JP2008249702A - Acoustic measurement device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an acoustic measurement device and method capable of accurately and easily measuring directional information of sound represented by acoustic intensity. <P>SOLUTION: The acoustic measurement device includes a sound receiving section 11 for receiving sound using an unidirectional microphone pair arranged so that the directivity is inverted by 180 degrees, an arithmetic processing section 12 for performing arithmetic processing of directional information of the sound including the arrival direction of the sound using the output difference of the unidirectional microphone pair, and an output section 13 for outputting directional information of the sound. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an acoustic measurement apparatus and an acoustic measurement method for measuring sound direction information represented by acoustic intensity, for example.

  The sound intensity is a quantity that represents a flow of sound energy per unit time, and is a vector quantity having a magnitude and a direction. This sound intensity is applied to, for example, noise identification for searching for a noise source, a monitoring camera system, and the like.

  Conventionally, a special microphone called an intensity microphone in which two microphones are combined at very close intervals has been used for measuring sound intensity. And the sound intensity was measured by using this intensity microphone and a 2-channel type FFT analyzer or a 1 / N octave analyzer.

  Further, the present inventor has proposed to obtain the sound source direction and the sound source level using a database of difference in level of microphones arranged in opposite directions (see, for example, Patent Document 1).

International Publication No. 06/054599 pamphlet

  Conventional intensity measurement methods are mainly performed by combining two omnidirectional sound pressure type microphones, so they are sensitive to the sensitivity difference and phase difference of the microphones, and these must be strictly managed. In addition, the microphone interval had to be changed according to the frequency. In addition, when the sound intensity is obtained using a database stored in advance, the accuracy of the sound intensity may not be satisfied.

  The present invention has been proposed in view of such a conventional situation, and provides an acoustic measurement device and an acoustic measurement method capable of measuring sound direction information typified by acoustic intensity with high accuracy and ease. The goal is to provide.

  In order to achieve the above-described object, an acoustic measurement apparatus according to the present invention includes an output difference calculation unit that calculates an output difference of a unidirectional microphone positioned 180 degrees opposite on a predetermined axis, and the output difference. And an axial direction component calculating means for calculating a direction component of the predetermined axis based on the axis.

  The acoustic measurement method according to the present invention includes an output difference calculating step of calculating an output difference of a unidirectional microphone positioned 180 degrees opposite on a predetermined axis, and the direction of the predetermined axis based on the output difference. And an axial direction component calculating step for calculating the component.

  In addition, the acoustic measurement apparatus according to the present invention includes a sound receiving means for receiving sound using a unidirectional microphone pair in which directivity is arranged in the opposite direction by 180 degrees on each axis of orthogonal coordinates, and the unidirectional Output difference calculating means for calculating the output difference obtained from the pair of directional microphones, direction information orthogonal basis component calculating means for calculating the orthogonal basis component of the direction information of each axis based on the output difference, and the direction of each axis Sound direction information calculating means for calculating sound direction information including the direction of arrival of the sound based on the information orthogonal basis component.

  Further, the acoustic measurement method according to the present invention includes a sound receiving process for receiving sound using a unidirectional microphone pair in which directivity is arranged in opposite directions by 180 degrees on each axis of orthogonal coordinates, and the unidirectional An output difference calculation step for calculating an output difference obtained from the pair of directional microphones, a direction information orthogonal basis component calculation step for calculating an orthogonal basis component of direction information of each axis based on the output difference, and a direction of each axis A sound direction information calculating step of calculating sound direction information including the direction of arrival of the sound based on the information orthogonal basis component.

  The acoustic measurement apparatus according to the present invention includes a sound receiving unit that receives sound by rotating a directional microphone, a squared sound pressure calculating unit that calculates a squared sound pressure of the directional microphone, and a Fourier sound pressure that is a Fourier transform of the squared sound pressure. The present invention is characterized by comprising arithmetic means for performing series expansion and calculating a primary Fourier coefficient, and acoustic intensity calculating means for calculating an acoustic intensity including a sound arrival direction based on the primary Fourier coefficient.

  According to the present invention, in order to measure the sound direction information including the sound arrival direction using the output difference obtained from the pair of unidirectional microphones arranged in opposite directions on each axis of the orthogonal coordinates, Sound intensity can be measured accurately and easily.

  First, the principle of measurement of sound intensity based on Fourier series expansion of sound direction information according to the present invention will be described. In this specification, sound direction information refers to a physical quantity related to sound whose direction is a parameter, such as directivity characteristics of speakers and microphones, and reflected sound energy by direction of the sound field.

  FIG. 1 is a conceptual diagram for explaining sound direction information. The direction information can be expressed by a time waveform x (θ, t) for each direction or a complex amplitude X (θ, ω) for each direction. Here, for the sake of simplicity, a function f (θ) in which parameters of time and angular frequency are omitted will be considered as direction information, and two-dimensional will be described as an example.

  Since the basic period T is 2π, the two-dimensional direction information f (θ) is represented by the Fourier series expansion of the equation (1) using the direction (angle) θ as a parameter. Further, the Fourier coefficients are expressed as in the equations (2) to (4), respectively.

  Here, since n represents the number of vibrations per round, it can be considered as the frequency of the direction information. Therefore, it is possible to evaluate a viewpoint different from the conventional one based on the frequency characteristics of the direction information. For example, if a microphone or speaker has a sharp directivity, it contains a lot of high direction frequency components, so that it can be applied as an index of directivity sharpness. Conversely, since omnidirectionality is a direct current component of direction information, the degree of omnidirectionality of the speaker and microphone and the isotropy of the sound field can be evaluated by the ratio of the direct current component to the total energy.

  Further, as shown in FIG. 1, when sound direction information f (θ) is received in the φ direction by a microphone having directivity characteristics m (θ) at a certain sound receiving point, the microphone output rφ obtained is It is expressed as equation (5).

  When the microphones are sequentially rotated and the direction information r (φ) in all directions is received using φ as a parameter, the equation (6) is obtained.

  This is a cyclic convolution on the direction axis between the direction information f (φ) of the sound field and m (−φ) obtained by reversely rotating the directional characteristic of the microphone.

  The present inventor proposes a method for measuring sound intensity using these principles.

  Hereinafter, specific embodiments to which the present invention is applied will be described in detail with reference to the drawings.

(First embodiment)
FIG. 2 is a block diagram illustrating a configuration of the acoustic measurement device according to the first embodiment. The acoustic measurement apparatus 10 includes a sound receiving unit 11 that receives sound using a unidirectional microphone pair in which directivity is arranged in the opposite direction by 180 degrees on each axis of orthogonal coordinates, and an output difference between the unidirectional microphone pair. Is provided with an arithmetic processing unit 12 that performs arithmetic processing of sound direction information including the direction of sound arrival, and an output unit 13 that outputs sound direction information.

  The sound receiving unit 11 has a plurality of microphones that have a predetermined directivity characteristic D and receive a signal coming from a sound source. These microphones are arranged in pairs with opposite directivities of 180 degrees on the axis of the orthogonal coordinate serving as a reference, for example, on the x axis, the y axis, and the z axis.

  FIG. 3 is a schematic diagram illustrating an example of the configuration of the sound receiving unit 11. The sound receiving unit 11 has a microphone 21A in which the directivity characteristic D is directed toward the 0 degree direction ((reference) forward direction), and the directivity characteristic D is directed toward the 90 degree direction (right direction) with respect to the microphone 21A. The microphone 21B that is disposed, the microphone 21C that is disposed with the directivity characteristic D 180 degrees (rearward) with respect to the microphone 21A, and the directivity D that is 270 degrees with respect to the microphone 21A (leftward). The microphone 21D arranged in the direction, the microphone 21E (upper vertical direction) in which the directivity characteristic D is oriented in the vertical direction with respect to the microphones 21A to 21D, and the directivity characteristic D in the 180-degree direction with respect to the microphone 21E. It consists of a microphone 21F arranged in the (downward vertical direction). Note that the sound receiving unit 11 may include four microphones 21A to 21D. Also, in the microphone arrangement, the acoustic center becomes a problem, but all microphones may be installed at substantially the same point, and the acoustic centers may be aligned, or the distance from any one point to the acoustic center of all microphones is equal. You may arrange | position so that it may become.

  In addition, the directivity characteristic D of the microphones 21A to 21F may be unidirectional, and may be any of cardioid, super cardioid, hyper cardioid, and ultra cardioid.

  As will be described later, the arithmetic processing unit 12 calculates an orthogonal basis component of the direction information using the output difference of the unidirectional microphone pair in the opposite direction in the sound receiving unit 11 and obtains a primary Fourier coefficient of the direction information. Thus, the sound direction information including the sound arrival direction is calculated.

  The output unit 13 outputs sound direction information such as a difference in phase angle, amplitude, and squared sound pressure calculated by the calculation processing unit 12 to, for example, a display monitor.

  According to such an acoustic measurement device 10, it is possible to measure the direction information of the sound represented by acoustic intensity with high accuracy and ease.

  Subsequently, specific arithmetic processing in the arithmetic processing unit 12 will be described in detail. Here, by forming an orthogonal base using a unidirectional microphone, the direction information of the sound field is decomposed into orthogonal components, and the sound source direction is estimated.

  Intensity is defined by the product of sound pressure p and particle velocity u as shown in equation (7). Although the current mainstream intensity measurement method is almost entirely based on this definition formula, this embodiment is based on the definition of formula (8) that does not use particle velocity.

  In equation (8), the right side is a scalar quantity even though the intensity on the left side is a vector. Therefore, the following is considered in order to vectorize the right side. First, consider a one-dimensional (x-axis) sound field. It is assumed that sound waves propagate in the positive and negative directions of the x axis. The relationship between instantaneous sound pressure and particle velocity in the interference sound field can be expressed by the following equation.

  However, the symbol with a shoulder represents a component propagating in the plus direction or the minus direction with respect to the x direction. Substituting these into equation (7) yields equation (12).

  The sound pressure itself is a scalar quantity, but this equation (12) indicates that if the instantaneous squared sound pressure for each propagation direction is known, the intensity of the vector quantity can be measured without measuring the particle velocity. ing. The average (active) sound intensity may be obtained by replacing the instantaneous square sound pressure of the equation (12) with the mean square sound pressure. Further, equation (12) is a one-dimensional intensity, but in order to expand this to three dimensions, an orthogonal basis component of the intensity with respect to the x, y, and z axes may be obtained by equation (12).

  Next, a method for calculating the sound intensity will be described.

  Since cardioids and hypercardioids are a combination of omnidirectionality and bidirectionality, the directivity characteristic m (θ) can be expressed as in equation (13) in the dimension of the sound pressure amplitude.

  However, A + B = 1, the cardioid is A = 0.5, the super cardioid is A = 0.37, and the hyper cardioid is A = 0.25.

  As shown in FIG. 4, for example, the sensitivity difference between two microphones directed to 0 degrees and 180 degrees on the reference orthogonal coordinate axis is expressed by the equation (14) in the dimension of sound pressure, which is a cos function.

  Further, as shown in FIG. 5, the sensitivity difference between the microphone pair at 90 degrees and 270 degrees is expressed by equation (15), which is a sin function.

  In the case of a cardioid as shown in FIGS. 4 and 5, the equations (14) and (15) are cos θ and sin θ, respectively.

  Further, the sensitivity difference of the microphone pair toward 0 degree and 180 degrees in the square sound pressure dimension is expressed by the equation (16), which is also a cos function.

  Similarly, the sensitivity difference of the microphone pair toward 90 degrees and 270 degrees is a sin function as shown in equation (17).

  FIG. 6 is a graph showing a difference in squared sound pressure sensitivity between cardioid microphone pairs at 0 degrees and 180 degrees. That is, the equation (16) in the case of cardioid is graphed.

  As can be seen from FIG. 6, since cos and sin are orthogonal to each other, calculating two pairs of opposite microphone sensitivity differences is equivalent to decomposing direction information into orthogonal components. Using the above relationship, the first-order Fourier coefficient of the direction information f (θ) in the above formulas (3) and (4) is directly calculated using the output difference of the unidirectional microphone in the opposite direction as follows. can do.

  When the equation (14) is transformed into the equation (18) and n = 1 is substituted into the equation (3), the equation (19) is obtained.

  Further, when the formula (5) is applied to this, the formula (20) is obtained.

Further, b ₁ can be obtained in the same manner.

  Further, from the equations (16) and (17), the difference between the square sound pressures of the microphones facing in the opposite direction can be decomposed into orthogonal components as in the equations (22) and (23) using the same concept as described above.

Since the difference a ₁ and b ₁ of the squared sound pressures by the two unidirectional microphones in the opposite directions are the direction information orthogonal basis components of the x-axis and the y-axis, respectively, the phase angle φ is expressed by equation (24). Is calculated by

  This phase angle φ coincides with the direction of arrival of sound when there is one sound source in free space. That is, this phase angle φ is set as the sound arrival angle θ.

  The amplitude D of the squared sound pressure is calculated by equation (25).

Further, the magnitude I _xy of the two-dimensional sound intensity is calculated by the equation (26).

  In addition, when expanding in three dimensions, another pair of microphones may be added in the vertical direction and the difference in sensitivity may be calculated.

  Decomposing the direction information into orthogonal bases in this way corresponds to obtaining each component of the vector. That is, according to the above equation (22), the difference between the squared sound pressures of the two unidirectional microphones in the x axis ± direction becomes the x-axis direction information orthogonal basis component as it is. Then, by substituting this square sound pressure difference into the equation (12), the x component of the three-dimensional intensity can be obtained. Also, the direction information orthogonal basis components of the y-axis and z-axis can be obtained in the same procedure.

  Next, the detection accuracy of the sound source direction was verified in the acoustic measurement of the first embodiment.

  FIG. 7 is a diagram schematically illustrating a sound source direction estimation experiment. As the microphone system, a total of six unidirectional microphones as shown in FIG. 3 were installed facing the ± direction of the three-dimensional orthogonal coordinates. A speaker was installed at a horizontal angle of 0 degrees and at a distance of 2 m from the microphone. Then, the microphone was rotated 11.25 degrees in the horizontal direction, and the impulse response of the microphone for 6 channels was measured in a total of 32 directions. The energy for each frequency was obtained from the impulse response of each channel, and the sound source direction was analyzed using the above-mentioned difference in squared sound pressure.

  8 to 10 are graphs showing estimation results at frequencies of 125 Hz, 1 kHz, and 8 kHz, respectively. FIG. 11 is a graph showing the maximum value and the average value of the estimation error with respect to the frequency.

  From these results, it was found that an average error up to 8 kHz can be estimated within 5 degrees. In particular, in the band of 4 kHz or less, it was found that detection can be performed with an accuracy within an average error of 1.5 degrees. In addition, it is thought that the average error becomes large at 8 kHz due to the influence of the microphone jig existing between the speaker and the microphone.

  As described above, according to the acoustic measurement device according to the first embodiment, the direction information is decomposed into orthogonal bases and the acoustic intensity is measured, so that it is possible to perform a broadband measurement without changing the microphone interval. Also, it is not necessary to align the microphone phases. In addition, the microphone intervals do not have to be strictly aligned. That is, the sound intensity can be measured with high accuracy and ease.

(Second Embodiment)
Next, a second embodiment to which the present invention is applied will be described.

  In the first embodiment, the directional microphone pair installed in opposite directions on each axis of the orthogonal coordinates is used. However, the acoustic measurement device shown as a specific example of the second embodiment is 1 The sound field is measured by rotating two microphones, and the primary component is analyzed by Fourier series expansion of the obtained direction information.

  FIG. 12 is a block diagram illustrating a configuration of an acoustic measurement device according to the second embodiment. The acoustic measurement device 30 includes a sound receiving unit 31 that receives sound by rotating a directional microphone, and a calculation processing unit 32 that performs Fourier transform on the direction information obtained by the sound receiving unit 31 and performs sound intensity calculation processing. And an output unit 33 for outputting sound intensity.

  The sound receiving unit 31 has a predetermined directivity characteristic D, rotates a microphone that receives a signal coming from a sound source, and measures direction information. In this embodiment, any microphone may be used as long as the direction information includes 1 Hz. Here, 1 Hz of the direction information means a fluctuation with an angle 2π as one cycle.

  FIG. 13 is a schematic diagram illustrating an example of the configuration of the sound receiving unit 31. The sound receiving unit 31 includes a horizontal rotation driving unit 42 that rotates the microphone 41 in the horizontal direction, a vertical rotation driving unit 43 that rotates the microphone 41 in the vertical direction, a horizontal rotation driving unit 42, and a vertical direction. And a control unit 44 that controls the rotation driving unit 43.

  The horizontal rotation drive unit 42 rotates the microphone 41 at an arbitrary angle in the horizontal direction under the control of the control unit 44.

  The vertical rotation drive unit 43 drives the microphone 41 to rotate at an arbitrary angle in the vertical direction under the control of the control unit 44.

  According to such a configuration, the microphone 41 can be driven three-dimensionally, and it is possible to obtain the same effect as in the case where the microphone 41 is configured as in the configuration example shown in FIG.

  Further, the sound receiving unit 31 may have a configuration in which a microphone 41 is disposed on a turntable 51 as shown in FIG. The turntable 51 is controlled by the drive unit 52 and is driven to rotate.

  Further, when the microphone 41 is rotated by the turntable 51, the acoustic center becomes a problem. However, the acoustic center of the microphone 41 is aligned with the rotation axis A of the turntable 51 as shown in FIG. Alternatively, as shown in FIG. 16, the microphone 41 may be rotationally driven while keeping the acoustic center of the microphone 41 at a constant distance from the rotation axis A of the turntable 51.

  As will be described later, the arithmetic processing unit 32 calculates a first-order Fourier coefficient of the squared sound pressure waveform received by the sound receiving unit 31 to calculate a phase angle indicating the sound arrival direction. In addition, the spectrum of the direction information can be analyzed, and the evaluation index of the microphone directivity can be calculated. For example, the degree of omnidirectionality can be analyzed by the ratio of the zeroth-order component and the first-order or higher-order component.

  The output unit 33 outputs the phase angle, the amplitude, the squared sound pressure, the microphone directivity evaluation index, and the like calculated by the calculation processing unit 32 to a display monitor, for example.

  According to such an acoustic measurement device 30, it is possible to measure acoustic intensity with high accuracy and ease.

  Next, specific arithmetic processing in the arithmetic processing unit 32 will be described in detail. Here, the sound source direction is estimated by obtaining the first-order Fourier coefficient of the squared sound pressure waveform obtained by the rotation.

  Intensity is energy that passes through a unit plane perpendicular to the traveling direction of sound waves per unit time. As shown in FIG. 17, when sound waves are incident on the unit surface at an angle θ, the acoustic energy S passing through the unit surface is the inner product of the unit normal vector n and the sound intensity I of the surface.

  This indicates that when the unit surface is rotated, the acoustic energy passing through the unit surface vibrates with a period of 1. Therefore, the intensity is nothing but the primary component (direction frequency: 1 Hz) of the direction information of the sound field. In other words, the sound field is measured by rotating an arbitrary directional microphone (however, it is necessary to include a direction frequency of 1 Hz, so a bidirectional microphone cannot be used), and the Fourier series of the obtained direction information (square sound pressure waveform) is measured. By examining the first-order Fourier coefficients of the expansion, the sound intensity can be known. Since the method using the rotating microphone is also based on the above equation (12), it is necessary to measure the square sound pressure while rotating the microphone.

That is, the arithmetic processing unit 32 squares the sound pressure obtained from the directional microphone, the arithmetic means for expanding the square sound pressure waveform by Fourier series expansion and calculating the first order Fourier coefficient, and the first order Fourier coefficient. Sound intensity calculating means for calculating the sound intensity including the direction of arrival of the sound based on the sound intensity. Here, the primary Fourier coefficient is a set of the coefficient a _{1 in the} expression (3) and the coefficient b _{1 in} the expression (4). The a ₁ and b ₁ (24) angle of arrival Substituting can be calculated in the expression (26) substituting can square sound pressure calculated if the equation, the magnitude of sound intensity by substituting in (27) Can be calculated.

Further, instead of the Fourier series expansion of the squared sound pressure waveform, a fast Fourier transform (FFT) can be used. When FFT is used, the real part of the first order Fourier coefficient corresponds to the coefficient a ₁ in the expression (3), and the imaginary part of the first order Fourier coefficient corresponds to the coefficient b ₁ in the expression (4).

  The sound intensity can be easily analyzed by rotating one microphone in this way and expanding the obtained squared sound pressure waveform in a Fourier series.

  Next, the measurement method using the rotating microphone according to the second embodiment was verified by computer simulation. White noise was presented from the 240 degree direction, and further, a condition was set so that uncorrelated background noise came from all directions and SN was -3 dB. And this was picked up by rotating one hyper cardioid microphone.

  18 and 19 show the time series data collected and the square waveform thereof, respectively. The phase angle of the first-order Fourier coefficient of the square waveform was 247 degrees, which almost coincided with the sound source direction. Since background noise is separated as a direct current component, sound having directionality can be extracted even under such a poor SN condition.

  As described above, according to the acoustic measurement device according to the second embodiment, the sound intensity is calculated based on the Fourier series expansion using one microphone, so that the direction information of 1 Hz is obtained regardless of the directivity characteristics of the microphone. Any microphone that includes can be used. In addition, since one microphone is rotated, inconsistencies such as sensitivity differences and phase differences between a plurality of microphones do not occur. Furthermore, if an acute directional microphone is used, information on the first and higher frequencies can be obtained.

(Third embodiment)
According to the sound intensity measurement method (CC method) using a unidirectional microphone pair of 180 degrees opposite to each other as described above, a method of combining two omnidirectional sound pressure type microphones (PP method), which has been the mainstream in the past, is used. In addition, it is not necessary to strictly manage microphone sensitivity difference, phase difference, microphone interval, and the like. This is because the former uses directivity information, but uses the difference in microphone position as information. Here, the cardioid microphone pair is used as the unidirectional microphone pair, and the effectiveness of the unidirectional microphone pair will be described in more detail.

[Intensity measurement principle]
First, as shown in FIG. 20, a sound field in which a single plane wave P (t) arrives at an angle θ with respect to the r direction is assumed. At that time, the particle velocity u (t) in the sound wave traveling direction is expressed by equation (28).

Then, the particle velocity u _r (t) in the r direction is expressed by equation (29).

  Therefore, the r-direction component of the intensity is expressed by the following equation.

Next, this sound field is measured with a cardioid microphone pair. The responses P ₁ (t) and P ₂ (t) measured by the pair of cardioid microphones (Mic.1, Mic.2) are as follows.

  When both are added, an omnidirectional response is obtained as in the following equation.

  Also, the difference between the two is given by equation (34).

  Here, comparing the equation (34) with the equation (29), the particle velocity u (t) can be obtained from the difference between the responses of the pair of cardioid microphones (Mic.1, Mic.2) as in the following equation. I understand.

  Therefore, the instantaneous intensity is as follows.

  Equation (36) can also be written as follows:

  That is, it becomes the same as the above equation (12).

  Next, it is confirmed whether or not the interference sound field as shown in FIG. The sound pressure P (t) and particle velocity u (t) at the observation point are as follows.

  Therefore, the instantaneous intensity is as follows.

Next, the intensity of this sound field is measured with a cardioid microphone pair (Mic.1, Mic.2). Responses P ₁ (t) and P ₂ (t) measured by each microphone are as follows.

Here, the addition and difference of P ₁ (t) and P ₂ (t) are as follows.

  Substituting these equations (43) and (44) into equations (35) and (36) to obtain the particle velocity and instantaneous intensity agree with the theoretical equations of equations (39) and (40). Further, even if the formulas (41) and (42) are substituted into the formula (37), they agree with the theoretical formula of the formula (40). That is, it can be seen that the C-C method is established even in an interference sound field.

  Further, the average intensity of the C-C method can be obtained by averaging the instantaneous intensity in the same manner as the P-P method.

  As described above, according to the C-C method, various averaging methods can be considered, so that a higher degree of freedom than that of the P-P method can be obtained.

[Microphone spacing and phase mismatch]
Next, the influence of the mismatch between the microphone intervals and the phase characteristics will be described. First, let us consider a case where the sound field of FIG. 1 is analyzed based on equations (45) and (46) using cardioid microphone pairs (Mic.1, Mic.2) whose phase characteristics do not match.

{P1 (t) -P ₂ (t)} and {P ₁ (t) + P ₂ (t)} due to the r-direction component of the incoming sound affect the amplitude due to the phase difference between Mic.1 and Mic.2. The relative phase relationship is maintained because the phase is always shifted by the same amount. On the other hand, {P ₁ (t) -P ₂ (t)} and {P ₁ (t) + P ₂ (t)}, which are components orthogonal to the r direction, show which phase characteristics of Mic.1 and Mic.2 Even if they are different, they are always orthogonal. Therefore, the average value of the components orthogonal to the r direction in {P ₁ (t) −P ₂ (t)} × {P ₁ (t) + P ₂ (t)} is always zero. Therefore, when calculating the expressions (45) and (46), the influence of the microphone phase mismatch is automatically canceled.

Next, consider the case of analysis based on equation (47). This equation shows that the difference can be calculated after square averaging P ₁ (t) and P ₂ (t) to obtain the average intensity. The mean square value is independent of the phase characteristics of the microphone. In addition, the equation (47) also means that if the measurement sound field is stationary, the intensity can be calculated by post-processing by sequentially changing the direction of one microphone and performing post-processing. In this case, it is also possible to measure P ₁ (t) and P ₂ (t) completely at the same position.

  As described above, in the CC method, when calculating the average intensity, the action of automatically canceling the mismatch of the phase characteristics of the microphones works. Therefore, when calculating the average intensity, the influence of the mismatch of the phase characteristics is affected. I do n’t accept anything. Therefore, calibration of the phase characteristics of the microphone is not necessary. Note that when obtaining the instantaneous intensity, it is necessary for the C-C method to match the phase characteristics of the two microphones as in the P-P method.

  In addition, the C-C method is not affected by the wavelength of sound in principle and has no frequency dependence. Therefore, wideband direction information can be obtained without the need to change the interval between the microphone probes according to the frequency band of the measurement object as in the P-P method. However, since a plurality of microphones cannot be installed at the same position in practice, the microphone interval is an error factor.

[Simulation result]
As described in the first embodiment, if a CC-type microphone pair is installed on an orthogonal coordinate axis, an independent component in each coordinate axis direction is obtained. When using a two-dimensional 4ch probe and a three-dimensional 6ch probe, when adding the responses of a plurality of cardioid microphones to obtain an omnidirectional response P (t), two microphones on the same axis as in equation (33) It is also possible to add the responses of all microphones instead of combining In this case, for example, the intensity in the case of a two-dimensional 4ch probe is as follows.

  Although the expressions (45) and (48) are essentially the same, there is a possibility that the influence of the phase mismatch of the microphones and the influence of the microphone interval is different. Therefore, the effect of the microphone interval and phase mismatch due to the above-described algorithm was examined by simulation. Here, a simple two-dimensional sound field in which a single plane wave (100 Hz sine wave) arrives was assumed. As shown in FIG. 22, the sound wave arrival angle was changed in 10 degree steps from 0 degrees to 350 degrees, and the sound arrival direction and intensity were analyzed with a 4ch probe in which CC microphone pairs were orthogonalized. .

  First, the waveform received by each cardioid microphone was simulated and analyzed based on the waveform for 4 channels. At that time, the influence was examined by changing the phase of the microphone and the microphone interval Δd. The calculated average intensity calculation algorithms are (45), (46), (47), and (48).

  23 and 24 are graphs showing the arrival direction and the absolute value of the intensity when the phase of 1ch is shifted by + π / 2, respectively. 25 and 26 are graphs showing the arrival direction and the absolute value of the intensity when the microphone interval is set to the half wavelength λ / 2 of the analysis target sound, respectively. In addition, the absolute value of the intensity in the figure is a value before the amplitude of the sound pressure is normalized by 1 and divided by ρc.

  Looking at these results, it was confirmed that the simulation results by the equations (45), (46), and (47) are not affected by the phase mismatch and the microphone interval in both the arrival direction and the intensity absolute value. However, the analysis based on the equation (48) is affected by them, and it can be seen that there are errors in both the arrival direction and the intensity absolute value. In other words, it was found that when determining the omnidirectional response with a multichannel probe, it is better to calculate using only the same microphone pair as the particle pair whose particle velocity was obtained.

  That is, if the output difference of the unidirectional microphone positioned 180 degrees opposite on the predetermined axis is used, the direction component of the predetermined axis can be measured with high accuracy.

  Next, a unidirectional microphone pair installed on the r-direction axis and an r-direction component intensity measuring apparatus will be described. Here, the intensity refers to the instantaneous intensity and the average intensity.

FIG. 27 is a block diagram showing a configuration of the first intensity measurement device. This first intensity measuring device is configured to obtain a predetermined value from a pair of microphones 61 in which a cardioid microphone 1 and a cardioid microphone 2 are arranged 180 degrees opposite to each other on an r-direction axis, and an output p ₁ (t) of the cardioid microphone 1. A band-pass filter 62 that extracts a signal in the frequency band, a band-pass filter 63 that extracts a signal in a predetermined frequency band from the output p ₂ (t) of the cardioid microphone 2, and a signal of a predetermined frequency from the band-pass filters 62 and 63 are added. , An adder 64 that obtains a difference between signals of a predetermined frequency from the bandpass filters 62 and 63, a multiplier 66 that multiplies the outputs of the adder 64 and the differencer 65, and an output of the multiplier 66 And an averaging circuit 67.

According to the first intensity measuring apparatus, the average intensity in the r direction based on the sound pressure difference can be obtained as shown in Equation (45). Here, the omnidirectional response p (t), which is a scalar quantity, can be obtained from the output from the adder 64. Further, the particle velocity u _r (t) in the r direction can be obtained from the output from the subtractor 65. Further, the instantaneous intensity I _r (t) in the r direction can be obtained from the output from the multiplier 66.

FIG. 28 is a block diagram showing the configuration of the second intensity measuring device. This second intensity measuring device is configured to obtain a predetermined value based on a microphone pair 61 in which the cardioid microphone 1 and the cardioid microphone 2 are disposed on the r-direction axis in opposite directions by 180 degrees, and the output p ₁ (t) of the cardioid microphone 1. A band-pass filter 62 that extracts a signal in the frequency band, a band-pass filter 63 that extracts a signal in a predetermined frequency band from the output p ₂ (t) of the cardioid microphone 2, and a square that squares a signal in a predetermined frequency from the band-pass filter 62 A circuit 71, a square circuit 72 that squares a signal of a predetermined frequency from the band-pass filter 63, and a difference for obtaining a difference between the square responses P ² ₁ (t) and P ² ₂ (t) from the square circuits 71 and 72 And an averaging circuit 74 for averaging the outputs from the differentiator 73.

According to the second intensity measuring device, the average intensity in the r direction based on the squared sound pressure difference can be obtained as shown in Equation (46). Here, the instantaneous intensity I _r (t) in the r direction can be obtained from the output from the differentiator 73.

FIG. 29 is a block diagram showing the configuration of the third intensity measuring device. The third intensity measuring device is configured to obtain a predetermined value from a microphone pair 61 in which a cardioid microphone 1 and a cardioid microphone 2 are arranged on the r-direction axis in opposite directions by 180 degrees, and an output p ₁ (t) of the cardioid microphone 1. A band-pass filter 62 that extracts a signal in the frequency band, a band-pass filter 63 that extracts a signal in a predetermined frequency band from the output p ₂ (t) of the cardioid microphone 2, and a square that squares a signal in a predetermined frequency from the band-pass filter 62 A circuit 71, a square circuit 72 that squares a signal of a predetermined frequency from the bandpass filter 63, an averaging circuit 81 that averages the square response P ² ₁ (t) from the square circuit 71, and a square from the square circuit 72 An averaging circuit 82 that averages the response P ² ₂ (t), and a differentiator 83 that obtains a difference between the outputs of the averaging circuits 81 and 82. And.

  According to the third intensity measuring apparatus, the average intensity in the r direction based on the mean square sound pressure difference can be obtained as shown in the equation (47).

  As described above, the direction information of the predetermined axis can be obtained with high accuracy by using the output difference of the unidirectional microphone positioned 180 degrees opposite to the predetermined axis.

  For example, when a unidirectional microphone pair is used, the intensity can be calculated based on the output difference of the unidirectional microphone pair. In this case, the average intensity may be calculated from the sound pressure difference as shown in equation (45), or may be calculated as the squared sound pressure difference as shown in equation (46). In particular, if the sound pressure difference is used, particle velocity and omnidirectional response can be obtained.

  Further, when the unidirectional microphone is rotated, the average intensity can be calculated based on the output difference of the microphone located on the predetermined axis. In this case, the average intensity may be calculated from the mean square sound pressure difference as shown in equation (47). When this unidirectional microphone is rotated, measurement can be performed at the same position, that is, with a microphone interval of zero.

  In addition, if the direction component of each axis of the orthogonal coordinates is calculated, the sound direction information including the sound arrival direction can be calculated. For example, if two sets of unidirectional microphone pairs are orthogonalized and the direction component of each axis is calculated based on the output difference of each unidirectional microphone pair, the direction information of the two-dimensional sound can be calculated. it can. In addition, if the unidirectional microphone is rotated, the direction component of each axis is calculated based on the output difference between the 0 ° position and the 180 ° position on the orthogonal coordinate axis and the output difference between the 90 ° position and the 270 ° position. Two-dimensional sound direction information can be calculated.

  Although the best mode for carrying out the present invention has been described above, the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the gist of the present invention. Of course.

It is a conceptual diagram for demonstrating sound direction information. It is a block diagram which shows the structure of the acoustic measuring device which concerns on 1st Embodiment. It is a schematic diagram which shows an example of a structure of the sound receiving part which concerns on 1st Embodiment. It is a schematic diagram for demonstrating the sensitivity difference of the microphone pair toward 0 degree | times and 180 degree | times. It is a schematic diagram for demonstrating the sensitivity difference of the microphone pair toward 90 degree | times and 270 degree | times. It is a graph which shows the sensitivity difference of the square sound pressure in the cardioid microphone pair toward 0 degree | times and 180 degree | times. It is a schematic diagram which shows the estimation experiment of a sound source direction. It is a graph which shows the estimation result in frequency 125Hz. It is a graph which shows the estimation result in frequency 1kHz. It is a graph which shows the estimation result in frequency 8kHz. It is a graph which shows the maximum value and average value of the estimation error with respect to a frequency. It is a block diagram which shows the structure of the acoustic measuring device which concerns on 2nd Embodiment. It is a schematic diagram which shows an example of a structure of the sound receiving part which concerns on 2nd Embodiment. It is a figure which shows the structural example which arrange | positions a microphone on a turntable. It is a figure which shows an example of the position of the acoustic center in a turntable. It is a figure which shows the other example of the position of the acoustic center in a turntable. It is a conceptual diagram for demonstrating intensity. It is a graph which shows the time series data in the acoustic measuring device which concerns on 2nd Embodiment. It is a graph which shows the square waveform in the acoustic measuring device which concerns on 2nd Embodiment. It is a figure which shows the example of a single plane wave sound field. It is a figure which shows the example of an interference sound field. It is a figure for demonstrating the 4ch probe arrange | positioned in a two-dimensional sound field. It is a graph which shows the arrival direction when the phase of 1ch is shifted + π / 2. It is a graph which shows the absolute value of intensity at the time of shifting + pi / 2 phase of 1ch. It is a graph which shows the arrival direction at the time of making microphone interval into the half wavelength (lambda) / 2 of the sound for analysis. It is a graph which shows the absolute value of an intensity | strength when a microphone space | interval is made into the half wavelength (lambda) / 2 of an analysis object sound. It is a block diagram which shows the structure of a 1st intensity measurement apparatus. It is a block diagram which shows the structure of a 2nd intensity measurement apparatus. It is a block diagram which shows the structure of a 3rd intensity measurement apparatus.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 Acoustic measuring device, 11 Sound receiving part, 12 Computation processing part, 13 Output part, 30 Acoustic measuring device, 31 Sound receiving part, 32 Computation processing part, 33 Output part, 41 Microphone, 42 Horizontal direction rotation drive part, 43 Vertical Direction rotation drive unit, 44 control unit, 51 turntable, 52 drive unit, 61 microphone pair, 62 band pass filter, 63 band pass filter, 64 adder, 65 differentiator, 66 multiplier, 67 averaging circuit, 71 square Circuit, 72 square circuit, 73 differentiator, 74 averaging circuit, 81 averaging circuit, 82 averaging circuit, 83 differentiator

Claims (16)

An output difference calculating means for calculating an output difference of a unidirectional microphone located 180 degrees opposite on a predetermined axis;
An acoustic measurement apparatus comprising: an axial direction component calculation unit that calculates a direction component of the predetermined axis based on the output difference. Sound receiving means for receiving sound using a unidirectional microphone pair in which directivity is arranged 180 degrees opposite to the predetermined axis;
2. The acoustic measurement apparatus according to claim 1, wherein the axial direction component calculation means calculates the direction component of the predetermined axis based on an output difference of the unidirectional microphone pair.   The axial direction component calculation means calculates particle velocity, omnidirectional response, and intensity based on the sound pressure difference and the sound pressure addition of the unidirectional microphone pair. Acoustic measuring device.   The acoustic measurement apparatus according to claim 2, wherein the axial direction component calculation means calculates an intensity based on a squared sound pressure difference of the unidirectional microphone pair.   The acoustic measurement apparatus according to claim 2, wherein the axial direction component calculation means calculates an average intensity based on an average square sound pressure difference of the unidirectional microphone pair. A sound receiving means for receiving sound by rotating a unidirectional microphone is provided.
2. The acoustic measurement apparatus according to claim 1, wherein the axial direction component calculating means calculates the direction component of the predetermined axis based on an output difference of a unidirectional microphone located on the predetermined axis.   The acoustic measurement apparatus according to claim 6, wherein the axial direction component calculation means calculates an average intensity based on a mean square sound pressure difference of the unidirectional microphone located on the predetermined axis. Sound direction information calculation means for calculating sound direction information including the direction of arrival of sound based on the direction component of each axis of orthogonal coordinates,
2. The acoustic measurement apparatus according to claim 1, wherein the axial direction component calculation means calculates a direction component of each axis of the orthogonal coordinates.   The acoustic measurement apparatus according to claim 8, wherein the orthogonal coordinates are two-dimensional or three-dimensional. An output difference calculating step of calculating an output difference of a unidirectional microphone located 180 degrees opposite on a predetermined axis;
And an axial direction component calculating step of calculating a direction component of the predetermined axis based on the output difference. Sound receiving means for receiving sound using a unidirectional microphone pair in which directivity is arranged 180 degrees opposite to each other on the axes of orthogonal coordinates;
An output difference calculating means for calculating an output difference obtained from the unidirectional microphone pair;
Direction information orthogonal basis component calculating means for calculating an orthogonal basis component of direction information of each axis based on the output difference;
An acoustic measurement device comprising: sound direction information calculating means for calculating sound direction information including a direction of sound arrival based on the direction information orthogonal basis component of each axis.   12. The direction information orthogonal basis component calculation means calculates a first-order Fourier coefficient of Fourier series expansion using an angle as a parameter using a sensitivity difference relationship of the unidirectional microphone pair. Acoustic measurement device. The output difference calculation means calculates an output difference of the square sound pressure,
The direction information orthogonal basis component calculation means calculates an orthogonal basis component of the direction information of each axis based on the output difference of the square sound pressure,
12. The acoustic measurement apparatus according to claim 11, wherein the sound direction information calculation means calculates an acoustic intensity including a sound arrival direction based on the direction information orthogonal basis component of each axis.   12. The acoustic measurement apparatus according to claim 11, wherein the directional characteristic of the unidirectional microphone is one selected from a cardioid, a super cardioid, a hyper cardioid, and an ultra cardioid. A sound receiving process for receiving sound using a unidirectional microphone pair in which directivity is arranged 180 degrees opposite to each other on the axes of orthogonal coordinates;
An output difference calculating step of calculating an output difference obtained from the unidirectional microphone pair;
A direction information orthogonal basis component calculating step for calculating an orthogonal basis component of the direction information of each axis based on the output difference;
A sound direction information calculating step of calculating sound direction information including a sound arrival direction based on the direction information orthogonal basis component of each axis. Sound receiving means for receiving sound by rotating a directional microphone;
A square sound pressure calculating means for calculating the square sound pressure of the directional microphone;
An arithmetic means for expanding the squared sound pressure by Fourier series and calculating a first-order Fourier coefficient;
And an acoustic intensity calculating means for calculating an acoustic intensity including a direction of arrival of the sound based on the first order Fourier coefficient.

Images