A sound field separation system and method based on spherical measurement surface

Abstract

The present invention specifically relates to a sound field separation system and method based on a spherical measuring surface, the sound field separation system includes a target sound source and an interfering sound source, the mixed sound field generated by the target sound source and the interfering sound source is used as the sound field to be separated; a spherical measuring surface is arranged between the target sound source and the interfering sound source, and measurement points are arranged on the spherical measuring surface; the spherical measuring surface is composed of a spherical array, and a hemispherical equivalent source surface is arranged on each side of the spherical array, and equivalent source points are arranged on the equivalent source surface; the sound pressure signal measured by the spherical measuring surface and the propagation function of the equivalent source surface are brought into the sound field information separation model to obtain the equivalent source strength, thereby obtaining the sound field generated by the sound sources on both sides of the spherical measuring surface, and realizing sound field separation. The sound field separation system and method of the present invention adopts a single-layer spherical measuring surface to measure the sound pressure, and uses the measured sound pressure information to achieve sound field separation under the condition of fewer measurement points, with higher separation accuracy.

Description

Technical Field

[0001] The invention belongs to the technical field of near-field acoustic holography and relates to an acoustic field separation system and method based on a spherical measuring surface. Background Art

[0002] In many acoustic scenes, due to the presence of multiple sound sources or partial scatterers, the sound field at a specific location in space is usually formed by the superposition of the outward radiated sound of the sound source or the superposition of the reflected sound. Based on the outward radiated waves of the sound source, directly reconstructing the sound field using traditional near-field acoustic holography technology will produce large reconstruction errors or even wrong results. In order to accurately reconstruct the sound field, sound field separation is required.

[0003] Among the existing sound field separation methods, E. Fernandez-Granded et al. realized sound field separation based on a single-layer vector microphone array. The vector microphone used in this method is relatively expensive. F. Ma et al. used a double-layer spherical array to measure sound pressure to achieve sound field separation. This method requires the arrangement of two layers of spherical sound pressure sensor arrays, which has a complex structure. At the same time, the measurement arrays used in the above sound field reconstruction methods are all rigid spherical arrays, which separate the incident sound field and the scattered sound field, and fail to separate multiple sound sources to be separated in space.

[0004] The existing sound field separation method based on sparse measurement can reduce the measurement points, but it needs to set up two measurement planes; using a single measurement plane, the sound field separation can be performed, but the premise is that the approximate location of the sound source needs to be clear or pre-measurement prediction is required. Therefore, the operation is relatively cumbersome. Summary of the invention

[0005] The purpose of the present invention is to provide a sound field separation system and method based on a spherical measuring surface. The technical solution adopted by the present invention is:

[0006] A sound field separation system based on a spherical measuring surface, comprising: a target sound source and an interfering sound source, wherein a mixed sound field generated by the target sound source and the interfering sound source is used as a sound field to be separated, a spherical measuring surface is arranged between the target sound source and the interfering sound source, and measurement points are arranged on the spherical measuring surface;

[0007] The spherical measurement surface is composed of a spherical array, and a hemispherical equivalent source surface is arranged on each side of the spherical array, and equivalent source points are arranged on the equivalent source surface;

[0008] The sound pressure signal measured by the spherical measurement surface and the propagation function of the equivalent source surface are brought into the sound field information separation model to obtain the equivalent source intensity, thereby obtaining the sound field generated by the sound sources on both sides of the spherical measurement surface to achieve sound field separation.

[0009] Furthermore, the radii of the equivalent source surfaces are the same.

[0010] A sound field separation method based on a spherical measurement surface comprises the following steps:

[0011] Step 1: The mixed sound field generated by the target sound source and the interference sound source is used as the sound field to be separated. A spherical measurement surface is set between the target sound source and the interference sound source. The center of the sphere is used as the coordinate origin. The spherical measurement surface is randomly divided into two parts. The sound pressure signals measured in the two parts are randomly divided into two groups, which are recorded as P1 and P2 respectively.

[0012] Step 2: Obtain the frequency characteristics of the sound source according to the sound pressure data measured by the spherical measurement surface, and determine the corresponding sound wave number k;

[0013] Step 3: determining a truncation order N according to the number of measurement points on the spherical measurement surface, the radius of the spherical measurement surface and the acoustic wave number;

[0014] Step 4: a hemispherical equivalent source surface is arranged on each side of the spherical measurement surface, and equivalent source points are arranged on the equivalent source surface, and the equivalent source strengths are respectively recorded as Q1 and Q2;

[0015] Step 5: The transfer matrix between the measurement point of the spherical measurement surface and the equivalent source point is a free field Green's function;

[0016] Step 6: Establish a sound field separation model to obtain the sound pressure at the measurement point;

[0017] Step 7: Solve the equivalent source strength;

[0018] Step 8: Obtain the radiated sound field of the target sound source on the measurement surface to complete the sound field separation.

[0019] Furthermore, the step 1 further comprises: randomly dividing the spherical measurement surface into two parts, namely a first spherical measurement surface H1 and a second spherical measurement surface H2, wherein the sound pressure signals measured by the first spherical measurement surface H1 and the second spherical measurement surface H2 are respectively recorded as P1 and P2; wherein P1=P 11 +P 21 , P2=P 12 +P 22 , P 11 and P 21 represents the sound pressure generated by the target sound source and the interference sound source on the first spherical measurement surface H1, P 12 and P 22 It represents the sound pressure generated by the target sound source and the interference sound source on the second spherical measurement surface H2.

[0020] Furthermore, the truncation order N of step 3 is:

[0021]

[0022] in, represents rounding up, k is the sound wave number, r h is the radius of the spherical measuring surface.

[0023] Furthermore, the step 4 further comprises: arranging a spherical measuring surface with a radius of r on each side. s The hemispherical equivalent source surfaces are S1 and S2 respectively.

[0024] Furthermore, the free field Green's function of step 5 is:

[0025]

[0026] Among them, Ω h is the azimuth of the measuring point, Ω s is the azimuth of the equivalent source point, j is an imaginary number, k is the wave number, s is the number of equivalent sources, N is the cutoff order, j n (·) is the nth-order Bessel function of the first kind, h n (·) is the nth-order first-kind Hankel function, is a spherical harmonic function of order n and degree m, and the superscript '*' is the complex conjugate.

[0027] Furthermore, the step 6 also includes:

[0028] Establish an acoustic field separation model, and transform the sound pressure P generated by the target sound source on the first spherical measurement surface H1 into 11 Expressed as: P 11 =G 11 Q1, where G 11 is the free-field Green's function generated by the target sound source on the first spherical measurement surface H1, and the sound pressure P generated by the interference sound source on the first spherical measurement surface H1 is 21 Expressed as: P 21 =G 21 Q2, where G 21 is the free-field Green's function generated by the interfering sound source on the first spherical measurement surface H1, and the sound pressure P generated by the target sound source on the second spherical measurement surface H2 is 21 Expressed as: P 12 =G 12 Q2, where G 12 is the free-field Green's function generated by the target sound source on the second spherical measurement surface H2, and the sound pressure P generated by the interference sound source on the second spherical measurement surface H2 is 22 Represented as P 22 =G 22 Q2 Among them, G 22is the free-field Green's function generated by the interfering sound source on the second spherical measuring surface H2;

[0029] The sound pressure at the measurement point is expressed as:

[0030]

[0031] Furthermore, the step 7 further comprises: solving the equivalent source strength, and writing the expression of the sound pressure vector in step 6 as: P H =GQ, solve the sparse solution of Q with equivalent source strength min||Q||1subject to||P by sparse regularization method H -GQ||2≤ε, where ε is the noise floor.

[0032] Furthermore, the step 8 further comprises: according to the function matrix The radiated sound field of the target sound source on the measurement surface is obtained to complete the sound field separation.

[0033] Beneficial effects of the present invention:

[0034] The sound field separation system and method of the present invention are both based on spherical harmonic function expansion and equivalent source method, and adopt a single-layer spherical measurement surface to measure the sound pressure. By using the measured sound pressure information, the sound field separation can be completed under the condition of fewer measurement points, and the radiation information of the sound source can be accurately obtained. It has high separation accuracy and is simple and convenient to operate. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A distribution diagram of measurement points on a spherical measurement surface according to an embodiment of the present invention;

[0036] Figure 2 A spherical measurement surface, an equivalent source surface, and a distribution diagram of sound source positions in an embodiment of the present invention;

[0037] Figure 3 This is a flow chart of a method for separating an acoustic field based on a spherical measuring surface according to the present invention;

[0038] Figure 4 It is a comparison diagram of the real parts of the sound pressure of the unseparated sound field at 1000 Hz, the theoretical value of the sound pressure of the sound field of the target sound source (sound source 1) on the measurement surface, and the sound pressure value of the sound field of the target sound source (sound source 1) on the measurement surface after the interference sound source (sound source 2) is separated by the method of the present invention at the corresponding measurement points;

[0039] Figure 5 It is a comparison diagram of the imaginary parts of the sound pressure of the unseparated sound field at 1000 Hz, the theoretical value of the sound pressure of the sound field of the target sound source (sound source 1) on the measurement surface, and the sound pressure value of the sound field of the target sound source (sound source 1) on the measurement surface after the interference sound source (sound source 2) is separated by the method of the present invention at the corresponding measuring points;

[0040] Figure 6 The figure is a theoretical sound pressure level distribution diagram of the sound field of the target sound source (sound source 1) at the measurement surface at 1000 Hz according to an embodiment of the present invention;

[0041] Figure 7 This is a sound pressure level diagram of the sound field of the target sound source (sound source 1) at the measurement surface after the method of the present invention separates the interference sound source (sound source 2) at 1000 Hz in an embodiment of the present invention;

[0042] Figure 8 This is a reconstruction error diagram after sound field separation at frequencies from 100 Hz to 2000 Hz according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] Embodiment 1:

[0045] The present embodiment provides a sound field separation system based on a spherical measurement surface, and the sound field separation system includes: a target sound source and an interfering sound source, and the mixed sound field generated by the target sound source and the interfering sound source is used as the sound field to be separated; a spherical measurement surface is arranged between the target sound source and the interfering sound source, and measurement points are arranged on the spherical measurement surface; the spherical measurement surface is composed of a spherical array, and a hemispherical equivalent source surface is arranged on each side of the spherical array, and equivalent source points are arranged on the equivalent source surface; combining the equivalent source principle and the wave superposition principle, the sound pressure signal measured by the spherical measurement surface and the propagation function of the equivalent source surface are brought into the sound field information separation model, and the equivalent source strength is obtained by the sparse regularization method, thereby obtaining the sound field generated by the sound sources on both sides of the spherical measurement surface, and realizing sound field separation. In order to achieve a better separation effect, the radius of the equivalent source surface of the present embodiment is the same.

[0046] The sound field separation system of the present invention can separate the mixed sound field generated by the target sound source and the interference sound source in a certain space by only arranging a few measurement points on a single-layer spherical measurement surface, thereby obtaining relevant information of the target sound field, and has high separation accuracy.

[0047] Embodiment 2:

[0048] This embodiment provides a sound field separation method based on a spherical measurement surface, and the sound field separation method includes the following steps:

[0049] Step 1: The mixed sound field generated by the target sound source and the interference sound source is used as the sound field to be separated. A spherical measurement surface is set between the target sound source and the interference sound source. The center of the sphere is used as the coordinate origin. The spherical measurement surface is randomly divided into two parts, namely the first spherical measurement surface H1 and the second spherical measurement surface H2. The spherical measurement surface is randomly divided into two parts. The sound pressure signals measured by the first spherical measurement surface H1 and the second spherical measurement surface H2 are recorded as P1 and P2 respectively; where P1 = P 11 +P 21 , P2=P 12 +P 22 , P 11 and P 21 represents the sound pressure generated by the target sound source and the interference sound source on the first spherical measurement surface H1, P 12 and P 22 Indicates the sound pressure generated by the target sound source and the interference sound source on the second spherical measurement surface H2;

[0050] Step 2: Obtain the frequency characteristics of the sound source according to the sound pressure data measured by the spherical measurement surface, and determine the corresponding sound wave number k;

[0051] Step 3: Determine the truncation order N according to the number of measurement points on the spherical measurement surface, the radius of the spherical measurement surface, and the number of acoustic waves. The truncation order N is: in, represents rounding up, k is the sound wave number, r h is the radius of the spherical measuring surface;

[0052] Step 4: Arrange a hemispherical equivalent source surface on both sides of the spherical measurement surface. The radius of the hemispherical equivalent source surface is r s , are S1 and S2 respectively, equivalent source points are arranged on the equivalent source surface, and the equivalent source strengths are recorded as Q1 and Q2 respectively;

[0053] Step 5: The transfer matrix between the measurement point and the equivalent source point on the spherical measurement surface is the free-field Green's function, which is:

[0054]

[0055] Among them, Ω h is the azimuth of the measuring point, Ω s is the azimuth of the equivalent source point, j is an imaginary number, k is the wave number, s is the number of equivalent sources, N is the cutoff order, j n (•) is the nth-order Bessel function of the first kind, h n (•) is the nth-order first-kind Hankel function, is a spherical harmonic function of order n and degree m, and the superscript '*' is the complex conjugate;

[0056] Step 6: Establish an acoustic field separation model and transform the sound pressure P generated by the target sound source on the first spherical measurement surface H1 11 Expressed as: P 11 =G 11 Q1, where G 11 is the free field Green's function generated by the target sound source on the first spherical measurement surface H1, and the sound pressure P generated by the interference sound source on the first spherical measurement surface H1 is 21 Expressed as: P 21 =G 21 Q2, where G 21 is the free field Green's function generated by the interference sound source on the first spherical measurement surface H1, and the sound pressure P generated by the target sound source on the second spherical measurement surface H2 21 Expressed as: P 12 =G 12 Q2, where G 12 is the free field Green's function generated by the target sound source on the second spherical measurement surface H2, and the sound pressure P generated by the interference sound source on the second spherical measurement surface H2 is 22 Represented as P 22 =G 22 Q2 Among them, G 22 The free field Green's function generated by the interference sound source on the second spherical measurement surface H2 is used to obtain the sound pressure at the measurement point. The sound pressure at the measurement point can be expressed as:

[0057]

[0058] Step 7: Solve for the equivalent source strength and write the expression of the sound pressure vector in step 6 as: P H =GQ, solve the sparse solution of Q with equivalent source strength min||Q||1subject to||P by sparse regularization method H -GQ||2≤ε, where ε is the background noise;

[0059] Step 8: According to the function matrix Obtain the radiated sound field of the target sound source on the measurement surface and complete the sound field separation.

[0060] In the sound field separation method of this embodiment, the sound source of the sound field to be separated is of any shape and can be located at any position in the three-dimensional space. The sound field separation method of this embodiment is suitable for separating the sound field of a strong interference background environment.

[0061] The following is a simulation test to illustrate:

[0062] Two point sound sources are used, the sound source frequency is 1000 Hz, and the point sound source at (0.5, π / 2, π / 2) is the target sound source (sound source 1), and its volume velocity is 1×10 -4 m3 / s, the point sound source at (0.5, π / 2, π*4 / 3) is the interference sound source (sound source 2), and its volume velocity is 10×10 -4 m 3 / s. The radius of the measuring sphere is 0.309m. 64 measuring points are arranged on the measuring surface. The measuring points are randomly divided into two categories, which are recorded as measuring surfaces H1 and H2. Figure 1 As shown, Gaussian white noise with a signal-to-noise ratio of 30 dB is added to the 64 microphones on the measurement surface, and the sound pressures measured at the measurement surfaces H1 and H2 are recorded as P1 and P2.

[0063] According to 64 measuring points, the radius of the measuring sphere and the wave number k = 18.32, the truncation order is N = 7. The radius of the equivalent source surface is 0.45m. The positional relationship between the measuring surface, the equivalent source surface and the sound source is as follows: Figure 2 As shown. Through the sound pressure signal P H The sparse solution of the equivalent source Q is solved with the propagation function G. The sound field generated by the target sound source (sound source 1) at the measurement surface is solved according to the equivalent source strength Q to complete the sound field separation. The specific process is as follows: Figure 3 shown.

[0064] Figure 4 and Figure 5 They are the real and imaginary comparison diagrams of the sound pressure of the unseparated sound field at a frequency of 1000Hz, the theoretical value of the sound pressure of the target sound source (sound source 1) on the measurement surface, and the sound pressure value of the target sound source (sound source 1) on the measurement surface after the interference sound source (sound source 2) is separated by the method of the present invention at the corresponding measurement point. Figure 4 and Figure 5 It can be seen from the figure that when the target sound source (sound source 1) and the interference sound source (sound source 2) work together, the sound pressure on the measuring surface is greatly different from the theoretical value of the sound pressure of the target sound source (sound source 1) at the measuring surface due to the influence of the interference sound source (sound source 2). The interference sound source (sound source 2) cannot be separated, and the sound pressure of the sound field of the target sound source (sound source 1) cannot be directly measured. However, the sound pressure value after separating the interference sound source (sound source 2) using the method of the present invention is consistent with the real and imaginary parts of the theoretical value.

[0065] Figure 6 The theoretical sound pressure level distribution diagram of the target sound source (sound source 1) at the measurement surface at 1000Hz. Figure 7 The sound pressure level distribution diagram of the sound field after the interference sound source (sound source 2) is separated at the measurement surface by the method of the present invention at 1000Hz. Figure 6 and Figure 7 In the figure, '+' indicates the location of the target sound source (sound source 1), and '*' indicates the location of the interference sound source (sound source 2). Figure 6 and Figure 7 It can be seen that the method of the present invention can achieve sound field separation and accurately identify the specific location of the target sound source (sound source 1). Figure 8This is the sound field reconstruction error diagram at different frequencies after separating the interfering sound source (sound source 2).

[0066] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A sound field separation method based on a spherical measurement surface, characterized in that: The method is applied to a sound field separation system based on a spherical measuring surface, so that the mixed sound field generated by the target sound source and the interference sound source in a certain space can be separated by arranging only a few measuring points on a single-layer spherical measuring surface. The system comprises: a target sound source and an interfering sound source, wherein a mixed sound field generated by the target sound source and the interfering sound source is used as a sound field to be separated, a spherical measuring surface is arranged between the target sound source and the interfering sound source, and measuring points are arranged on the spherical measuring surface; the spherical measuring surface is composed of a spherical array, and a hemispherical equivalent source surface is arranged on each side of the spherical array, and equivalent source points are arranged on the equivalent source surface; the sound pressure signal measured by the spherical measuring surface and the propagation function of the equivalent source surface are brought into a sound field information separation model to obtain an equivalent source strength, thereby obtaining the sound field generated by the sound sources on both sides of the spherical measuring surface, and realizing sound field separation; the radius of the equivalent source surface is the same; The method comprises the following steps: Step 1: The mixed sound field generated by the target sound source and the interference sound source is used as the sound field to be separated. A spherical measurement surface is set between the target sound source and the interference sound source. The center of the sphere is used as the coordinate origin. The spherical measurement surface is randomly divided into two parts. The sound pressure signals measured in the two parts are randomly divided into two groups, which are respectively recorded as P1 and P2; the spherical measurement surface is randomly divided into two parts, namely the first spherical measurement surface H1 and the second spherical measurement surface H2. The sound pressure signals measured by the first spherical measurement surface H1 and the second spherical measurement surface H2 are respectively recorded as P1 and P2; wherein P1=P 11 +P 21 , P2=P 12 +P 22 , P 11 and P 21 represents the sound pressure generated by the target sound source and the interference sound source on the first spherical measurement surface H1, P 12 and P 22 represents the sound pressure generated by the target sound source and the interference sound source on the second spherical measurement surface H2; Step 2: Obtain the frequency characteristics of the sound source according to the sound pressure data measured by the spherical measurement surface, and determine the corresponding sound wave number k; Step 3: Determine the truncation order N according to the number of measurement points on the spherical measurement surface, the radius of the spherical measurement surface and the acoustic wave number; truncation order N: in, represents rounding up, k is the sound wave number, r h is the radius of the spherical measuring surface; Step 4: a hemispherical equivalent source surface is arranged on each side of the spherical measurement surface, and equivalent source points are arranged on the equivalent source surface, and the equivalent source strengths are respectively recorded as Q1 and Q2; Step 5: The transfer matrix between the measurement point of the spherical measurement surface and the equivalent source point is a free field Green's function; Step 6: Establish a sound field separation model to obtain the sound pressure at the measurement point; Step 7: Solve the equivalent source strength; Step 8: Obtain the radiated sound field of the target sound source on the measurement surface to complete the sound field separation.

2. The method for separating the acoustic field based on a spherical measuring surface according to claim 1, characterized in that: The step 4 further includes: arranging a spherical measuring surface with a radius of r on each side. s The hemispherical equivalent source surfaces are S1 and S2 respectively.

3. The method for separating the acoustic field based on a spherical measuring surface according to claim 2, characterized in that: The free field Green's function of step 5 is: Among them, Ω h is the azimuth of the measuring point, Ω s is the azimuth of the equivalent source point, j is an imaginary number, k is the wave number, s is the number of equivalent sources, N is the cutoff order, j n (·) is the nth-order Bessel function of the first kind, h n (·) is the nth-order first-kind Hankel function, is a spherical harmonic function of order n and degree m, and the superscript '*' is the complex conjugate.

4. The method for separating the acoustic field based on a spherical measuring surface according to claim 3, characterized in that: The step 6 also includes: Establish an acoustic field separation model, and transform the sound pressure P generated by the target sound source on the first spherical measurement surface H1 into 11 Expressed as: P 11 =G 11 Q1, where G 11 is the free-field Green's function generated by the target sound source on the first spherical measurement surface H1, and the sound pressure P generated by the interference sound source on the first spherical measurement surface H1 is 21 Expressed as: P 21 =G 21 Q2, where G 21 is the free-field Green's function generated by the interfering sound source on the first spherical measurement surface H1, and the sound pressure p generated by the target sound source on the second spherical measurement surface H2 is 12 Expressed as: P 12 =G 12 Q2, where G 12 is the free-field Green's function generated by the target sound source on the second spherical measurement surface H2, and the sound pressure P generated by the interference sound source on the second spherical measurement surface H2 is 22 Represented as P 22 =G 22 Q2 Among them, G 22 is the free-field Green's function generated by the interfering sound source on the second spherical measuring surface H2; The sound pressure at the measurement point is expressed as:

5. The method for separating the acoustic field based on a spherical measuring surface according to claim 4, characterized in that: The step 7 also includes: solving the equivalent source strength, and writing the expression of the sound pressure vector in step 6 as: P H =GQ, solve the sparse solution of Q with equivalent source strength min||Q||1subject to||P by sparse regularization method H -GQ||2≤ε, where ε is the noise floor.

6. The method for separating the acoustic field based on a spherical measuring surface according to claim 5, characterized in that: The step 8 also includes: according to the function matrix The radiated sound field of the target sound source on the measurement surface is obtained to complete the sound field separation.

Images