speaker 13, and the proximity characteristic detection microphone 12 can be handled as being arranged on the straight line X2. Therefore, with the movement of the noise source, the control point P3 moves in a direction that is point-symmetric with the position of the noise source N2 with the control speaker 13 and the neighborhood characteristic detection microphone 12 as the symmetry points, and becomes the control point P4. At the control point P4, the propagation direction of the sound Ci2 from the noise source N2 and the propagation direction of the sound Ci1 from the control speaker 13 can be overlapped, so that a region having a silencing effect can be provided in a wide range downstream of the control point P4. Can be formed. On the other hand, as shown in FIG. 10, when the distance L between the control speaker 13 and the proximity characteristic detection microphone 12 is large, the propagation direction of the noise output from the noise source N <b> 2 after the movement and the output from the control speaker 13. Since the direction of propagation of the control sound intersects with an angle, it produces a silencing effect near the control point P4, while increasing the sound in other areas. As described above, according to the active noise control device and the active noise control method according to the present embodiment, the proximity characteristic detection microphone 12 can be disposed close to the control speaker 13, so that the area area that provides the silencing effect is reduced. It becomes possible to avoid doing. The effect can be increased as the distance L between the control speaker 13 and the proximity characteristic detection microphone 12 is decreased.

Further, according to the active noise control apparatus according to the present embodiment, the transmission path characteristic T _n from the noise source S to the proximity characteristic detection microphone 12, and the transmission path characteristic T _f from the noise source S to the control point P Are compared to calculate a phase difference and an amplitude difference, and the characteristic C _f of the error path from the control speaker 13 to the control point P can be linearly converted using the phase difference and the amplitude difference. In this way, characteristics that are not affected by the structure of the sound pressure driving surface or the support structure of the speaker, utilizing the properties of the linear phase system that has a similar waveform with the temporal and spatial progression of sound waves Thus, the characteristic C _m of the error path from the control speaker 13 to the proximity characteristic detecting microphone 12 can be calculated.

Further, according to the active noise control apparatus according to the present embodiment, the error path characteristic C _f from the control speaker 13 to the control point P and the transmission path characteristic T _n from the noise source S to the nearby characteristic detection microphone 12 are The ratio with the characteristic T _f of the transmission path from the noise source S to the control point P identified in advance by measurement can be integrated. As a result, the sound waveform of the error path from the control speaker 13 to the control point P is maintained, and the characteristic T _n of the transmission path from the noise source S to the nearby characteristic detection microphone 12 and the measurement are identified in advance. It is possible to shift only the phase difference from the characteristic T _f of the transmission path from the noise source S to the control point P, that is, the position of the proximity characteristic detecting microphone 12. Therefore, the characteristics of the error path that is not affected by the structure of the sound pressure driving surface and the speaker support structure can be calculated by a simple calculation.

  The embodiment described above shows an example of the active noise control device and the active noise control method according to the present invention. The active noise control device and the active noise control method according to the present invention are not limited to the active noise control device and the active noise control method according to the embodiment, and the embodiment is within a range not changing the gist described in each claim. The active noise control device and the active noise control method according to the above may be modified or applied to others.

For example, in the above-described embodiment, as Method 1, an example of calculating the phase and intensity of the sound field characteristic C _f using the average value of the group delay difference and the average value of the amplitude difference of the sound field characteristics T _n and T _f is used. Although described, it is not limited to the average value, and may be a median value or the like.

  Hereinafter, examples and comparative examples implemented by the present inventors will be described in order to explain the above effects.

Example 1
A silencer system having an active noise control device shown in FIGS. 11 and 12 was used. 11 and 12 are schematic views of the experimental environment, FIG. 11 is a diagram of the experimental environment viewed from the side, and FIG. 12 is a diagram of the experimental environment viewed from above. As shown in FIG. 11, a simulated speaker 13a (primary sound source N1) was substituted for the noise source. The primary sound source N1 was fixed at a height h1 (1240 mm). An active noise control device having a reference microphone 11, a secondary sound source as a control speaker 13, and a proximity characteristic detection microphone 12 was installed. The heights of the reference microphone 11, the secondary sound source, and the proximity characteristic detection microphone 12 were adjusted to the height h1 of the primary sound source N1. The reference microphone 11 was provided at a position 300 mm away from the installation point of the primary sound source N1. The secondary sound source was provided at the upper end of the partition 21 at a position d ₁ (1200 mm) away from the installation point of the primary sound source N1. The proximity characteristic detection microphone 12 was provided at a position d ₂ (150 mm) away from the installation point of the secondary sound source. The control point P was provided at a position separated by d ₃ (1500 mm) from the installation point of the secondary sound source. At the control point P, a verification microphone 12a is disposed. Then, the error filter C _n , the error filter C _f , the transfer filter T _n , and the transfer filter T _f were identified, noise was output from the primary sound source N1, and feedforward control was performed by the control circuit shown in FIG. The correction filter _Cm was obtained by changing the phase and amplitude of the error filter _Cf (method 1). Then, as shown in FIG. 12, the origin is O, and the noise reduction effect in the region d5 × d5 (3000 mm × 3000 mm) including the control point P was evaluated. Note that the evaluation of the silencing effect was based on the difference in sound pressure level (dB) when the control sound was output and when it was not output. The results are shown in FIG.

(Example 2)
The correction filter _Cm was obtained using the ratio of the transfer characteristics _Tn and _Tf (method 2). Others are the same as the first embodiment. The results are shown in FIG.

(Comparative Example 1)
In the silencing system shown in FIG. 11, general active noise control by a direct method is performed by causing the verification microphone 12 a arranged at the control point P to function as an error microphone.
Others are the same as the first embodiment. The results are shown in FIG.

(Comparative Example 2)
Without using the correction filter C _m, it was subjected to active noise control using an error filter C _n. Others are the same as in the first embodiment. The results are shown in FIG.

  FIGS. 13 to 16 are diagrams showing the difference in sound pressure level (dB) in shades when the control sound is output and when it is not output. As shown in FIG. 15, in the active noise control device of Comparative Example 1 in which the microphone is actually arranged at the control point P, it has been confirmed that the area where the silencing effect is exerted is widened at the control point P and its downstream side. . On the other hand, as shown in FIG. 16, in Comparative Example 2 in which the proximity characteristic detection microphone 12 is placed in the vicinity of the secondary sound source without placing a microphone at the control point P, the silencing effect is played only in the vicinity of the proximity characteristic detection microphone 12. It was confirmed that

  On the other hand, as shown in FIGS. 13 and 14, in Examples 1 and 2, it was confirmed that the control point P and the region where the silencing effect is exerted spread on the downstream side. Moreover, it was confirmed that FIGS. 13 and 14 have no difference in the effect even when compared with FIG. That is, it was confirmed that Examples 1 and 2 have the same effect as the case where the microphone is installed at the control point P without arranging the microphone at the control point P.

  DESCRIPTION OF SYMBOLS 11 ... Reference microphone, 12 ... Proximity characteristic detection microphone, 13 ... Control speaker (control sound source), 14 ... DSP device (identification means, error path calculation means, control filter construction means, feedforward control means).

Claims (4)

An active noise control device that outputs a control sound to a control region and controls noise in the control region,
A control sound source that is disposed between a noise source and the control region and outputs a control sound to the control region side;
A reference microphone that is arranged between the noise source and the control sound source and detects noise output from the noise source ;
Juxtaposed prior Symbol control sound source while being disposed between the control sound source and the control area, and the neighboring property detecting microphone for detecting noise and the control sound,
Identifying means for identifying a characteristic of a transmission path from the noise source to the nearby characteristic detection microphone;
Characteristics of the transmission path from the noise source to the proximity characteristic detection microphone, characteristics of the transmission path from the noise source identified in advance by measurement to a predetermined control point in the control region, and identification in advance by measurement An error path calculation means for calculating a characteristic of an error path from the control sound source to the neighboring characteristic detection microphone based on a characteristic of an error path from the control sound source to the control point;
Based on the characteristics of the transmission path from the noise source to the proximity characteristic detection microphone and the characteristics of the error path from the control sound source to the proximity characteristic detection microphone, the control sound and noise are caused to phase interfere at the control point and Control filter constituting means constituting a control filter for reducing noise in the control region;
Based on the detection result of the reference microphone and the control filter, feedforward control means for feedforward controlling the control sound of the control sound source;
An active noise control device comprising: The error path calculation means includes
Based on the characteristics of the transmission path from the noise source to the nearby characteristic detection microphone, calculate the phase and amplitude related to the frequency band to be controlled,
Based on the characteristics of the transmission path from the noise source identified in advance by measurement to the control point, the phase and amplitude relating to the frequency band to be controlled are calculated,
Based on the calculated phase and amplitude of both, the neighboring characteristics are detected from the control sound source by correcting the phase and amplitude among the characteristics of the error path from the control sound source to the control point identified in advance by measurement. Calculating the characteristics of the error path to the microphone,
The active noise control apparatus according to claim 1. The error path calculation means includes
The characteristic of the transmission path from the noise source to the proximity characteristic detection microphone is divided by the characteristic of the transmission path from the noise source identified in advance by measurement to the control point, and the value obtained by the division is measured. Calculating an error path characteristic from the control sound source to the neighboring characteristic detection microphone by integrating the error path characteristic from the control sound source identified in advance to the control point;
The active noise control apparatus according to claim 1. A control sound source that is arranged between a noise source and a control region and outputs control sound to the control region side, and is arranged between the noise source and the control sound source to detect noise output from the noise source. Active noise control of an active noise control apparatus , comprising: a reference microphone that is arranged between the control sound source and the control region, and a proximity characteristic detection microphone that is arranged in parallel with the control sound source and detects noise and control sound A method,
An identification step for identifying a characteristic of a transmission path from the noise source to the proximity characteristic detection microphone;
The characteristics of the transmission path from the noise source identified by the identification step to the proximity characteristic detection microphone, the characteristics of the transmission path from the noise source identified in advance by measurement to a predetermined control point in the control region, and An error path calculation step of calculating an error path characteristic from the control sound source to the neighboring characteristic detection microphone based on an error path characteristic from the control sound source identified in advance by measurement to the control point;
Based on the characteristics of the transmission path from the noise source to the proximity characteristic detection microphone and the characteristics of the error path from the control sound source to the proximity characteristic detection microphone, the control sound and noise are caused to phase interfere at the control point and A control filter configuration step for configuring a control filter for reducing noise in the control region;
Based on the detection result of the reference microphone and the control filter, a feedforward control step for feedforward controlling the control sound of the control sound source;
An active noise control method comprising: