Active noise control system and active noise control method

Abstract

To provide an active noise control (ANC) system and method that achieve ANC with robustness to noise source location.SOLUTION: In an ANC system, a control unit 110 comprises: a reference microphone 130 that detects sound and generates a reference signal x; a specific microphone 150 that detects sound and generates a specific signal j; a position estimator 160 that generates position information (position discrimination function G) of a noise source based on the reference signal x and the specific signal j; a memory 170 that stores the position discrimination function and a noise control filter W1 in association with each other; a processor 180; and a first speaker 10A that radiates a first sound wave. The processor generates a control signal y1 that defines response characteristics of the first sound wave to the reference signal x based on the reference signal x and the noise control filter W1, and achieves active noise control by causing the first speaker 10A to emit the first sound wave based on the control signal y1.SELECTED DRAWING: Figure 6A

Description

[Technical Field]

[0001] The present invention relates to an active noise control system and an active noise control method. [Background technology]

[0002] Active noise control (ANC) is known. In ANC, noise is reduced by using an anti-phase sound. Examples of ANC are described in Patent Document 1 and Patent Document 2.

[0003] Furthermore, there is known a technique for reducing noise at a control point distant from the microphone. Examples of this technique are described in Patent Documents 4 and 5 and Non-Patent Documents 1 to 4. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2020-190599 [Patent Document 2] Japanese Patent Publication No. 2022-047766 [Patent Document 3] Japanese Patent Application Laid-Open No. 2016-122187 [Patent Document 4] Japanese Patent Application Laid-Open No. 2011-123330 [Patent Document 5] International Publication No. 2019 / 106748 [Non-patent literature]

[0005] [Non-Patent Document 1] Shota Toyooka, Yoshinobu Kajikawa, "Comparison of noise reduction performance and robustness of virtual sensing methods in ANC systems," 2022, Information Processing Society of Japan [Non-patent document 2] Shoma Edamoto and Yoshinobu Kajikawa, "Effectiveness Verification of Feedforward ANC System Implementation Using Virtual Sensing," 2018, Institute of Electronics, Information and Communication Engineers [Non-patent document 3] SJ Elliott and J. Cheer, “Modeling local active sound control with remote sensors in randomly random pressure fields,” J. Acoust. Soc. Am., vol. 137, no. 4, pp. 1936-1946, Apr. 2015. [Non-patent document 4] CD Petersen et al., “ACTIVE NOISE CONTROL AT A MOVING LOCATION USING VIRTUAL SENSING” ICSV13-July 2-6, 2006 Summary of the Invention [Problem to be solved by the invention]

[0006] The present invention provides a technique suitable for realizing ANC that is robust to the location of the noise source. [Means for solving the problem]

[0007] The present invention provides a reference microphone unit that detects sound and generates a reference signal, the reference microphone unit including one or more reference microphones; a specific microphone unit that detects sound and generates a specific signal, the specific microphone unit including one or more specific microphones; a first speaker that emits a first sound wave; a control device that determines a control filter that defines a response characteristic of the first sound wave to the reference signal based on the reference signal and the specific signal, and realizes active noise control by radiating the first sound wave from the first speaker based on the reference signal and the control filter, Active noise control system. [Effects of the Invention]

[0008] The technique according to the present invention is suitable for realizing ANC that is robust to the location of the noise source. [Brief explanation of the drawings]

[0009] [Figure 1A] 1 is a perspective view of an ANC system according to an embodiment. FIG. [Figure 1B] FIG. 1 is a top view of an ANC system according to an embodiment. [Figure 1C] FIG. 1 is a side view of an ANC system according to an embodiment. [Figure 2A] FIG. 1 is an explanatory diagram of noise propagation in a symmetrical arrangement. [Figure 2B] FIG. 1 is an explanatory diagram of noise propagation in an asymmetric arrangement. [Figure 2C] FIG. 1 is an explanatory diagram of sounds emitted by speakers arranged in the left-right direction. [Figure 3A] FIG. 10 is a schematic explanatory diagram of control of a first sound wave in a control stage. [Figure 3B] FIG. 10 is a schematic explanatory diagram of the control of the second sound wave in the control stage. [Figure 4] FIG. 10 is a schematic diagram showing an example of a plurality of reference positions. [Figure 5A] FIG. 10 is a signal processing block diagram of the control device in the tuning stage, which is related to preparation for control of the first sound wave. [Figure 5B] FIG. 10 is a signal processing block diagram of the control device in the tuning stage, which is related to preparation for control of the second sound wave. [Figure 6A] FIG. 2 is a signal processing block diagram of a control device in a control stage, which is related to the control of a first sound wave. [Figure 6B] FIG. 10 is a signal processing block diagram of a control device in a control stage, which is related to the control of a second sound wave. [Figure 7A] FIG. 10 is a signal processing block diagram of a position estimation unit and memory in the tuning stage, relating to control of a first sound wave. [Figure 7B]FIG. 10 is a signal processing block diagram of a position estimation unit and memory in the tuning stage, relating to control of a second sound wave. [Figure 7C] FIG. 10 shows a block diagram for explaining the operation of a correction unit in a tuning stage. [Figure 8A] FIG. 10 is a signal processing block diagram of a position estimation unit and memory in the control stage, which is related to the control of the first sound wave. [Figure 8B] FIG. 10 is a signal processing block diagram of a position estimation unit and memory in the control stage, which is related to the control of the second sound wave. [Figure 8C] FIG. 10 shows a block diagram for explaining the operation of a correction unit in the control stage. [Figure 9] FIG. 10 shows a specific example of a signal processing block diagram of a control device in a tuning stage. [Figure 10] FIG. 1 shows a specific example of a signal processing block diagram of a control device in a control stage. [Figure 11A] 4 is an explanatory diagram of a first support and a first radiation surface of a first speaker. FIG. [Figure 11B] 10 is an explanatory diagram of a second support and a second radiation surface of a second speaker. FIG. [Figure 12A] 4 is an explanatory diagram of a first support and a first radiation surface of a first speaker. FIG. [Figure 12B] 10 is an explanatory diagram of a second support and a second radiation surface of a second speaker. FIG. [Figure 13A] 1 is a cross-sectional view of a piezoelectric speaker taken along a plane parallel to its thickness direction. [Figure 13B] FIG. 10 is a view of the piezoelectric speaker as viewed from the opposite side to the fixing surface. [Figure 14] 10A and 10B are diagrams illustrating a piezoelectric speaker according to another configuration example. [Figure 15] 10 is a flowchart showing a specific example of signal processing by a control device that can be applied to the embodiment. [Figure 16] 1 is a flowchart illustrating an ANC method according to an embodiment. [Figure 17]FIG. 10 is a top view of an ANC system according to another embodiment. [Figure 18A] FIG. 1 is a top view of the first experimental setup. [Figure 18B] FIG. 1 is a side view of the first experimental arrangement. [Figure 19A] FIG. 10 is a top view of the second experimental setup. [Figure 19B] FIG. 10 is a side view of the second experimental setup. [Figure 20A] FIG. 10 is a top view of the third experimental setup. [Figure 20B] FIG. 10 is a side view of the third experimental setup. [Figure 21] FIG. 1 is an explanatory diagram of the arrangement of speakers in the first to third experimental arrangements. [Figure 22] FIG. 10 is an explanatory diagram of a reference position. [Figure 23] FIG. 10 is an explanatory diagram of a reference position. [Figure 24A] 1 is a graph showing the change over time in the silencing effect of Experimental Examples 1-A, 1-B, and 1-C. [Figure 24B] 1 is a graph showing the change over time in the silencing effect of Experimental Examples 1-A, 1-B, and 1-C. [Figure 25A] 10 is a graph showing the change over time in the silencing effect of Experimental Examples 2-A, 2-B, and 2-C. [Figure 25B] 10 is a graph showing the change over time in the silencing effect of Experimental Examples 2-A, 2-B, and 2-C. [Figure 26A] 10 is a graph showing the change over time in the noise reduction effect of Experimental Examples 3-A, 3-B, and 3-D. [Figure 26B] 10 is a graph showing the change over time in the silencing effect of Experimental Examples 3-A, 3-B, and 3-D. [Figure 27A] 10 is a graph showing the change over time in the silencing effect of Experimental Examples 4-A, 4-B, and 4-D. [Figure 27B] 10 is a graph showing the change over time in the noise reduction effect of Experimental Examples 4-A, 4-B, and 4-D. [Figure 28]10 is a graph showing the change over time in the position corresponding to the noise control filter in the control stage of Experimental Example 1-A. DETAILED DESCRIPTION OF THE INVENTION

[0010] (Contents of study by the inventors) Active noise control (ANC) systems can be configured to reduce sound from a noise source located at a certain location. When an ANC system is configured in this way, the noise reduction characteristics can deteriorate if the noise source moves from that location. The noise reduction characteristics can be determined, for example, by a noise control filter.

[0011] It is not necessarily easy to implement ANC that is robust against the position of a noise source. Therefore, the present inventors have investigated an ANC that is suitable for realizing ANC that is robust against the position of a noise source.

[0012] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, the following is merely an example of an embodiment of the present invention and is not intended to limit the present invention. In the following, terms such as "upper," "lower," "left," "right," "front," "rear," and "height" are used to specify the relative positions of elements and are not intended to limit the orientation of these elements when ANC is performed. Furthermore, in the following, the same or similar components will be denoted by the same reference numerals, and their description may be omitted.

[0013] [Active Noise Control System] 1A to 1C show an active noise control (ANC) system 500. The ANC system 500 performs ANC. The ANC system 500 is provided on a floor 85. The ANC system 500 includes a structure 80, a first support 90A, a second support 90B, a first speaker 10A, and a second speaker 10B. The structure 80 defines a space 86 therein.

[0014] The structure 80 includes a first wall 80A, a second wall 80B, and a third wall 80C. The first wall 80A and the second wall 80B are connected via the third wall 80C. The first support 90A is attached to the first wall 80A. The second support 90B is attached to the second wall 80B. A gap 93 exists between the first support 90A and the second support 90B. The first support 90A supports the first speaker 10A. The second support 90B supports the second speaker 10B. The first support 90A and the second support 90B are plate-like bodies. The plate-like bodies may be flat or curved. The gap 93 may or may not be present.

[0015] The first speaker 10A is attached to one of the two main surfaces of the first support 90A that faces the outside of the space 86. The second speaker 10B is attached to one of the two main surfaces of the second support 90B that faces the outside of the space 86. The first speaker 10A has a first radiation surface 15A. The second speaker 10B has a second radiation surface 15B. The first radiation surface 15A and the second radiation surface 15B face in a direction away from the space 86. Here, the direction that the radiation surface faces refers to the direction in which the area of ​​the radiation surface appears to be the largest.

[0016] 1A to 1C show a vertical direction D1, a left-right direction D2, and a front-rear direction D3. The vertical direction D1, the left-right direction D2, and the front-rear direction D3 are perpendicular to one another. The front-rear direction D3 is a direction in which a reference axis O extends. The reference axis O passes through the space 86 and between the first radiation surface 15A and the second radiation surface 15B. The reference axis O is a virtual axis. Specifically, the reference axis O does not intersect with either the first radiation surface 15A or the second radiation surface 15B.

[0017] The first wall 80A and the second wall 80B face each other in the left-right direction D2. The space 86 is located between the first wall 80A and the second wall 80B in the left-right direction D2. The first radiation surface 15A and the second radiation surface 15B are located between the first wall 80A and the second wall 80B in the left-right direction D2.

[0018] In the front-rear direction D3, the space 86 is located between the combination of the first support 90A and the second support 90B and the third wall 80C. In the front-rear direction D3, the space 86 is located between the combination of the first speaker 10A and the second speaker 10B and the third wall 80C.

[0019] The type of the first speaker 10A is not particularly limited. The first speaker 10A may be a piezoelectric speaker or a dynamic speaker. The piezoelectric speaker may be a film speaker. The same applies to the second speaker 10B. In FIGS. 1A to 1C and 11A to 12B, the first speaker 10A and the second speaker 10B are depicted as piezoelectric speakers.

[0020] 1B, a noise source 200 may be located inside the space 86. Noise from the noise source 200 is emitted to the outside of the space 86. The noise source 200 may move in any direction. Specifically, the direction of movement of the noise source 200 may include at least one selected from the group consisting of an up-down direction D1 component, a left-right direction D2 component, and a front-back direction D3 component.

[0021] In one specific example, a booth is configured including a structure 80, a first support 90A, and a second support 90B. The first support 90A and the second support 90B are doors. A person inside the booth is the noise source 200. The booth may be a personal booth. The first wall 80A, the second wall 80B, and the third wall 80C may be components such as partitions, and the booth as the structure 80 may be configured by assembling these components.

[0022] The ANC system 500 can reduce noise from the noise source 200 outside the space 86. Specifically, the first radiation surface 15A radiates a first sound wave for silencing the noise that travels away from the space 86. The second radiation surface 15B radiates a second sound wave for silencing the noise that travels away from the space 86.

[0023] The first radiation surface 15A and the second radiation surface 15B are arranged so that noise from the noise source 200 can be easily reduced outside the space 86 by the first sound wave and the second sound wave even if the position of the noise source 200 changes in the left-right direction D2. This point will be explained below with reference to FIGS. 2A to 2C. Note that the first speaker 10A and the second speaker 10B are not shown in FIGS. 2A and 2B. In FIG. 2C, the first speaker 10A and the second speaker 10B are depicted as dynamic speakers.

[0024] In the example of Fig. 2A, noise source 200 is arranged in the center of space 86 in the left-right direction D2. Hereinafter, this arrangement of noise source 200 will be referred to as a symmetrical arrangement. In contrast, in the example of Fig. 2B, noise source 200 is arranged in a position shifted from the center of space 86 in the left-right direction D2. Hereinafter, this arrangement of noise source 200 will be referred to as an asymmetrical arrangement. Specifically, in the example of Fig. 2B, noise source 200 is arranged in a position shifted to the left from the center of space 86 in the left-right direction D2.

[0025] 2A and 2B, reference numeral 200w0 denotes a wavefront formed by the noise source 200 near the noise source 200. Reference numeral 200w1 denotes a wavefront formed by the noise source 200 near the first support 90A. Reference numeral 200w2 denotes a wavefront formed by the noise source 200 near the second support 90B.

[0026] In the example of Fig. 2A, a wavefront from noise source 200 propagates evenly in the left-right direction D2 and forward in the front-rear direction D3. Wavefronts 200w1 and 200w2 have approximately the same phase and amplitude. In contrast, in the example of Fig. 2B, wavefronts 200w1 and 200w2 have different phases and amplitudes.

[0027] When the noise source 200 is movable in the left-right direction D2, the noise source 200 can be arranged either symmetrically or asymmetrically. In order to reduce the sound generated by the noise source 200 outside the space 86 whether the noise source 200 is arranged symmetrically or asymmetrically, it is possible to change the sound waves for silencing emitted by the ANC system 500 depending on whether the noise source 200 is arranged symmetrically or asymmetrically. Specifically, the ANC by the ANC system 500 In the case of a symmetrical arrangement, a wavefront having an opposite phase and an equal amplitude to the wavefront 200w1 in FIG. 2A is formed in the vicinity of the first support 90A, In the case of a symmetrical arrangement, a wavefront having an opposite phase and an equal amplitude to the wavefront 200w2 in FIG. 2A is formed in the vicinity of the second support 90B, In the case of an asymmetric arrangement, a wavefront having an opposite phase and an equal amplitude to the wavefront 200w1 in FIG. 2B is formed in the vicinity of the first support 90A, In the case of an asymmetric arrangement, a wavefront having an opposite phase and an equal amplitude to the wavefront 200w2 in FIG. 2B is formed in the vicinity of the second support 90B. This is because, in both the symmetrical and asymmetrical arrangements, outside the space 86, the sound originating from the noise source 200 is cancelled out by the sound originating from the ANC, making the space quieter.

[0028] In other words, when noise source 200 moves in left-right direction D2, the difference between the sound waves that noise source 200 forms near first support 90A and the sound waves that noise source 200 forms near second support 90B changes. In order to maintain the cancellation state described above and keep the outside of space 86 quiet even when this change occurs, it is effective to change the difference between the sound waves that ANC forms near first support 90A and the sound waves that ANC forms near second support 90B in accordance with this change.

[0029] In this regard, in this embodiment, as shown in FIGS. 1A to 1C and 2C, at least a portion of the first radiation surface 15A and at least a portion of the second radiation surface 15B are located at different positions in the left-right direction D2. The difference between the first sound wave on the first radiation surface 15A and the second sound wave on the second radiation surface 15B changes depending on the sound traveling from the inside of the space 86 to the outside of the space 86, so that the sound from the inside of the space 86 is reduced outside the space 86. This configuration is suitable for achieving ANC that is robust against the position of the noise source 200 in the left-right direction D2. Specifically, it is suitable for achieving a wide sweet spot in the left-right direction D2. Here, the sweet spot is a range within which the noise source 200 is allowed to be located in order to reduce the sound from the noise source 200 to a practical level at a desired control point. In this embodiment, the control point is set outside the space 86.

[0030] Specifically, in this embodiment, the first speaker 10A and the second speaker 10B are individually controlled while the positions of the first radiation surface 15A and the second radiation surface 15B are at least partially shifted in the left-right direction D2. This makes it easy to adjust the difference between the first sound wave on the first radiation surface 15A and the second sound wave on the second radiation surface 15B. Through this adjustment, it is possible to change the difference between the sound wave near the first support 90A and the sound wave near the second support 90B formed by ANC. This allows the cancellation state described above to be maintained.

[0031] In the illustrated example, at least a portion of the first speaker 10A and at least a portion of the second speaker 10B are located at different positions in the left-right direction D2. Specifically, the first speaker 10A and the second speaker 10B are aligned in the left-right direction D2. Furthermore, the first radiation surface 15A and the second radiation surface 15B are aligned in the left-right direction D2.

[0032] Here, the phase of the first sound wave at the first radiation surface 15A is defined as the first phase. The sound pressure of the first sound wave at the first radiation surface 15A is defined as the first sound pressure. The phase of the second sound wave at the second radiation surface 15B is defined as the second phase. The sound pressure of the second sound wave at the second radiation surface 15B is defined as the second sound pressure. Specifically, in this embodiment, the difference between the first phase and the second phase and the difference between the first sound pressure and the second sound pressure change depending on the sound going out from inside the space 86 to outside the space 86 so that the sound from inside the space 86 is reduced outside the space 86.

[0033] As shown in FIGS. 1A to 1C, the ANC system 500 is provided with an upper opening 87 and a lower opening 88. The upper opening 87 is defined by an upper end 901A of the first support 90A, an upper end 901B of the second support 90B, the first wall 80A, and the second wall 80B. The lower opening 88 is defined by a lower end 90mA of the first support 90A, a lower end 90mB of the second support 90B, the first wall 80A, and the second wall 80B. In the illustrated example, the upper opening 87 is also defined by a third wall 80C. The lower opening 88 is also defined by a floor 85.

[0034] In the above configuration, ANC that reduces diffracted sound from the upper opening 87 and the lower opening 88 can be performed while preventing sound leakage from the left-right direction D2 by the first wall 80A and the second wall 80B (that is, with a small number of diffraction paths), which is advantageous from the viewpoint of ensuring the accuracy of ANC.

[0035] ANC that reduces diffracted sound can be performed effectively when the first speaker 10A and the second speaker 10B are piezoelectric speakers, as described in Patent Documents 1 and 2.

[0036] In one specific example, the first speaker 10A and the second speaker 10B are piezoelectric speakers. The first radiation surface 15A has a central region and regions on both sides thereof (i.e., upper and lower regions) in the longitudinal direction D1A (see FIG. 11A, etc., described later). The second radiation surface 15B has a central region and regions on both sides thereof (i.e., upper and lower regions) in the longitudinal direction D1B (see FIG. 11B, etc., described later). The ANC system 500 includes: (1) A sound wave traveling forward in the front-rear direction D3 and downward in the up-down direction D1 is emitted from the upper region of the first radiation surface 15A, (2) A sound wave traveling forward in the front-rear direction D3 and upward in the up-down direction D1 is emitted from the lower region of the first radiation surface 15A, (3) A sound wave traveling forward in the front-rear direction D3 and downward in the up-down direction D1 is emitted from the upper region of the second radiation surface 15B, (4) Sound waves traveling forward in the front-rear direction D3 and upward in the up-down direction D1 can be emitted from the lower region of the second radiation surface 15B. The propagation directions of the sound waves (1), (2), (3), and (4) are close to the propagation directions of the diffracted waves from the upper end 901A, the lower end 90mA, the upper end 901B, and the lower end 90mB originating from the noise source 200. This is advantageous from the viewpoint of canceling out these diffracted waves originating from the noise source 200 and reducing the diffracted sound originating from the noise source 200.

[0037] In the illustrated example, the upper opening 87 includes a front opening and a ceiling opening. The front opening is defined by the upper end 90lA of the first support 90A, the upper end 90lB of the second support 90B, the front end 80fA of the first wall 80A, and the front end 80fB of the second wall 80B. The front opening opens toward the front in the front-to-rear direction D3. In a modified example in which the structure 80 includes a top plate, the front opening may also be defined by the top plate. The ceiling opening is defined by the upper end 80lA of the first wall 80A and the upper end 80lB of the second wall 80B. Specifically, the ceiling opening is also defined by the upper end 80lC of the third wall 80C. The ceiling opening opens toward the upper side in the up-down direction D1. In the illustrated example, the upper end 90lA of the first support 90A and the upper end 90lB of the second support 90B do not contribute to defining the ceiling opening. However, the ceiling opening may also be defined by upper end 901A and upper end 901B.

[0038] In the illustrated example, the lower opening 88 includes a front opening. The front opening is defined by a lower end 90mA of the first support 90A, a lower end 90mB of the second support 90B, a front end 80fA of the first wall 80A, and a front end 80fB of the second wall 80B. Specifically, the front opening is also defined by the floor 85. The front opening opens toward the front in the front-to-back direction D3. The lower opening 88 may include a bottom opening. The bottom opening may be defined by a lower end 80mA of the first wall 80A and a lower end 80mB of the second wall 80B. Specifically, the bottom opening may also be defined by a lower end 80mC of the third wall 80C. The bottom opening opens toward the downward in the up-down direction D1. The floor 85 does not have to contribute to defining the front opening. The lower end 90mA of the first support 90A and the lower end 90mB of the second support 90B may not contribute to the definition of the bottom opening, and the bottom opening may also be defined by the lower end 90mA and the lower end 90mB.

[0039] The upper opening 87 may include both a front opening and a ceiling opening, or may include only one of them. The lower opening 88 may include both a front opening and a bottom opening, or may include only one of them.

[0040] In the ANC system 500, the area of ​​the front opening of the upper opening 87 may be larger than the area of ​​the front opening of the lower opening 88, may be smaller than the area of ​​the front opening of the lower opening 88, or may be the same as the area of ​​the front opening of the lower opening 88. Only one of the upper opening 87 and the lower opening 88 may be provided.

[0041] As shown in FIG. 1B , the ANC system 500 has an angled state. The angled state is a state in which at least one condition selected from the group consisting of a first condition and a second condition is satisfied. The first condition is that the first radiation surface 15A faces in a direction deviating from the front-rear direction D3 toward the reference axis O. The second condition is that the second radiation surface 15B faces in a direction deviating from the front-rear direction D3 toward the reference axis O. In FIG. 1B , the first angle θ1 is the angle of deviation of the direction in which the first radiation surface 15A faces from the front-rear direction D3 toward the reference axis O. The second angle θ2 is the angle of deviation of the direction in which the second radiation surface 15B faces from the front-rear direction D3 toward the reference axis O. In the angled state, the first angle θ1 and / or the second angle θ2 are positive values. The first angle θ1 and the second angle θ2 may be the same or different.

[0042] The above configuration is advantageous from the viewpoint of exhibiting good sound deadening characteristics. Specifically, high-frequency components of sound emitted by the noise source 200 in the space 86 tend to be reflected by the first wall 80A and / or the second wall 80B and then exit the space 86. The propagation direction of the reflected waves that exit in this manner may be a direction deviated from the front-rear direction D3 toward the reference axis O. In this regard, in the above configuration, the first radiation surface 15A and / or the second radiation surface 15B face in a direction deviated from the front-rear direction D3 toward the reference axis O. This makes it easy for the first sound wave and / or the second sound wave to propagate in the deviated direction. This makes it easy to align the propagation direction of the reflected wave that exits the space 86 with the propagation direction of the first sound wave and / or the second sound wave. Therefore, good sound deadening characteristics can be exhibited.

[0043] The above expression "the ANC system 500 has an angled state" will be explained. This expression is not intended to necessarily mean that the state of the ANC system 500 is always in the angled state. This expression encompasses the case where the ANC system 500 is configured so that the first angle θ1 and the second angle θ2 are variable and can take on values ​​of zero or positive values. For example, the first support 90A is a door, and the first angle θ1 changes depending on the degree to which the door is opened. Also, for example, the second support 90B is a door, and the second angle θ2 changes depending on the degree to which the door is opened.

[0044] Specifically, the first condition is that the first radiation surface 15A faces in a direction deviating from the front-rear direction D3 toward the reference axis O by an angle of more than 0° and not more than 45°. The second condition is that the second radiation surface 15B faces in a direction deviating from the front-rear direction D3 toward the reference axis O by an angle of more than 0° and not more than 45°.

[0045] More specifically, the first condition is that the first radiation surface 15A faces in a direction deviated from the front-rear direction D3 by 10° to 45° toward the reference axis O. The second condition is that the second radiation surface 15B faces in a direction deviated from the front-rear direction D3 by 10° to 45° toward the reference axis O.

[0046] In the angled state, only one of the first condition and the second condition may be satisfied, or both may be satisfied. Also, the ANC system 500 does not have to have an angled state.

[0047] The ANC system 500 may have a no-angle state. The no-angle state is a state in which neither the first condition nor the second condition is met. In other words, the no-angle state is a state in which the first angle θ1 and the second angle θ2 are 0°. The ANC system 500 may have both the angled state and the no-angle state.

[0048] In one specific example, the first support 90A is connected to the first wall 80A via an attachment (not shown) so as to be rotatable about the vertical direction D1 as the rotation axis. The second support 90B is connected to the second wall 80B via an attachment (not shown) so as to be rotatable about the vertical direction D1 as the rotation axis. This specific example makes it possible to realize an ANC system 500 that has both an angled state and a non-angled state.

[0049] The ANC system 500 may have a locking mechanism that fixes the first angle θ1 and the second angle θ2. The ANC system 500 may have a stopper that limits the range of the first angle θ1 and the second angle θ2 so that the first angle θ1 and the second angle θ2 do not deviate from an appropriate range.

[0050] The first angle θ1 and / or the second angle θ2 may be changed automatically. For example, a configuration can be adopted in which the position of noise source 200 is acquired by a sensor such as an imaging device, and the first angle θ1 and / or the second angle θ2 is automatically changed based on the acquired position.

[0051] 1B, the ANC system 500 includes a reference microphone 130, a first error microphone 140A, a second error microphone 140B, a specific microphone 150, a first amplifier 111A, a second amplifier 111B, and a control device 110. The control device 110 realizes ANC of the ANC system 500. Note that the first amplifier 111A, the second amplifier 111B, and the control device 110 are not shown in FIGS. 1A and 1C. The reference microphone 130, the first error microphone 140A, the second error microphone 140B, and the specific microphone 150 are not shown in FIG. 1A.

[0052] A reference microphone 130 and a specific microphone 150 are used in the tuning stage and the control stage. In this embodiment, the reference microphone 130 is located inside the space 86. The specific microphone 150 is located outside the space 86. A first error microphone 140A and a second error microphone 140B are used in the tuning stage. In this embodiment, the first error microphone 140A and the second error microphone 140B are located outside the space 86. The first error microphone 140A and the second error microphone 140B may be removed in the control stage.

[0053] In the tuning stage, the controller 110 prepares to control the first sound wave in the control stage using the first amplifier 111A, the reference microphone 130, the first error microphone 140A, the second error microphone 140B, and the particular microphone 150. In the control stage, the controller 110 controls the first sound wave using the first amplifier 111A, the reference microphone 130, and the particular microphone 150.

[0054] In the tuning stage, the controller 110 prepares to control the second sound wave in the control stage using the second amplifier 111B, the reference microphone 130, the first error microphone 140A, the second error microphone 140B, and the particular microphone 150. In the control stage, the controller 110 controls the second sound wave using the second amplifier 111B, the reference microphone 130, and the particular microphone 150.

[0055] The control of the ANC system 500 will be described in detail below with reference to Figures 3A to 10. In the following, the terms noise control filter W1, noise control filter W2, reference signal x, first error signal e1, second error signal e2, specific signal j, signal level ratio j / x, and position discrimination function G may be used. In this embodiment, these terms are defined as follows: The noise control filter W1 is a filter that defines the response characteristic RC1 of the first sound wave with respect to the reference signal x. The noise control filter W2 is a filter that defines the response characteristic RC2 of the second sound wave with respect to the reference signal x. The reference signal x is the detection signal of the reference microphone 130 . The first error signal e1 is the detection signal of the first error microphone 140A. The second error signal e2 is the detection signal of the second error microphone 140B. The specific signal j is a detection signal of the specific microphone 150 . The signal level ratio j / x is the ratio of the level of the reference signal x to the level of the specific signal j, specifically, the ratio of the level of the specific signal j to the level of the reference signal x. The position discrimination function G is a function that represents the signal level ratio j / x. In this embodiment, the level of the reference signal x, the level of the specific signal j, the signal level ratio j / x, and the position discriminant function G are vectors.

[0056] The control of the first and second sound waves in the control stage is shown schematically in FIGS. 3A and 3B, respectively. As shown in FIGS. 3A and 3B, the control device 110 stores a priori estimate 73. The priori estimate 73 is prepared in the tuning stage and is used for ANC in the control stage. The priori estimate 73 includes multiple parameter sets 72. Each parameter set 72 includes a portion defining a noise control filter W1, a portion defining a noise control filter W2, and a portion defining a position discriminant function G. The multiple parameter sets 72 are associated one-to-one with multiple reference positions 71. The number of combinations of the associated reference positions 71 and parameter sets 72 is not particularly limited. The designer of the ANC system 500 can set multiple reference positions 71 to cover the expected range of movement of the noise source 200.

[0057] In the control stage, the control device 110 calculates a position discriminant function G. From among a plurality of parameter sets 72, the control device 110 selects a parameter set 72 including a position discriminant function G that is closest to the calculated position discriminant function G. The position discriminant function G can be considered as position information of the noise source 200. The selection of the parameter set 72 can be considered as a process of determining, from a plurality of reference positions 71, the reference position 71 that is closest to the position of the noise source 200 (hereinafter referred to as a position discrimination process).

[0058] As shown in Fig. 3A, the control device 110 controls the first sound wave using the reference signal x and the noise control filter W1 included in the selected parameter set 72. As shown in Fig. 3B, the control device 110 controls the second sound wave using the reference signal x and the noise control filter W2 included in the selected parameter set 72. As a result, the noise from the noise source 200 is reduced by the first sound wave from the first speaker 10A and the second sound wave from the second speaker 10B at the first control point 74A and the second control point 74B. In this embodiment, the first control point 74A and the second control point 74B are set outside the space 86.

[0059] The calculation of the position discriminant function G is performed sequentially. Therefore, the selected parameter set 72 can be changed sequentially according to the movement of the noise source 200. That is, the noise control filter W1 and the noise control filter W2 to be used can be changed sequentially according to the movement of the noise source 200. In this way, the first sound wave and the second sound wave that are suited to the position of the noise source 200 are formed, and a state in which noise is effectively reduced at the first control point 74A and the second control point 74B can be maintained.

[0060] 3A and 3B, the space 86 includes a plurality of reference positions 71. The plurality of reference positions 71 includes the reference positions 71. α , reference position 71 β and reference position 71 γ The plurality of parameter sets 72 includes the parameter sets 72 α , parameter set 72 β and parameter set 72 γ Includes reference position 71α But parameter set 72 α It is associated with reference position 71. β But parameter set 72 β It is associated with reference position 71. γ But parameter set 72 γ 3A and 3B, the reference position 71 α There is a noise source 200 in the parameter set 72 α It depicts how it is used.

[0061] The arrangement direction of at least two of the multiple reference positions 71 may include at least one selected from the group consisting of a vertical direction D1 component, a horizontal direction D2 component, and a front-to-back direction D3 component. FIG. 4 shows an example of the multiple reference positions 71. In this example, the multiple reference positions 71 include nine reference positions 71. The nine reference positions 71 are arranged in an array of three rows and three columns. The rows extend in the horizontal direction D2, and the columns extend in the front-to-back direction D3. Note that the first speaker 10A and the second speaker 10B are not shown in FIG. 4.

[0062] 5A and 5B are signal processing block diagrams of the control device 110 in the tuning stage, which are related to the preparation for controlling the first and second acoustic waves, respectively. As shown in FIGS. 5A and 5B, the control device 110 includes a position estimator 160, a memory 170, and a processor 180.

[0063] In the tuning stage, a noise source 200 is placed at one of a plurality of reference positions 71. As shown in Figures 5A and 5B, a first error microphone 140A is placed at a first control point 74A, and a second error microphone 140B is placed at a second control point 74B. The reference microphone 130 detects sound and generates a reference signal x.

[0064] As shown in Fig. 5A, in the tuning stage, the processor 180 generates a control signal y1 based on the noise control filter W1 and the reference signal x of the reference microphone 130. The control signal y1 is amplified by a first amplifier 111A (not shown in Fig. 5A). The first speaker 10A emits a first sound wave based on the amplified control signal y1.

[0065] As shown in Fig. 5B, in the tuning stage, the processor 180 generates a control signal y2 based on the noise control filter W2 and the reference signal x of the reference microphone 130. The control signal y2 is amplified by a second amplifier 111B, which is not shown in Fig. 5B. The second speaker 10B emits a second sound wave based on the amplified control signal y2.

[0066] 5A, in the tuning stage, the first error microphone 140A detects sound and generates a first error signal e1. The processor 180 identifies the filter coefficients of the noise control filter W1 so that the first error signal e1 is reduced. The identification can be performed using an adaptive algorithm.

[0067] 5B, in the tuning stage, the second error microphone 140B detects sound and generates a second error signal e2. The processor 180 identifies the filter coefficients of the noise control filter W2 so that the second error signal e2 is small. The identification can be performed using an adaptive algorithm.

[0068] 5A and 5B, in the tuning stage, the position estimation unit 160 generates position information of the noise source 200 based on the reference signal x and the specific signal j. The position information is specifically a position discriminant function G.

[0069] In the tuning stage of this embodiment, the position estimation unit 160 generates a position discriminant function G as position information of the noise source 200 based on the reference signal x, the specific signal j, the noise control filter W1, and the noise control filter W2. The noise control filter W1 and the noise control filter W2 are acquired from the processor 180. Specifically, in the position estimation unit 160, the filter coefficients of the noise control filters W1 and W2 are used for correction to reduce or cancel out the influence of the first sound wave and the second sound wave in the specific signal j. Then, the position estimation unit 160 generates the position discriminant function G based on the reference signal x and the corrected specific signal j.

[0070] In the tuning stage, the memory 170 stores the position discriminant function G, the noise control filter W1 having the identified filter coefficients, and the noise control filter W2 having the identified filter coefficients, in a state where they are associated with each other.

[0071] Such tuning stages are performed by sequentially changing the position of the noise source 200 to a plurality of reference positions 71. As a result, a plurality of noise control filters W1, a plurality of noise control filters W2, and a plurality of position discriminant functions G are obtained. In the memory 170, the plurality of position discriminant functions G and the plurality of noise control filters W1 are stored in one-to-one correspondence, and the plurality of position discriminant functions G and the plurality of noise control filters W2 are also stored in one-to-one correspondence. In this way, a plurality of parameter sets 72 are obtained that are in one-to-one correspondence with the plurality of reference positions 71. As described above, each parameter set 72 includes a portion that defines the noise control filter W1, a portion that defines the noise control filter W2, and a portion that defines the position discriminant function G.

[0072] After the tuning stage, the control stage is executed. The signal processing block diagrams of the control device 110 in the control stage in Figures 6A and 6B relate to the control of the first and second sound waves, respectively. In the description of the control stage, descriptions that overlap with the tuning stage may be omitted.

[0073] 6A and 6B, in the control stage, the position estimation unit 160 generates current position information of the noise source 200 based on the current reference signal x and the current specific signal j. The current position information is specifically the current position discriminant function G. Of the multiple position discriminant functions G stored in the tuning stage, the one that is closest to the current position discriminant function G is identified.

[0074] As shown in Fig. 6A, in the control stage, a noise control filter W1 associated with the identified position discriminant function G is selected from the multiple noise control filters W1 stored in the tuning stage. The processor 180 generates a control signal y1 based on the selected noise control filter W1 and the reference signal x of the reference microphone 130. The control signal y1 is amplified by a first amplifier 111A (not shown in Fig. 6A). The first speaker 10A emits a first sound wave based on the amplified control signal y1.

[0075] As shown in Fig. 6B, in the control stage, a noise control filter W2 associated with the identified position discriminant function G is selected from the multiple noise control filters W2 stored in the tuning stage. The processor 180 generates a control signal y2 based on the selected noise control filter W2 and the reference signal x of the reference microphone 130. The control signal y2 is amplified by a second amplifier 111B, which is not shown in Fig. 6B. The second speaker 10B emits a second sound wave based on the amplified control signal y2.

[0076] In the control stage of this embodiment, the position estimation unit 160 generates a position discriminant function G as current position information of the noise source 200 based on the current reference signal x, the current specific signal j, the current (specifically, determined in the most recent control cycle) noise control filter W1, and the current (specifically, determined in the most recent control cycle) noise control filter W2. The noise control filter W1 and the noise control filter W2 are acquired from the processor 180. Specifically, in the position estimation unit 160, the filter coefficients of the noise control filters W1 and W2 are used for correction to reduce or cancel the effects of the first sound wave and the second sound wave in the current specific signal j. The position estimation unit 160 then generates the current position discriminant function G based on the current reference signal x and the corrected specific signal j. Of the multiple position discriminant functions G stored in the tuning stage, the one closest to the current position discriminant function G is identified. Of the multiple noise control filters W1 and multiple noise control filters W2 stored in the tuning stage, the noise control filter W1 and multiple noise control filters W2 associated with the identified position discriminant function G are selected. In this way, the noise control filters W1 and W2 used for ANC are updated to match the position of the noise source 200.

[0077] In this way, the control stage selects noise control filter W1 and noise control filter W2 that are suited to the position of noise source 200, and forms first and second sound waves that are suited to the position of noise source 200, thereby maintaining a state in which noise is effectively reduced at first control point 74A and second control point 74B.

[0078] The operation of the position estimator 160 and memory 170 in the tuning stage will be explained in more detail with reference to Figures 7A and 7B for controlling the first and second acoustic waves, respectively.

[0079] In the following description, "(n)" indicates an expression in the time domain, specifically relating to time step n. For example, the position discriminant function G at time step n may be expressed as the position discriminant function G(n). The reference signal x of the reference microphone 130 at time step n may be expressed as the reference signal x(n). The specific signal j of the specific microphone 150 at time step n may be expressed as the specific signal j(n).

[0080] 7A and 7B, the position estimation unit 160 includes an adaptive filter 161 as a position discriminant function G(n), an updating unit 162, a subtraction unit 163, and a correction unit 164. The memory 170 includes a filter holding unit 171 and a position information holding unit 172.

[0081] In the tuning stage, the coefficients of the position discriminant function G(n) are updated by an updating unit 162, a subtracting unit 163, and a correcting unit 164. In this embodiment, the updating unit 162 identifies the coefficients of the position discriminant function G(n) based on the NLMS (Normalized Least Mean Square) algorithm.

[0082] In the illustrated example, x(n)G(n), which is the product of x(n) and G(n), is calculated. The correction unit 164 corrects j(n) according to x(n). This correction reduces or cancels the influence of the first sound wave and the second sound wave on j(n).

[0083] As shown in Fig. 7C, the correction unit 164 includes a convolution unit 164a and a subtraction unit 164b. The correction unit 164 receives noise control filters W1 and W2 from the processor 180. In Fig. 7C, S1 # is a secondary path filter from the first speaker 10A to the specific microphone 150. S2 # is a secondary path filter from the second speaker 10B to the specific microphone 150. The convolution unit 164a passes x(n) through the noise control filter W1 and the secondary path filter S1 #The signal obtained by convolution processing with x(n) is passed through the noise control filter W2 and the secondary path filter S2. # The pseudo signal k(n) is generated by adding the signal obtained by convolution processing with the noise control filter W1 and the signal obtained by convolution processing with the noise control filter W2. The subtraction unit 164b subtracts k(n) from j(n) generated by the specific microphone 150. This corrects j(n) and reduces or cancels out the influence of the first sound wave and the second sound wave in j(n). In this way, the convolution unit 164a uses the noise control filter W1, the secondary path filter S1, # , noise control filter W2 and secondary path filter S2 # Then, j(n) is corrected according to x(n) by convolution processing using the correction filter. This correction improves the estimation accuracy of the noise source 200, and makes it possible to calculate G(n) that matches the position of the noise source 200 with high accuracy. Note that the secondary path filter S1 # and secondary path filter S2 # can be pre-modeled and implemented in the correction unit 164 before the tuning stage.

[0084] A subtraction unit 163 calculates q(n)=j(n)-x(n)G(n), which is the difference obtained by subtracting x(n)G(n) from the corrected j(n). An update unit 162 determines G(n) so that q(n) converges to zero. Specifically, G(n) is calculated to satisfy the following equation 1. In equation 1, μ is the step size.

number

[0085] q(n)=j(n)-x(n). The convergence of G(n) to zero means that G(n) converges to j(n) / x(n). As can be understood from this, G(n) calculated by adaptive filter 161 represents the ratio j(n) / x(n) of specific signal j(n) to reference signal x(n). Specifically, G(n) represents the signal level ratio j(n) / x(n), which is the ratio of the level of specific signal j(n) to the level of reference signal x(n).

[0086] G(n) is the position discriminant function G expressed in the time domain. Here, the position discriminant function G expressed in the frequency domain is written as G(z). G(n) can be converted to G(z), and G(z) can be converted to G(n). Furthermore, G(z) = J(z) / X(z). Here, X(z) is the transfer function from the noise source 200 to the reference microphone 130. J(z) is the transfer function from the noise source 200 to the specific microphone 150.

[0087] As described above, in the tuning stage, processor 180 identifies the filter coefficients of noise control filter W1 and the filter coefficients of noise control filter W2, thereby obtaining noise control filter W1 with the identified filter coefficients and noise control filter W2 with the identified filter coefficients.

[0088] Such tuning stages are performed by sequentially changing the position of the noise source 200 to a plurality of reference positions 71. In this way, a plurality of noise control filters W1 associated one-to-one with the plurality of reference positions 71, a plurality of noise control filters W2 associated one-to-one with the plurality of reference positions 71, and a plurality of position discriminant functions G associated one-to-one with the plurality of reference positions 71 are obtained. The plurality of noise control filters W1 and the plurality of noise control filters W2 are stored in the filter storage unit 171. The plurality of position discriminant functions G are stored in the position information storage unit 172.

[0089] In the following explanation, *" indicates that it is an element obtained in the tuning stage. For example, the position discriminant function G obtained in the tuning stage is * The noise control filter W1 obtained in the tuning stage is sometimes referred to as the noise control filter W1 * The noise control filter W2 obtained in the tuning stage is sometimes referred to as the noise control filter W2 * It is sometimes written as:

[0090] In this embodiment, the plurality of noise control filters W1 obtained in the tuning stage * is the noise control filter W 1α * , noise control filter W 1β * , noise control filter W 1γ * and noise control filter W 1δ * The multiple noise control filters W2 obtained in the tuning stage * is the noise control filter W 2α * , noise control filter W 2β * , noise control filter W 2γ * and noise control filter W 2δ * The multiple position discriminant functions G obtained in the tuning stage * is the location discriminant function G α * , the position discriminant function G β * , the position discriminant function G γ * and the position discriminant function G δ * These are shown in Figures 8A and 8B, which will be described later. Note that the noise control filter W1 obtained in the tuning stage * The number of the filters is not particularly limited, and may be four, less than four, or more than four. In this respect, the noise control filter W2 obtained in the tuning stage *and the position discriminant function G obtained in the tuning stage * The same is true for the number of

[0091] The operation of the position estimator 160 and the memory 170 in the control stage will be described in more detail with reference to Figures 8A and 8B, which relate to the control of the first acoustic wave and the control of the second acoustic wave, respectively. Hereinafter, the reference signal x at time step n may be referred to as reference signal x(n). Also, the current time step n may be referred to as the time step n. c It is sometimes written as:

[0092] 8A and 8B, the control device 110 includes a selector 190. The selector 190 includes an error calculation unit 191, a minimum value determination unit 192, and a filter selection unit 193.

[0093] In the control stage, as shown in FIGS. 8A and 8B, the position estimation unit 160 calculates the position of the current time step n c The reference signal x(n c ) and the current time step n c A specific signal j(n c ) based on the current time step n of the noise source 200 c The position discriminant function G(n c In this embodiment, in the control stage as well as in the tuning stage, the position estimator 160 generates j(n c ) is corrected.

[0094] 8C, the correction unit 164 acquires the noise control filters W1 and W2 from the processor 180. The convolution unit 164a calculates x(n c ) to the noise control filter W1 and the secondary path filter S1 # and the signal obtained by convolution with x(n c ) to the noise control filter W2 and the secondary path filter S2 # By adding the signal obtained by convolution processing with cThe subtraction unit 164b generates j(n c ) to k(n c ) is subtracted. This gives j(n c ) is corrected, and j(n c ) the effects of the first and second sound waves are reduced or cancelled out.

[0095] In this way, the correction unit 164 performs convolution processing using a correction filter to obtain x(n c ) depending on j(n c ) is corrected. Then, the position estimation unit 160 calculates the reference signal x(n c ) and the corrected specific signal j(n c ) based on the location discriminant function G(n c )

[0096] The error calculation unit 191 calculates a plurality of position discriminant functions G * For each of these, the location discriminant function G * and the location discriminant function G(n c Specifically, the error calculation unit 191 calculates the proximity of a plurality of position discriminant functions G * For each of these, the location discriminant function G * and the location discriminant function G(n c ) and calculate the normalized mean square displacement (NMSD) between them.

number

[0097] The minimum value determination unit 192 determines the smallest closeness among the calculated closenesses. Specifically, the minimum value determination unit 192 determines the smallest NMSD among the calculated NMSDs.

[0098] The filter selection unit 193 selects the location discriminant function G that results in the smallest closeness. * Noise control filter W1 corresponding to * and noise control filter W2 *Specifically, the filter selection unit 193 selects the position discriminant function G that results in the minimum NMSD. * Noise control filter W1 corresponding to * and noise control filter W2 * Select .

[0099] The processor 180 synchronizes the control filter CF1 with the selected noise control filter W1 * Specifically, the processor 180 sets the selected noise control filter W1 * The filter coefficient of the control filter CF1 is substituted into the control filter CF1. As a result, the filter coefficient of the control filter CF1 is changed to the filter coefficient of the selected noise control filter W1. * The processor 180 calculates the filter coefficients of the set control filter CF1 and the current time step n. c The reference signal x(n c ) based on the current time step n c The control signal y1(n c ) to generate the control signal y1(n c ) is amplified by a first amplifier 111A, which is not shown in FIG. 8A. The first speaker 10A outputs the amplified control signal y1(n c ) and emits the first sound wave.

[0100] Processor 180 then applies control filter CF2 to the selected noise control filter W2 * Specifically, processor 180 sets the selected noise control filter W2 * The filter coefficient of the control filter CF2 * As a result, the filter coefficient of the control filter CF2 is set to the filter coefficient of the selected noise control filter W2. The processor 180 then performs a calculation based on the set control filter CF2 and the current time step n c The reference signal x(n c ) based on the current time step n c Control signal y2(n c ) to generate the control signal y2(n c) is amplified by a second amplifier 111B, which is not shown in FIG. 8B. The second speaker 10B outputs the amplified control signal y2(n c ) and emits a second sound wave.

[0101] In this way, in the control stage, a noise control filter W1 that is adapted to the position of the noise source 200 is * and noise control filter W2 * are selected and used to control the first and second sound waves. In this way, the first and second sound waves that are suited to the position of the noise source 200 are formed, and a state in which noise is satisfactorily reduced at the first control point 74A and the second control point 74B can be maintained.

[0102] In the control stage, a certain control filter CF1 is set by the selector 190, and then another control filter CF1 is set by the selector 190. In this embodiment, the certain control filter CF1 operates as a fixed filter until the other control filter CF1 is set. Here, the fixed filter is a filter whose filter coefficients are not updated.

[0103] Specific examples of signal processing block diagrams of the control device 110 are described with reference to Figures 9 and 10, which relate to the tuning stage and control stage, respectively. In the following description, "(z)" indicates a frequency domain representation. In Figure 10, dotted line 145 schematically indicates that the first error microphone 140A and the second error microphone 140B may be removed from the control stage.

[0104] In Figures 9 and 10, The reference signal x(n) is the detection signal of the reference microphone 130. The specific signal j(n) is the detection signal of the specific microphone 150 . The first error signal e1(n) is the detection signal of the first error microphone 140A. The second error signal e2(n) is the detection signal of the second error microphone 140B. ·W1(z) is a noise control filter that defines the response characteristic RC1 of the first sound wave to the reference signal x(n). ·W2(z) is a noise control filter that defines the response characteristic RC2 of the second sound wave to the reference signal x(n). The control signal y1(n) is a signal obtained by applying a noise control filter W1(z) to the reference signal x(n). The control signal y2(n) is a signal obtained by applying a noise control filter W2(z) to the reference signal x(n). P1(z) is a transfer function that represents the transfer characteristics of sound in the path from the output of the reference microphone 130 to the input of the first error microphone 140A. P2(z) is a transfer function that represents the transfer characteristics of sound in the path from the output of the reference microphone 130 to the input of the second error microphone 140B. ·S 11 (z) is a transfer function that represents the transfer characteristics of sound in the path from the output of noise control filter W1(z) to first error microphone 140A. ·S 21 (z) is a transfer function that represents the transfer characteristics of sound in the path from the output of noise control filter W1(z) to second error microphone 140B. ·S 12 (z) is a transfer function that represents the transfer characteristics of sound in the path from the output of noise control filter W2(z) to first error microphone 140A. ·S 22 (z) is a transfer function that represents the transfer characteristics of sound in the path from the output of noise control filter W2(z) to second error microphone 140B.

[0105] In Figure 9, ·r 11 (n) is the transfer function S 11 (z) is a filtered reference signal obtained by applying ·r 21(n) is the transfer function S 21 (z) is a filtered reference signal obtained by applying ·r 12 (n) is the transfer function S 12 (z) is a filtered reference signal obtained by applying ·r 22 (n) is the transfer function S 22 (z) is a filtered reference signal obtained by applying The update unit 135A uses the NLMS algorithm, the filtered reference signal r 11 (n) and the filtered reference signal r 21 The filter coefficients of the noise control filter W1(z) are updated based on (n). The update unit 135B uses the NLMS algorithm and the filtered reference signal r 12 (n) and the filtered reference signal r 22 The filter coefficients of the noise control filter W2(z) are updated based on (n). ·y' 11 (n) is the control signal y1(n) with the transfer function S 11 This is the signal obtained by applying (z). ·y' 21 (n) is the control signal y1(n) with the transfer function S 21 This is the signal obtained by applying (z). ·y' 12 (n) is the control signal y2(n) with the transfer function S 12 This is the signal obtained by applying (z). ·y' 22 (n) is the control signal y2(n) with the transfer function S 22 This is the signal obtained by applying (z). d1(n) is a signal obtained by applying a transfer function P1(z) to a reference signal x(n). ·d2(n) is a signal obtained by applying the transfer function P2(z) to the reference signal x(n).

[0106] In the tuning stage, with the noise source 200 placed at one of the reference positions 71, the signal processing block diagram of FIG. 9 is performed as follows: The filter coefficient of the noise control filter W1 is updated by the update unit 135A, The filter coefficient of the noise control filter W2 is updated by the update unit 135B, Noise control filter W1 with updated filter coefficients * is stored in the filter storage unit 171 of the memory 170, Noise control filter W2 with updated filter coefficients * is stored in the filter storage unit 171 of the memory 170, The stored position discriminant function G generated by the position estimation unit 160 * is stored in the position information storage unit 172 of the memory 170. The above updating and storing are performed by sequentially changing the position of the noise source 200 to a plurality of reference positions 71.

[0107] Then, in the control stage, according to the signal processing block diagram of FIG. 10, control signals y1(n) and y2(n) adapted to the position of the noise source 200 are generated, the first sound wave and the second sound wave are formed, and a state in which the noise is effectively reduced at the first control point 74A and the second control point 74B can be maintained.

[0108] The ANC system 500 according to this embodiment will be further described below.

[0109] The dimension of the space 86 in the up-down direction D1 is, for example, 40 cm or more and 300 cm or less, and may be 50 cm or more and 250 cm or less. The dimension of the space 86 in the left-right direction D2 is, for example, 20 cm or more and 500 cm or less, and may be 30 cm or more and 400 cm or less. The dimension of the space 86 in the front-rear direction D3 is, for example, 30 cm or more and 500 cm or less, and may be 40 cm or more and 400 cm or less.

[0110] The dimensions of the above-mentioned space 86 will now be described. Consider the smallest rectangular parallelepiped that encloses the space 86, having a first side extending in the up-down direction D1, a second side extending in the left-right direction D2, and a third side extending in the front-to-back direction D3. The dimension of the space 86 in the up-down direction D1 is the length of the first side. The dimension of the space 86 in the left-to-right direction D2 is the length of the second side. The dimension of the space 86 in the front-to-back direction D3 is the length of the third side.

[0111] As shown in FIG. 11A, the first support 90A has a first vertical direction DvA, a first horizontal direction DhA, and a first thickness direction DtA. The first vertical direction DvA, the first horizontal direction DhA, and the first thickness direction DtA are perpendicular to one another. The first support 90A has an upper end 90lA and a lower end 90mA on one side and the other side of the first vertical direction DvA, respectively. The first support 90A has a left end 90jA and a right end 90kA on one side and the other side of the first horizontal direction DhA, respectively. The first thickness direction DtA is the thickness direction of the first support 90A. In a typical example, the first vertical direction DvA is the same as the up-down direction D1.

[0112] The dimension of the first support 90A in the first vertical direction DvA is, for example, 25 cm or more and 200 cm or less, and may be 50 cm or more and 120 cm or less. The dimension of the first support 90A in the first horizontal direction DhA is, for example, 20 cm or more and 400 cm or less, and may be 20 cm or more and 200 cm or less. The dimension of the first thickness direction DtA of the first support 90A (i.e., the thickness of the first support 90A) is, for example, 0.1 cm or more and 15 cm or less. The dimension of the first vertical direction DvA of the first support 90A and the dimension of the first horizontal direction DhA of the first support 90A may be the same or different. In the illustrated example, the dimension of the first vertical direction DvA of the first support 90A is larger than the dimension of the first horizontal direction DhA of the first support 90A.

[0113] As shown in FIG. 11B, the second support 90B has a second vertical direction DvB, a second horizontal direction DhB, and a second thickness direction DtB. The second vertical direction DvB, the second horizontal direction DhB, and the second thickness direction DtB are perpendicular to one another. The second support 90B has an upper end 90lB and a lower end 90mB on one side and the other of the second vertical direction DvB, respectively. The second support 90B has a left end 90jB and a right end 90kB on one side and the other of the second horizontal direction DhB, respectively. The second thickness direction DtB is the thickness direction of the second support 90B. In a typical example, the second vertical direction DvB is the same as the up-down direction D1.

[0114] The dimension of the second support 90B in the second vertical direction DvB is, for example, 25 cm or more and 200 cm or less, and may be 50 cm or more and 120 cm or less. The dimension of the second support 90B in the second horizontal direction DhB is, for example, 20 cm or more and 400 cm or less, and may be 20 cm or more and 200 cm or less. The dimension of the second support 90B in the second thickness direction DtB (i.e., the thickness of the second support 90B) is, for example, 0.1 cm or more and 15 cm or less. The dimension of the second support 90B in the second vertical direction DvB and the dimension of the second horizontal direction DhB of the second support 90B may be the same or different. In the illustrated example, the dimension of the second support 90B in the second vertical direction DvB is larger than the dimension of the second horizontal direction DhB of the second support 90B.

[0115] The ratio of the sum of the dimension of the first horizontal direction DhA of the first support 90A and the dimension of the second horizontal direction DhB of the second support 90B to the dimension of the left-right direction D2 of the space 86 is, for example, 0.6 or more and 1.0 or less. This ratio may be 0.8 or more and 0.95 or less. The difference obtained by subtracting the sum of the dimension of the first horizontal direction DhA of the first support 90A and the dimension of the second horizontal direction DhB of the second support 90B from the dimension of the left-right direction D2 of the space 86 is, for example, 0 cm or more and 40 cm or less. This difference may be 5 cm or more and 20 cm or less.

[0116] As shown in FIG. 11A, the first radiation surface 15A has a longitudinal direction DlA and a lateral direction DsA. The first radiation surface 15A has an upper end 15lA and a lower end 15mA on one side and the other side of the longitudinal direction DlA, respectively. The first radiation surface 15A has a left end 15jA and a right end 15kA on one side and the other side of the lateral direction DsA, respectively. In this embodiment, the longitudinal direction DlA is the same as the first vertical direction DvA. The lateral direction DsA is the same as the first horizontal direction DhA. The shape of the first radiation surface 15A is rectangular.

[0117] However, as shown in FIG. 12A , the longitudinal direction DlA of the first radiation surface 15A may deviate from the first vertical direction DvA. The lateral direction DsA of the first radiation surface 15A may deviate from the first horizontal direction DhA. The deviation angle θp1 of the longitudinal direction DlA from the first vertical direction DvA may be, for example, 0° or more and 15° or less, or 0° or more and 5° or less. Similarly, the deviation angle θq1 of the lateral direction DsA from the first horizontal direction DhA may be, for example, 0° or more and 15° or less, or 0° or more and 5° or less. The shape of the first radiation surface 15A does not have to be rectangular. Another example of the first radiation surface 15A is a rectangle with rounded corners.

[0118] The dimension LxA of the first radiation surface 15A in the longitudinal direction DlA is, for example, 25 cm to 200 cm, and may be 50 cm to 120 cm. The dimension LyA of the first radiation surface 15A in the lateral direction DsA is, for example, 20 cm to 400 cm, and may be 20 cm to 200 cm.

[0119] The ratio of the dimension LxA to the dimension LyA, that is, the aspect ratio LxA / LyA, is, for example, 1.2 or more. The aspect ratio LxA / LyA may be 1.2 or more and 6 or less, or 1.5 or more and 4 or less.

[0120] As shown in FIG. 11B, the second radiation surface 15B has a longitudinal direction DlB and a lateral direction DsB. The second radiation surface 15B has an upper end 15lB and a lower end 15mB on one side and the other side of the longitudinal direction DlB, respectively. The second radiation surface 15B has a left end 15jB and a right end 15kB on one side and the other side of the lateral direction DsB, respectively. In this embodiment, the longitudinal direction DlB is the same as the second vertical direction DvB. The lateral direction DsB is the same as the second horizontal direction DhB. The shape of the second radiation surface 15B is rectangular.

[0121] However, as shown in FIG. 12B , the longitudinal direction DlB of the second radiation surface 15B may deviate from the second vertical direction DvB. The lateral direction DsB of the second radiation surface 15B may deviate from the second horizontal direction DhB. The deviation angle θp2 of the longitudinal direction DlB from the second vertical direction DvB is, for example, 0° or more and 15° or less, and may be 0° or more and 5° or less. Similarly, the deviation angle θq2 of the lateral direction DsB from the second horizontal direction DhB is, for example, 0° or more and 15° or less, and may be 0° or more and 5° or less. The shape of the second radiation surface 15B does not have to be rectangular. Another example of the second radiation surface 15B is a rectangle with rounded corners.

[0122] The dimension LxB of the second radiation surface 15B in the longitudinal direction DlB is, for example, 25 cm to 200 cm, and may be 50 cm to 120 cm. The dimension LyB of the second radiation surface 15B in the lateral direction DsB is, for example, 20 cm to 400 cm, and may be 20 cm to 200 cm.

[0123] The ratio of the dimension LxB to the dimension LyB, that is, the aspect ratio LxB / LyB, is, for example, 1.2 or more. The aspect ratio LxB / LyB ​​may be 1.2 or more and 6 or less, or 1.5 or more and 4 or less.

[0124] In an example where the first speaker 10A is a piezoelectric speaker, the first speaker 10A has a piezoelectric film. The longitudinal direction DlA of the first radiation surface 15A may be the piezoelectric expansion / contraction direction of the piezoelectric film.

[0125] In an example where the second speaker 10B is a piezoelectric speaker, the second speaker 10B has a piezoelectric film. The longitudinal direction D1B of the second radiation surface 15B may be the piezoelectric expansion / contraction direction of the piezoelectric film.

[0126] In this embodiment, the control device 110 is configured to output sounds in a first frequency range FR1 from the first speaker 10A and the second speaker 10B. The first frequency range FR1 is, for example, a range that includes a band from 100 Hz to 700 Hz. The first frequency range FR1 may be 50 Hz or more and 3000 Hz or less, or 100 Hz or more and 2000 Hz or less.

[0127] In one specific example, the control device 110 can set a second frequency range FR2. The control device 110 controls the frequencies of the sounds output from the first speaker 10A and the second speaker 10B to values ​​within the second frequency range FR2. The second frequency range FR2 is narrower than the first frequency range FR1. In reality, taking into consideration the size and calculation load of the ANC system 500, it may be desirable to utilize only a portion of the ANC system 500's capabilities rather than maximize its capabilities. This specific example can be used in such cases. Specifically, according to this specific example, a desired band can be selected as the second frequency range FR2.

[0128] As shown in FIG. 11A, in this embodiment, the first support 90A has an upper end 901A and a lower end 90mA that face each other. The first radiation surface 15A has an upper end 151A and a lower end 15mA that face each other. A first upper margin M1 between the upper end 901A and the upper end 151A is greater than or equal to zero and less than 1 / 10 of the reference wavelength. A first lower margin M2 between the lower end 90mA and the lower end 15mA is greater than or equal to zero and less than 1 / 10 of the reference wavelength. Here, the reference wavelength is the wavelength of the sound at the upper limit of the first frequency range FR1 or the second frequency range FR2. This arrangement is suitable for reducing diffracted sound generated when noise from the noise source 200 is diffracted at the upper end 901A and the lower end 90mA. Appropriate margin setting contributes to the reduction of diffracted sound, as described in Patent Documents 1 and 2.

[0129] In reality, there may be cases where the first upper margin M1 and the first lower margin M2 need to be increased to some extent for reasons of commercialization. Taking this into consideration, the upper limits of the first upper margin M1 and the first lower margin M2 may be greater than 1 / 10 of the reference wavelength. From the perspective of achieving reasonable commercialization while still achieving the effect of reducing diffracted sound, for example, the first upper margin M1 and the first lower margin M2 may be greater than zero and less than 1 / 3 of the reference wavelength. The first upper margin M1 and the first lower margin M2 may be, for example, greater than 0 cm and greater than 50 cm, or greater than 0 cm and less than 10 cm.

[0130] As shown in FIG. 11A , in this embodiment, the first support 90A has opposing left and right ends 90jA and 90kA. The first radiation surface 15A has opposing left and right ends 15jA and 15kA. A first left margin M3 between the left end 90jA and the left end 15jA is greater than or equal to zero and less than 1 / 10 of the reference wavelength. A first right margin M4 between the right end 90kA and the right end 15kA is greater than or equal to zero and less than 1 / 10 of the reference wavelength. The first left margin M3 and the first right margin M4 may be greater than or equal to zero and less than 1 / 3 of the reference wavelength. The first left margin M3 and the first right margin M4 may be, for example, greater than or equal to 0 cm and greater than 50 cm, or greater than or equal to 0 cm and less than 10 cm.

[0131] As shown in FIG. 12A, the longitudinal direction DlA of the first radiation surface 15A may be offset from the first vertical direction DvA, and the lateral direction DsA of the first radiation surface 15A may be offset from the first horizontal direction DhA. In such a case, the first top margin M1 is determined as the geometric mean value of the distance between the top end 90lA and the top end 15lA. The first bottom margin M2 is determined as the geometric mean value of the distance between the bottom end 90mA and the bottom end 15mA. The first left margin M3 is determined as the geometric mean value of the distance between the left end 90jA and the left end 15jA. The first right margin M4 is determined as the geometric mean value of the distance between the right end 90kA and the right end 15kA.

[0132] As shown in FIG. 11B, in this embodiment, the second support 90B has an upper end 901B and a lower end 90mB that face each other. The second radiation surface 15B has an upper end 151B and a lower end 15mB that face each other. A second upper margin M5 between the upper end 901B and the upper end 151B is greater than or equal to zero and less than 1 / 10 of the reference wavelength. A second lower margin M6 between the lower end 90mB and the lower end 15mB is greater than or equal to zero and less than 1 / 10 of the reference wavelength. Here, the reference wavelength is the wavelength of the sound at the upper limit of the first frequency range FR1 or the second frequency range FR2. This configuration is suitable for reducing diffracted sound generated when noise from the noise source 200 is diffracted at the upper end 901B and the lower end 90mB.

[0133] The upper limits of the second upper margin M5 and the second lower margin M6 may be greater than 1 / 10 of the reference wavelength. From the viewpoint of achieving reasonable commercialization while obtaining the effect of reducing diffracted sound, for example, the second upper margin M5 and the second lower margin M6 may be greater than or equal to zero and less than or equal to 1 / 3 of the reference wavelength. The second upper margin M5 and the second lower margin M6 may be, for example, greater than or equal to 0 cm and greater than or equal to 50 cm, or greater than or equal to 0 cm and less than or equal to 10 cm.

[0134] As shown in FIG. 11B , in this embodiment, the second support 90B has opposing left and right ends 90jB and 90kB. The second radiation surface 15B has opposing left and right ends 15jB and 15kB. A second left margin M7 between the left end 90jB and the left end 15jB is greater than or equal to zero and less than 1 / 10 of the reference wavelength. A second right margin M8 between the right end 90kB and the right end 15kB is greater than or equal to zero and less than 1 / 10 of the reference wavelength. The second left margin M7 and the second right margin M8 may be greater than or equal to zero and less than 1 / 3 of the reference wavelength. The second left margin M7 and the second right margin M8 may be, for example, greater than or equal to 0 cm and greater than 50 cm, or greater than or equal to 0 cm and less than 10 cm.

[0135] As shown in FIG. 12B, the longitudinal direction DlB of the second radiation surface 15B may be offset from the second vertical direction DvB, and the lateral direction DsB of the second radiation surface 15B may be offset from the second horizontal direction DhB. In such a case, the second top margin M5 is determined as the geometric mean value of the distance between the top end 90lB and the top end 15lB. The second bottom margin M6 is determined as the geometric mean value of the distance between the bottom end 90mB and the bottom end 15mB. The second left margin M7 is determined as the geometric mean value of the distance between the left end 90jB and the left end 15jB. The second right margin M8 is determined as the geometric mean value of the distance between the right end 90kB and the right end 15kB.

[0136] In a non-limiting example of this embodiment, the first speaker 10A is a piezoelectric speaker. The first radiation surface 15A has an upper end 151A and a lower end 15mA on one side and the other side in the longitudinal direction DIA, respectively. In this case, as understood from Patent Documents 1 and 2, the control device 110 can perform control to reverse the positive and negative phases between a central region of the first radiation surface 15A and regions on both sides of the central region (i.e., the upper region and the lower region) in the longitudinal direction DIA.

[0137] In the above non-limiting example, the accuracy of ANC by the ANC system 500 can be improved by reducing the absolute value of the difference between the first upper margin M1 and the first right margin M4 and by reducing the absolute value of the difference between the first lower margin M2 and the first right margin M4. Specifically, the sound waves emitted from the central region, the upper region, and the lower region of the first radiation surface 15A are guided around the first support 90A in the space 86 via the right end 90kA, the upper end 90lA, and the lower end 90mA of the first support 90A, respectively. When both absolute values ​​of these two are small, the phase difference between the sound waves emitted from the central region, the upper region, and the lower region is generally maintained during this guidance, and the phases of these sound waves are maintained in a matched or mismatched state. These sound waves that have been deflected propagate deeper into the space 86, away from the first support 90A. During this propagation, these sound waves cancel each other out. This cancellation reduces the amount of sound that is deflected from the first speaker 10A to the reference microphone 130. This makes it less likely that the sound emitted by the first speaker 10A will act as noise in ANC.

[0138] In a non-limiting example of this embodiment, the second speaker 10B is a piezoelectric speaker. The second radiation surface 15B has an upper end 151B and a lower end 15mB on one side and the other side in the longitudinal direction D1B, respectively. In this case, the control device 110 can perform control to reverse the positive and negative phases between a central region of the second radiation surface 15B and regions on both sides of the central region (i.e., the upper region and the lower region) in the longitudinal direction D1B.

[0139] In the above non-limiting example, the accuracy of ANC by the ANC system 500 can be improved by reducing the absolute value of the difference between the second upper margin M5 and the second left margin M7 and by reducing the absolute value of the difference between the second lower margin M6 and the second left margin M7. Specifically, the sound waves emitted from the central region, the upper region, and the lower region of the second radiation surface 15B are guided around the left end 90jB, the upper end 90lB, and the lower end 90mB of the second support 90B, respectively, and are guided around the vicinity of the second support 90B in the space 86. When both absolute values ​​of the above two are small, the phase difference between the sound waves emitted from the central region, the upper region, and the lower region is generally maintained during this guidance, and the relationship of agreement or disagreement between the positive and negative phases of these sound waves is maintained. These sound waves that have been deflected propagate deeper into the space 86, away from the second support 90B. During this propagation, these sound waves cancel each other out. This cancellation reduces the amount of sound that is deflected from the second speaker 10B to the reference microphone 130. This makes it less likely that the sound emitted by the second speaker 10B will act as noise in ANC.

[0140] The absolute value of the difference between the first top margin M1 and the first right margin M4 may be, for example, 1 / 8 or less of the reference wavelength, or 1 / 16 or less of the reference wavelength. The absolute value of the difference between the first bottom margin M2 and the first right margin M4 may be, for example, 1 / 8 or less of the reference wavelength, or 1 / 16 or less of the reference wavelength. The absolute value of the difference between the second top margin M5 and the second left margin M7 may be, for example, 1 / 8 or less of the reference wavelength, or 1 / 16 or less of the reference wavelength. The absolute value of the difference between the second bottom margin M6 and the second left margin M7 may be, for example, 1 / 8 or less of the reference wavelength, or 1 / 16 or less of the reference wavelength.

[0141] The absolute value of the difference between the first top margin M1 and the first right margin M4 may be, for example, 1 cm or more and 86 cm or less, or may be 1 cm or more and 43 cm or less. The absolute value of the difference between the first bottom margin M2 and the first right margin M4 may be, for example, 1 cm or more and 86 cm or less, or may be 1 cm or more and 43 cm or less. The absolute value of the difference between the second top margin M5 and the second left margin M7 may be, for example, 1 cm or more and 86 cm or less, or may be 1 cm or more and 43 cm or less. The absolute value of the difference between the second bottom margin M6 and the second left margin M7 may be, for example, 1 cm or more and 86 cm or less, or may be 1 cm or more and 43 cm or less.

[0142] Furthermore, when the first speaker 10A is a piezoelectric speaker, the vibration of the first speaker 10A when it emits sound is less likely to vibrate the first support 90A. Therefore, it is less likely that unwanted sound will be input from the first support 90A to the reference microphone 130 due to the vibration of the first support 90A. Similarly, when the second speaker 10B is a piezoelectric speaker, the vibration of the second speaker 10B when it emits sound is less likely to vibrate the second support 90B. Therefore, it is less likely that unwanted sound will be input from the second support 90B to the reference microphone 130 due to the vibration of the second support 90B. These factors also make it less likely that noise originating from the first speaker 10A and / or the second speaker 10B will be generated in the ANC.

[0143] Below, a configuration example of the first speaker 10A, which is a piezoelectric speaker, and a configuration example of the second speaker 10B, which is also a piezoelectric speaker, will be described. The configuration of the "piezoelectric speaker 10" in the following description can be used as the configuration of the first speaker 10A or the configuration of the second speaker 10B. As explained in Patent Documents 1 and 2, the "piezoelectric speaker 10" in the following description has good characteristics. These good characteristics can also be useful when the piezoelectric speaker 10 is incorporated into an ANC system 500. The same applies to the "piezoelectric speaker 310" in the following description.

[0144] [First Configuration Example of Piezoelectric Speaker 10] A piezoelectric speaker 10 according to a first configuration example will be described with reference to FIGS. 13A and 13B.

[0145] The piezoelectric speaker 10 includes a piezoelectric film 35, a first bonding layer 51, an intervening layer 40, and a second bonding layer 52. The first bonding layer 51, the intervening layer 40, the second bonding layer 52, and the piezoelectric film 35 are laminated in this order.

[0146] The piezoelectric film 35 includes a piezoelectric body 30 , a first electrode 61 , and a second electrode 62 .

[0147] The piezoelectric body 30 has a film shape. The piezoelectric body 30 vibrates when a voltage is applied. The piezoelectric body 30 can be made of a ceramic film, a resin film, or the like. Examples of materials for the ceramic film piezoelectric body 30 include lead zirconate, lead zirconate titanate, lead lanthanum zirconate titanate, barium titanate, Bi layered compounds, tungsten bronze structure compounds, and solid solutions of barium titanate and bismuth ferrite. Examples of materials for the resin film piezoelectric body 30 include polyvinylidene fluoride and polylactic acid. The resin film piezoelectric body 30 can also be made of polyolefins such as polyethylene and polypropylene. The piezoelectric body 30 can be either non-porous or porous.

[0148] The thickness of the piezoelectric body 30 is, for example, 10 μm or more and 300 μm or less, and may be 30 μm or more and 110 μm or less.

[0149] The first electrode 61 and the second electrode 62 are in contact with the piezoelectric body 30 so as to sandwich the piezoelectric body 30. The first electrode 61 and the second electrode 62 have a film shape. The first electrode 61 and the second electrode 62 are each connected to a lead wire (not shown). The first electrode 61 and the second electrode 62 can be formed on the piezoelectric body 30 by vapor deposition, plating, sputtering, or the like. Metal foils can also be used as the first electrode 61 and the second electrode 62. The metal foils can be attached to the piezoelectric body 30 using double-sided tape, pressure-sensitive adhesives, adhesives, or the like. Materials for the first electrode 61 and the second electrode 62 include metals, such as gold, platinum, silver, copper, palladium, chromium, molybdenum, iron, tin, aluminum, and nickel. Materials for the first electrode 61 and the second electrode 62 include carbon, conductive polymers, and the like. Materials for the first electrode 61 and the second electrode 62 also include alloys of these. The first electrode 61 and the second electrode 62 may contain a glass component, etc.

[0150] The thickness of each of the first electrode 61 and the second electrode 62 is, for example, not less than 10 nm and not more than 150 μm, and may be not less than 20 nm and not more than 100 μm.

[0151] 13A and 13B, the first electrode 61 covers the entirety of one main surface of the piezoelectric body 30. However, the first electrode 61 may cover only a portion of the one main surface of the piezoelectric body 30. The second electrode 62 covers the entirety of the other main surface of the piezoelectric body 30. However, the second electrode 62 may cover only a portion of the other main surface of the piezoelectric body 30.

[0152] In the first configuration example, the intervening layer 40 is disposed between the piezoelectric film 35 and the first bonding layer 51. The intervening layer 40 may be a layer other than an adhesive layer or a pressure-sensitive adhesive layer, and may be an adhesive layer or a pressure-sensitive adhesive layer. In the first configuration example, the intervening layer 40 is a porous layer and / or a resin layer. Here, the resin layer is a concept that includes a rubber layer and an elastomer layer, and therefore the intervening layer 40, which is a resin layer, may be a rubber layer or an elastomer layer. Examples of the intervening layer 40 that is a resin layer include an ethylene propylene rubber layer, a butyl rubber layer, a nitrile rubber layer, a natural rubber layer, a styrene butadiene rubber layer, a silicone layer, a urethane layer, and an acrylic resin layer. Examples of the intervening layer 40 that is a porous layer include a foam layer. Specifically, examples of the intervening layer 40 that is a porous layer and a resin layer include an ethylene propylene rubber foam layer, a butyl rubber foam layer, a nitrile rubber foam layer, a natural rubber foam layer, a styrene butadiene rubber foam layer, a silicone foam layer, and a urethane foam layer. Examples of the intervening layer 40 that is a resin layer but not a porous layer include an acrylic resin layer. Examples of the intervening layer 40 that is a porous layer but not a resin layer include a metal porous layer. Here, the term "resin layer" refers to a layer containing resin, and may contain 30% or more of resin, 45% or more of resin, 60% or more of resin, or 80% or more of resin. The same applies to rubber layers, elastomer layers, ethylene propylene rubber layers, butyl rubber layers, nitrile rubber layers, natural rubber layers, styrene butadiene rubber layers, silicone layers, urethane layers, acrylic resin layers, metal layers, etc. The same applies to resin films, ceramic films, etc. that can be used as the piezoelectric element 30. The intervening layer 40 may also be a blend layer of two or more materials.

[0153] The modulus of elasticity of the intermediate layer 40 is, for example, 10,000 N / m 2 More than 20000000N / m 2 Less than 20000N / m 2 More than 100000N / m 2 It may be the following:

[0154] In one example, the pore size of the porous intermediate layer 40 is 0.1 mm to 7.0 mm, or may be 0.3 mm to 5.0 mm. In another example, the pore size of the porous intermediate layer 40 is, for example, 0.1 mm to 2.5 mm, or may be 0.2 mm to 1.5 mm, or may be 0.3 mm to 0.7 mm. The porosity of the porous intermediate layer 40 is, for example, 70% to 99%, or may be 80% to 99%, or may be 90% to 95%.

[0155] A known foam can be used for the intervening layer 40 (for example, the foam disclosed in Patent Document 3 can be used). The intervening layer 40, which is a foam layer, may have an open-cell structure, a closed-cell structure, or a semi-closed semi-open-cell structure. An open-cell structure refers to a structure in which the open cell ratio is 100%. A closed-cell structure refers to a structure in which the open cell ratio is 0%. A semi-closed semi-open-cell structure refers to a structure in which the open cell ratio is greater than 0% and less than 100%. Here, the open cell ratio can be calculated, for example, by conducting a test in which the foam layer is submerged in water and using the formula: open cell ratio (%) = {(volume of absorbed water) / (volume of air bubble portion)} × 100. In one specific example, the "volume of absorbed water" is measured by submerging the foam layer in water and leaving it under a reduced pressure of -750 mmHg for 3 minutes, then measuring the mass of water that has replaced the air in the bubbles of the foam layer, and determining that the density of water is 1.0 g / cm. 3 The "bubble partial volume" can be calculated by the formula: bubble partial volume (cm 3 ) = {(mass of foam layer) / (apparent density of foam layer)} - {(mass of foam layer) / (material density)}. "Material density" is the density of the base material (solid body) that forms the foam layer.

[0156] The foaming ratio (density ratio before and after foaming) of the intervening layer 40, which is a foam layer, is, for example, 5 times or more and 40 times or less, and may be 10 times or more and 40 times or less.

[0157] The thickness of the intervening layer 40 in the uncompressed state is, for example, 0.1 mm to 30 mm, or alternatively, 1 mm to 30 mm, 1.5 mm to 30 mm, or alternatively, 2 mm to 25 mm. Typically, the intervening layer 40 is thicker than the piezoelectric film 35 in the uncompressed state. In the uncompressed state, the ratio of the thickness of the intervening layer 40 to the thickness of the piezoelectric film 35 is, for example, 3 times or more, 10 times or more, or even 30 times or more. Also, typically, the intervening layer 40 is thicker than the first bonding layer 51 in the uncompressed state.

[0158] The surface of the first bonding layer 51 forms the fixing surface 17. The first bonding layer 51 is a layer that is bonded to the first support 90A and / or the second support 90B. In the example of FIG. 13A , the first bonding layer 51 is bonded to the intermediate layer 40.

[0159] In the first configuration example, the first bonding layer 51 is an adhesive or sticky layer. In other words, the first bonding layer 51 is an adhesive or sticky layer. The fixing surface 17 is an adhesive or sticky surface. The first bonding layer 51 can be attached to the first support 90A and / or the second support 90B. In the example of FIG. 13A, the first bonding layer 51 is in contact with the intermediate layer 40.

[0160] The first bonding layer 51 may be a double-sided tape having a substrate and an adhesive applied to both sides of the substrate. The substrate of the double-sided tape used as the first bonding layer 51 may be a nonwoven fabric or the like. The adhesive of the double-sided tape used as the first bonding layer 51 may be an adhesive containing an acrylic resin or the like. However, the first bonding layer 51 may also be an adhesive layer without a substrate.

[0161] The thickness of the first bonding layer 51 is, for example, not less than 0.01 mm and not more than 1.0 mm, and may be not less than 0.05 mm and not more than 0.5 mm.

[0162] The second bonding layer 52 is disposed between the intervening layer 40 and the piezoelectric film 35. In the first configuration example, the second bonding layer 52 is an adhesive or sticky layer. In other words, the second bonding layer 52 is an adhesive or sticky layer. Specifically, the second bonding layer 52 is bonded to the intervening layer 40 and the piezoelectric film 35.

[0163] The second bonding layer 52 may be a double-sided tape having a substrate and an adhesive applied to both sides of the substrate. The substrate of the double-sided tape used as the second bonding layer 52 may be a nonwoven fabric or the like. The adhesive of the double-sided tape used as the second bonding layer 52 may be an adhesive containing an acrylic resin or the like. However, the second bonding layer 52 may also be an adhesive layer without a substrate.

[0164] The thickness of the second bonding layer 52 is, for example, not less than 0.01 mm and not more than 1.0 mm, and may be not less than 0.05 mm and not more than 0.5 mm.

[0165] In the first configuration example, the adhesive surface or sticky surface comes into contact with the piezoelectric film 35, thereby integrating the piezoelectric film 35 with the layer on the fixing surface 17 side. Specifically, in the first configuration example, the adhesive surface or sticky surface is a surface formed by the surface of the second bonding layer 52.

[0166] An ANC system 500 can be configured using the piezoelectric speaker 10 according to the first configuration example as the first speaker 10A and the second speaker 10B. Compared to a dynamic speaker, the piezoelectric speaker 10 has a shorter time (hereinafter sometimes referred to as delay time) from when an electrical signal reaches the speaker 10 until sound is emitted. For this reason, the piezoelectric speaker 10 is suitable for configuring a small ANC system not only because of its small size but also because the distance between the reference microphone 130 and the piezoelectric speaker 10 can be shortened.

[0167] With the piezoelectric speaker 10 fixed to the first support 90A and / or the second support 90B, a voltage is applied to the piezoelectric film 35 via the lead wires, causing the piezoelectric film 35 to vibrate and radiate sound waves.

[0168] The piezoelectric speaker 10 and the ANC system 500 to which the piezoelectric speaker 10 is applied will be further described.

[0169] The piezoelectric speaker 10 can be fixed to the first support 90A and / or the second support 90B by the fixing surface 17. In this manner, an ANC system 500 can be configured using the piezoelectric speaker 10. In the ANC system 500, the interposing layer 40 is disposed between the piezoelectric film 35 and the first support 90A and / or the second support 90B.

[0170] Although the details of the effect remain to be studied, it is possible that by appropriately constraining one of the principal surfaces of the piezoelectric film 35 with the intervening layer 40, the piezoelectric film 35 is more likely to generate low-frequency sounds in the audible range. Taking this into consideration, the intervening layer 40 can be arranged to occupy 25% or more of the area of ​​the piezoelectric film 35 when observed in a planar view. When observed in a planar view, the intervening layer 40 may be arranged to occupy 50% or more of the area of ​​the piezoelectric film 35, or 75% or more of the area of ​​the piezoelectric film 35, or the intervening layer 40 may be arranged to occupy the entire area of ​​the piezoelectric film 35. Furthermore, 50% or more of the principal surface 38 of the piezoelectric speaker 10 opposite the fixing surface 17 may be made of the piezoelectric film 35. 75% or more of the principal surface 38 may be made of the piezoelectric film 35, or the entire principal surface 38 may be made of the piezoelectric film 35.

[0171] In the first configuration example, the second bonding layer 52 prevents separation between the piezoelectric film 35 and the intervening layer 40. From the viewpoint of the above-mentioned "appropriate constraint," the second bonding layer 52 and the intervening layer 40 can be arranged over 25% or more of the area of ​​the piezoelectric film 35 when observed in a planar view. When observed in a planar view, the second bonding layer 52 and the intervening layer 40 may be arranged over 50% or more of the area of ​​the piezoelectric film 35, or over 75% or more of the area of ​​the piezoelectric film 35, or the second bonding layer 52 and the intervening layer 40 may be arranged over the entire area of ​​the piezoelectric film 35.

[0172] Here, when the intervening layer 40 is porous, the ratio of the area where the intervening layer 40 is disposed is determined from a more macroscopic perspective, rather than a microscopic perspective that takes into account the pores resulting from the porous structure. For example, when the piezoelectric film 35, the porous intervening layer 40, and the second bonding layer 52 are plate-like bodies that have a common outline in a planar view, it is expressed that the second bonding layer 52 and the intervening layer 40 are disposed in an area that occupies 100% of the area of ​​the piezoelectric film 35.

[0173] In the first configuration example, the constraint degree of the intermediate layer 40 is 5×10 9 N / m 3 The degree of constraint of the intermediate layer 40 is, for example, 1×10 4 N / m 3 The degree of constraint of the intermediate layer 40 is preferably 5×10 8 N / m 3 More preferably, it is 2×10 or less. 8 N / m 3 or less, and more preferably 1×10 5 5x10 or more 7 N / m 3 Here, the constraint degree (N / m 3 ) is calculated by the following formula: 2) and the surface filling rate of the intervening layer 40, divided by the thickness (m) of the intervening layer 40. The surface filling rate of the intervening layer 40 is the filling rate (1 minus the porosity) of the main surface of the intervening layer 40 facing the piezoelectric film 35. If the pores in the intervening layer 40 are evenly distributed, the surface filling rate can be considered to be equal to the three-dimensional filling rate of the intervening layer 40. Restraint degree (N / m 3 ) = Elastic modulus (N / m 2 ) × surface filling rate ÷ thickness (m)

[0174] The degree of constraint can be considered a parameter that represents the degree of constraint of the piezoelectric film 35 by the intervening layer 40. The above formula shows that the greater the elastic modulus of the intervening layer 40, the greater the degree of constraint. The above formula also shows that the greater the surface filling rate of the intervening layer 40, the greater the degree of constraint. The above formula also shows that the thinner the intervening layer 40, the greater the degree of constraint. Although the relationship between the degree of constraint of the intervening layer 40 and the sound generated by the piezoelectric film 35 needs further study, if the degree of constraint is excessively large, it is possible that the deformation of the piezoelectric film 35 necessary to generate low-frequency sound is hindered. Conversely, if the degree of constraint is excessively small, the piezoelectric film 35 does not deform sufficiently in its thickness direction, and expands and contracts only in its in-plane direction (direction perpendicular to the thickness direction), which may prevent the generation of low-frequency sound. By setting the degree of constraint of the intervening layer 40 within an appropriate range, the expansion and contraction of the piezoelectric film 35 in the in-plane direction is appropriately converted into deformation in the thickness direction, causing the piezoelectric film 35 to bend appropriately as a whole, making it easier to generate low-frequency sounds.

[0175] As can be understood from the above description, a layer other than the intervening layer 40 may be present between the piezoelectric film 35 and the fixing surface 17. The different layer is, for example, the second bonding layer 52.

[0176] In the illustrated example, the piezoelectric film 35 is not completely surrounded by the intervening layer 40. In the illustrated example, there is an imaginary straight line that passes through the intervening layer 40 and the piezoelectric film 35 in that order and then reaches the outside of the piezoelectric speaker 10 without passing through the intervening layer 40. Here, "there is an imaginary straight line" means that such a straight line can be drawn. In the illustrated example, the intervening layer 40 extends only toward the fixed surface 17 side when viewed from the piezoelectric film 35.

[0177] In the illustrated example, the principal surface 38 of the piezoelectric film 35 opposite the fixed surface 17 constitutes the radiation surface 15. In other words, the principal surface 38 of the piezoelectric film 35 opposite the intervening layer 40 constitutes the radiation surface 15. In this configuration, the principal surface of the piezoelectric film 35 facing the intervening layer 40 is constrained by the intervening layer 40, so that expansion and contraction in the in-plane direction of the piezoelectric film 35 can be appropriately converted into deformation in the thickness direction. However, other configurations can also be adopted.

[0178] Specifically, a first layer may be provided on the piezoelectric film 35 on the side opposite to the interposed layer 40. For example, the first layer is used to protect the piezoelectric film 35. In this case, the main surface of the first layer may form the radiation surface 15. Alternatively, a second layer separate from the first layer may form the radiation surface 15.

[0179] The thickness of the first layer is, for example, 0.05 mm or more and 5 mm or less. The material of the first layer is, for example, a polyester-based material. Here, polyester-based material refers to a material containing polyester, and may contain 30% or more of polyester, 45% or more of polyester, 60% or more of polyester, or 80% or more of polyester. In one example, the material of the intervening layer 40 is different from the material of the first layer. When the material of the intervening layer 40 is different from the material of the first layer, it is possible to create a difference in the degree of constraint between the main surface of the piezoelectric film 35 facing the intervening layer 40 and the main surface of the piezoelectric film 35 facing the first layer. This can make it possible to appropriately convert expansion and contraction in the in-plane direction of the piezoelectric film 35 into deformation in the thickness direction. The degree of constraint of the intervening layer 40 and the degree of constraint of the first layer may be different. Here, the degree of constraint of the first layer (N / m 3 ) is the elastic modulus of the first layer (N / m 2 The surface filling ratio of the first layer is the filling ratio (1 minus the porosity) of the main surface of the first layer facing the piezoelectric film 35. The difference between the degree of constraint of the intervening layer 40 and the degree of constraint of the first layer can appropriately convert the expansion and contraction of the piezoelectric film 35 in the in-plane direction into deformation in the thickness direction. In one specific example, the degree of constraint of the intervening layer 40 is greater than the degree of constraint of the first layer. The first layer may have a film shape. The first layer may be a nonwoven fabric.

[0180] In the first configuration example, the fixing surface 17 is arranged so that, when the piezoelectric film 35 is observed in a plan view, at least a portion of the piezoelectric film 35 overlaps with the fixing surface 17 (overlapping with the first bonding layer 51 in the example of FIG. 13A ). From the viewpoint of stably fixing the piezoelectric speaker 10 to the first support 90A and / or the second support 90B, the fixing surface 17 can be arranged over 50% or more of the area of ​​the piezoelectric film 35 when the piezoelectric film 35 is observed in a plan view. The fixing surface 17 may also be arranged over 75% or more of the area of ​​the piezoelectric film 35 when the piezoelectric film 35 is observed in a plan view, or the fixing surface 17 may be arranged over the entire area of ​​the piezoelectric film 35.

[0181] In the first configuration example, the piezoelectric film 35 and the intervening layer 40 each have a substantially constant thickness. This is often advantageous from various perspectives, such as storage of the piezoelectric speaker 10, ease of use, and control of the sound emitted from the piezoelectric film 35. Note that "substantially constant thickness" means, for example, that the minimum thickness is between 70% and 100% of the maximum thickness. The minimum thickness of the piezoelectric film 35 and the intervening layer 40 may each be between 85% and 100% of the maximum thickness.

[0182] Resin is a material that is less susceptible to cracking than ceramics and the like. In one specific example, the piezoelectric body 30 of the piezoelectric film 35 is a resin film, and the intervening layer 40 is a resin layer that does not function as a piezoelectric film. This is advantageous from the perspective of being able to cut the piezoelectric speaker 10 with scissors, by hand, or the like without causing cracks in the piezoelectric body 30 or the intervening layer 40 (the fact that the piezoelectric speaker 10 can be cut with scissors, by hand, or the like makes it easier to construct the ANC system 500). This also makes it less likely that cracks will occur in the piezoelectric body 30 or the intervening layer 40 even when the piezoelectric speaker 10 is bent. Having the piezoelectric body 30 be a resin film and the intervening layer 40 be a resin layer is also advantageous from the perspective of fixing the piezoelectric speaker 10 to a curved surface without causing cracks in the piezoelectric body 30 or the intervening layer 40.

[0183] 13A, the contours of the piezoelectric film 35, the intervening layer 40, the first bonding layer 51, and the second bonding layer 52 are the same in plan view, although it does not matter if these contours are misaligned.

[0184] 13A, the piezoelectric film 35, the interposed layer 40, the first bonding layer 51, and the second bonding layer 52 are rectangular in shape having a short side direction and a long side direction in a plan view, but may also be square, circular, elliptical, or the like.

[0185] Furthermore, the piezoelectric speaker 10 may include layers other than the layers shown in Fig. 13A. The layers other than the layers shown in Fig. 13A are, for example, the first and second layers described above.

[0186] [Second Configuration Example of Piezoelectric Speaker 10] Hereinafter, a piezoelectric speaker 310 according to the second configuration example will be described with reference to Fig. 14. In the following, descriptions of parts that are the same as those in the first configuration example may be omitted.

[0187] The piezoelectric speaker 310 includes a piezoelectric film 35, a fixing surface 117, and an intervening layer 140. The fixing surface 117 can be used to fix the piezoelectric film 35 to the first support 90A and / or the second support 90B.

[0188] The intervening layer 140 is disposed between the piezoelectric film 35 and the fixed surface 117 (here, "between" includes the fixed surface 117. The same applies to the first configuration example). The fixed surface 117 is formed by the surface (main surface) of the intervening layer 140.

[0189] The intervening layer 140 is a porous layer and / or a resin layer. The intervening layer 140 is an adhesive layer or a bonding layer. An adhesive containing an acrylic resin can be used as the intervening layer 140. Other adhesives, such as adhesives containing rubber, silicone, or urethane, may also be used as the intervening layer 140. The intervening layer 140 may also be a blend layer of two or more materials.

[0190] The modulus of elasticity of the intermediate layer 140 is, for example, 10,000 N / m2 More than 20000000N / m 2 Less than 20000N / m 2 More than 100000N / m 2 It may be the following:

[0191] The thickness of the intervening layer 140 in the uncompressed state is, for example, 0.1 mm to 30 mm, or alternatively, 1 mm to 30 mm, 1.5 mm to 30 mm, or 2 mm to 25 mm. Typically, the intervening layer 140 is thicker than the piezoelectric film 35 in the uncompressed state. In the uncompressed state, the ratio of the thickness of the intervening layer 140 to the thickness of the piezoelectric film 35 is, for example, 3 times or more, or alternatively, 10 times or more, or even 30 times or more.

[0192] In the second configuration example, the constraint degree of the intermediate layer 140 is 5×10 9 N / m 3 The degree of constraint of the intermediate layer 140 is, for example, 1×10 4 N / m 3 The degree of constraint of the intermediate layer 140 is preferably 5×10 8 N / m 3 More preferably, it is 2×10 or less. 8 N / m 3 or less, and more preferably 1×10 5 5x10 or more 7 N / m 3 The definition of the degree of constraint is as explained above.

[0193] In the second configuration example, the adhesive or sticky surface comes into contact with the piezoelectric film 35, thereby integrating the piezoelectric film 35 with the layer on the fixing surface 117 side. Specifically, in the second configuration example, the adhesive or sticky surface is a surface formed by the intervening layer 140.

[0194] The piezoelectric speaker 310 can also be fixed to the first support 90A and / or the second support 90B by the fixing surface 117. In this way, an ANC system 500 can be configured using the piezoelectric speaker 310 according to the second configuration example as the first speaker 10A and the second speaker 10B.

[0195] An ANC system 500 may be configured by using the piezoelectric speaker 10 according to the first configuration example as one of the first speaker 10A and the second speaker 10B, and using the piezoelectric speaker 310 according to the second configuration example as the other of the first speaker 10A and the second speaker 10B.

[0196] As can be understood from the first and second configuration examples, the piezoelectric speaker 10 and the piezoelectric speaker 310 can be in a film shape. Hereinafter, the film-shaped piezoelectric speaker 10 can be referred to as a film speaker 10. The film-shaped piezoelectric speaker 310 can be referred to as a film speaker 310.

[0197] This embodiment will be further described below. The description regarding the first channel of the ANC system 500 is applicable to the second channel of the ANC system 500. The description regarding the second channel is applicable to the first channel. The first channel includes a first error microphone 140A, a noise control filter W1, a first amplifier 111A, and a first speaker 10A. The second channel includes a second error microphone 140B, a noise control filter W2, a second amplifier 111B, and a second speaker 10B. The first and second channels share a reference microphone 130 and a specific microphone 150.

[0198] As can be understood from the above description, in this embodiment, the control device 110 determines the control filter CF1 based on the reference signal x and the specific signal j. The control filter CF1 defines the response characteristic RC1 of the first sound wave with respect to the reference signal x. The control device 110 realizes ANC by radiating the first sound wave from the first speaker 10A based on the reference signal x and the control filter CF1. This configuration is suitable for realizing ANC that is robust with respect to the position of the noise source 200. The reference signal x and the specific signal j may be vectors.

[0199] In the above configuration, the reference signal x may be a signal generated by the reference microphone 130, and the specific signal j may be a signal generated by the specific microphone 150. In the above configuration, because the positions of the two microphones 130 and 150 are different, when the noise source 200 moves, the way in which the reference signal x changes and the way in which the specific signal j changes differ. For this reason, the relationship between the reference signal x and the specific signal j is useful as information about the position of the noise source 200.

[0200] The control device 110 may determine the control filter CF1 based on a relationship, such as the ratio between the reference signal x and the specific signal j. In this embodiment, the control device 110 determines the control filter CF1 based on a function of the reference signal x and the specific signal j. Specifically, the function represents the ratio between the reference signal x and the specific signal j. The ratio between the reference signal x and the specific signal j is specifically the ratio between the level of the reference signal x and the level of the specific signal j. The function and the ratio may be vectors. The function may be a position discriminant function G.

[0201] In this embodiment, the determination of the control filter CF1 in the control stage after the tuning stage is not a process of converging the filter coefficients of the control filter CF1 using an adaptive algorithm. This configuration is advantageous from the perspective of ensuring a fast response time of the control in response to the movement of the noise source 200. This configuration is also advantageous from the perspective of avoiding divergence in the control stage. This can contribute to the stability of the control. Specifically, consider an ANC (hereinafter referred to as a reference ANC) in which the filter coefficients are sequentially updated using an adaptive algorithm in the control stage. In the reference ANC, control can become unstable and diverge when an abnormal or unexpected sound is input to the microphone. Specifically, in the reference ANC, the filter coefficients can diverge infinitely, resulting in a loud sound being output from the control speaker. In contrast, the above configuration of this embodiment can avoid such divergence in the control stage. Specifically, the determination of the control filter CF1 based on the selection of the response characteristic filter RF1 (described later) can be a process of switching the filter coefficients. Switching the filter coefficients is less likely to cause divergence in the control.

[0202] In this embodiment, the control filter CF1 associates the reference signal x with the first sound wave one-to-one. This configuration allows for simple ANC, which is advantageous from the viewpoint of ensuring a fast response time of control to the movement of the noise source 200. Specifically, this configuration allows for ANC with a smaller number of filters than techniques that use auxiliary filters in virtual sensing, etc., which is advantageous from the viewpoint of achieving good ANC.

[0203] In this embodiment, the determination of the control filter CF1 can be repeated at a predetermined cycle while the noise source 200 is moving. With this configuration, the control filter CF1 can be switched in real time in accordance with the movement of the noise source 200. This allows the control of the first sound wave to follow the movement of the noise source 200 in real time.

[0204] In this embodiment, the control device 110 includes a memory 170, a selector 190, and a processor 180. The memory 170 stores a plurality of response characteristic filters RF1 that define different response characteristics RC1. The selector 190 selects A response characteristic filters RF1 from the plurality of response characteristic filters RF1 based on a reference signal x and a specific signal j, where A is a natural number equal to or greater than 1. The processor 180 implements ANC using a control filter CF1 based on the A response characteristic filters RF1. This configuration makes it possible to determine the control filter CF1 depending on the position of the noise source 200. Specifically, in this embodiment, A is 1. However, A may also be a natural number equal to or greater than 2.

[0205] In this embodiment, the plurality of response characteristic filters RF1 are associated one-to-one with the plurality of corresponding values. Each of the plurality of response characteristic filters RF1 is capable of realizing ANC when the function of the reference signal x and the specific signal j takes the corresponding value corresponding to that response characteristic filter RF1. The A response characteristic filters RF1 correspond to A corresponding values ​​selected from the plurality of corresponding values ​​in order from the value closest to the function value. This configuration makes it possible to make the ANC follow the movement of the noise source 200. The corresponding value can be a vector. The corresponding value is calculated based on the position discriminant function G * In this context, "selected in order" is not an expression for indicating the action of "selection," but an expression for explaining which of a plurality of corresponding values ​​corresponds to A corresponding values. The same applies in other similar contexts.

[0206] In this embodiment, the selector 190 selects, as the control filter CF1, the response characteristic filter RF1 corresponding to the corresponding value that is closest to the function value among the plurality of corresponding values ​​from the plurality of response characteristic filters RF1. This configuration is simple as it enables the ANC to follow the movement of the noise source 200.

[0207] In this embodiment, the control device 110 achieves ANC by causing the first speaker 10A to emit a first sound wave and the second speaker 10B to emit a second sound wave. At least a portion of the first radiation surface 15A of the first speaker 10A and at least a portion of the second radiation surface 15B of the second speaker 10B are located at different positions in a predetermined direction. This configuration is suitable for achieving ANC that is robust against the position of the noise source 200 in the predetermined direction.

[0208] In this embodiment, the reference microphone 130 and the specific microphone 150 are aligned in a direction different from the predetermined direction. This arrangement of the reference microphone 130 and the specific microphone 150 makes it easy for the way the reference signal x and the way the specific signal j change to differ when the noise source 200 moves in a direction different from the predetermined direction. This is advantageous from the perspective of switching the control filter CF1 based on the reference signal x and the specific signal j when the noise source 200 moves in a direction different from the predetermined direction. Therefore, this configuration, in combination with the positional relationship between the first radiation surface 15A and the second radiation surface 15B, can achieve ANC that is robust against the two-dimensional position of the noise source 200.

[0209] In this embodiment, the ANC system 500 includes a first wall 80A, a second wall 80B, and a space 86. The first wall 80A and the second wall 80B face each other. The space 86 is located between the first wall 80A and the second wall 80B. In the ANC, the control device 110 causes the first speaker 10A to emit a first sound wave away from the space 86, thereby reducing noise from inside the space 86 to outside the space 86. This configuration can clearly demonstrate the benefits of the high robustness of the ANC of this embodiment. Specifically, suppose the ANC system is not designed to perform ANC that is robust to the position of the noise source. In this case, if the noise source moves between the two walls, the propagation characteristics of the noise are likely to change. This is because the noise is affected by the walls, such as by reflection from the walls. In contrast, the robustness of the ANC system 500 according to this embodiment is also exhibited against movement between the two walls 80A and 80B of the noise source 200.

[0210] In this embodiment, the control device 110 includes a correction unit 164. The correction unit 164 corrects the specific signal j according to the reference signal x by convolution processing using a correction filter. The control device 110 determines the control filter CF1 based on the reference signal x and the corrected specific signal j. This configuration is advantageous from the viewpoint of improving the accuracy of ANC.

[0211] 15 shows a specific example of signal processing by the control device 110, which is applicable to the embodiment. In this example, the control device 110 is configured so that a control flag is turned on in the control stage.

[0212] In step S101, the control device 110 sets the initial values ​​of the filter coefficients of the control filter CF1 and the initial values ​​of the filter coefficients of the control filter CF2.

[0213] Next, in step S102, the control device 110 determines whether or not the control flag is on. If the control flag is on, the process proceeds to step S103. If the control flag is off, the control flow ends.

[0214] In step S103, the control device 110 acquires the reference signal x and the specific signal j.

[0215] Next, in step S104, the control device 110 calculates a position discriminant function G(n c ) is calculated.

[0216] Next, in step S105, the control device 110 calculates a plurality of position discriminant functions G * For each of these, the location discriminant function G * and the location discriminant function G(n c ) and calculate the NMSD between

[0217] Next, in step S106, the control device 110 determines whether the noise source 200 has moved. If it is determined that the noise source 200 has moved, the process proceeds to step S107. If it is determined that the noise source 200 has not moved, the process proceeds to step S109. Specifically, in step S106, the control device 110 determines the position discriminant function G that results in the minimum NMSD. * There is a location discriminant function G * Another location discriminant function G * If the state changes to , it is determined that the noise source 200 has moved.

[0218] Next, in step S107, the control device 110 determines the position discriminant function G that results in the minimum NMSD. * Noise control filter W1 corresponding to * and noise control filter W2 * Select .

[0219] Next, in step S108, the control device 110 sets the filter coefficients of the control filter CF1 to the filter coefficients of the selected noise control filter W1. The control device 110 sets the filter coefficients of the control filter CF2 to the filter coefficients of the selected noise control filter W2. In this manner, the control device 110 sets the filter coefficients.

[0220] Next, in step S109, the processor 180 generates a control signal y1 based on the control filter CF1 and the reference signal x, and generates a control signal y2 based on the control filter CF2 and the reference signal x.

[0221] In this way, for each channel, the control filter CF1 and the control filter CF2 are determined according to the position of the noise source 200. Then, the control signal y1 and the control signal y2 according to the position of the noise source 200 are generated.

[0222] An ANC method according to an embodiment will be described with reference to Fig. 16. The description regarding the first channel of the ANC system 500 is applicable to the second channel of the ANC system 500. The description regarding the second channel is applicable to the first channel.

[0223] In step S201, the reference microphone 130 detects a sound and generates a reference signal x.

[0224] In the ANC method according to this embodiment, the response characteristic RC1 of the first sound wave with respect to the reference signal x is determined in accordance with the position of the noise source 200. This determination is realized in the following step S202.

[0225] In step S202, a response characteristic RC1 is selected from a plurality of different reference characteristics based on the position of the noise source 200.

[0226] Specifically, in this embodiment, the multiple reference characteristics are associated one-to-one with the multiple reference positions 71. Each of the multiple reference characteristics is capable of achieving ANC when noise source 200 is located at the reference position 71 corresponding to that reference characteristic. From the multiple reference characteristics, the reference characteristic corresponding to the reference position 71 that is closest to the position of noise source 200 is selected as response characteristic RC1. In this embodiment, the multiple reference positions 71 are located within space 86 of structure 80.

[0227] In step S203, ANC is achieved by radiating a first sound wave based on the reference signal x and the response characteristic RC1.

[0228] In this embodiment, the direction of movement of noise source 200 that causes a change in response characteristic RC1 includes a first direction and a second direction different from the first direction. This configuration is suitable for achieving ANC that is robust to the two-dimensional position of noise source 200. Specifically, in this embodiment, the multiple reference positions 71 include a first reference position 71, a second reference position 71, and a third reference position 71. The direction in which the first reference position 71 and the second reference position 71 are aligned is different from the direction in which the first reference position 71 and the third reference position 71 are aligned, and the direction in which the second reference position 71 and the third reference position 71 are aligned is different from one another.

[0229] In this embodiment, ANC is achieved by radiating a first sound wave from the first radiation surface 15A of the first speaker 10A and a second sound wave from the second radiation surface 15B of the second speaker 10B. At least a portion of the first radiation surface 15A and at least a portion of the second radiation surface 15B are located at different positions relative to each other in a predetermined direction. The direction of movement of the noise source 200 that causes a change in the response characteristic RC1 includes a direction different from the predetermined direction. This configuration, combined with the positional relationship between the first radiation surface 15A and the second radiation surface 15B, can achieve ANC that is robust against the two-dimensional position of the noise source 200. Specifically, in this embodiment, the multiple reference positions 71 include two reference positions 71 aligned in a direction different from the predetermined direction.

[0230] Techniques applicable to the above-described embodiments will be described below.

[0231] In the above embodiment, the ANC system 500 is a two-channel ANC system. The first channel of the ANC system 500 includes a first error microphone 140A, a noise control filter W1, a first amplifier 111A, and a first speaker 10A. The second channel of the ANC system 500 includes a second error microphone 140B, a noise control filter W2, a second amplifier 111B, and a second speaker 10B. The first and second channels share the reference microphone 130 and the specific microphone 150.

[0232] However, there is no particular limitation on the number of channels of the ANC system 500. The ANC system 500 may be a one-channel ANC system or an ANC system with three or more channels.

[0233] In the above embodiment, the specific microphone 150 is a microphone different from the first error microphone 140A and the second error microphone 140B. However, the specific microphone 150 may also serve as the first error microphone 140A or the second error microphone 140B.

[0234] In the above embodiment, the ANC system 500 includes a structure 80. A space 86 is defined inside the structure 80. The ANC system 500 reduces noise from a noise source 200 in the space 86 outside the space 86. In the above embodiment, the specific microphone 150 is disposed outside the space 86 defined by the structure 80.

[0235] However, the specific microphone 150 may be placed inside the space 86 defined by the structure 80. Furthermore, the ANC system 500 does not have to include the structure 80. The ANC system 500 may be configured in an open space.

[0236] In the above embodiment, the ANC system 500 includes one reference microphone 130. However, the ANC system 500 may include multiple reference microphones 130. In the example shown in FIG. 17, the number of reference microphones 130 is two. Hereinafter, one of the two reference microphones 130 shown in FIG. 17 will be referred to as the first reference microphone 130A, and the other will be referred to as the second reference microphone 130B. In the example of FIG. 17, a composite signal obtained by adding together the detection signals of the two reference microphones 130A and 130B can be treated as a single reference signal x. According to the example of FIG. 17, compared to when the number of reference microphones 130 is one, it is easier to avoid a situation in which the level of the reference signal x decreases due to the noise source 200 moving away from the reference microphone 130, thereby reducing the ANC effect.

[0237] In the above embodiment, the ANC system 500 has one specific microphone 150. However, the ANC system 500 may have more than one specific microphone 150. In the example shown in FIG. 17, the number of specific microphones 150 is two. Hereinafter, one of the two specific microphones 150 shown in FIG. 17 will be referred to as the first specific microphone 150A, and the other will be referred to as the second specific microphone 150B. In the example of FIG. 17, a composite signal obtained by adding together the detection signals of the two specific microphones 150A and 150B can be treated as one specific signal j.

[0238] The number of reference microphones 130 may be one and the number of specific microphones 150 may be one. The number of reference microphones 130 may be multiple and the number of specific microphones 150 may be one. The number of reference microphones 130 may be one and the number of specific microphones 150 may be multiple. The number of reference microphones 130 may be multiple and the number of specific microphones 150 may be multiple. A configuration including one or multiple reference microphones 130 may be referred to as a reference microphone unit. A configuration including one or multiple specific microphones 150 may be referred to as a specific microphone unit. The reference microphone unit may also be referred to as a reference microphone unit. The specific microphone unit may also be referred to as a specific microphone unit. When there are multiple reference microphones 130, a composite signal obtained by adding together the detection signals of those reference microphones 130 may be treated as a single reference signal x. When there are a plurality of specific microphones 150, a composite signal obtained by adding together the detection signals of these specific microphones 150 can be treated as one specific signal j.

[0239] In the above embodiment, the control device 110 selects a response characteristic filter RF1 according to the position of the noise source 200, and uses the selected response characteristic filter RF1 as the control filter CF1. The control device 110 selects a response characteristic filter RF2 according to the position of the noise source 200, and uses the selected response characteristic filter RF2 as the control filter CF2. The response characteristic filter RF1 defines a response characteristic RC1 of a first sound wave with respect to the reference signal x. The response characteristic filter RF2 defines a response characteristic RC2 of a second sound wave with respect to the reference signal x. The control device 110 uses the control filter CF1 and the control filter CF2 to generate the first sound wave and the second sound wave such that noise is reduced at the positions of the first error microphone 140A and the second error microphone 140B.

[0240] In the above embodiment, the response characteristic filter RF1 is the noise control filter W1, and the response characteristic filter RF2 is the noise control filter W2.

[0241] However, the control device 110 may be configured to generate the first sound wave and the second sound wave so that noise is reduced at a position shifted from the position of the first error microphone 140A and a position shifted from the position of the second error microphone 140B. That is, the first control point 74A may be shifted from the position of the first error microphone 140A, and the second control point 74B may be shifted from the position of the second error microphone 140B. Such a configuration can be realized, for example, by virtual sensing. In one example of virtual sensing, the response characteristic filter RF1 includes a noise control filter W1 and an auxiliary filter H1. The response characteristic filter RF2 includes a noise control filter W2 and an auxiliary filter H2. The control device 110 selects the response characteristic filter RF1 according to the position of the noise source 200 and uses the selected response characteristic filter RF1 as the control filter CF1. The control device 110 selects the response characteristic filter RF2 according to the position of the noise source 200 and uses the selected response characteristic filter RF2 as the control filter CF2. For more details on virtual sensing, please refer to prior art documents.

[0242] In the above context, "selecting response characteristic filter RF1" may mean selecting only the noise control filter W1 out of the noise control filter W1 and the auxiliary filter H1. "selecting response characteristic filter RF1" may mean selecting only the auxiliary filter H1 out of the noise control filter W1 and the auxiliary filter H1. "selecting response characteristic filter RF1" may mean selecting both the noise control filter W1 and the auxiliary filter H1.

[0243] In the above context, "selecting response characteristic filter RF2" may mean selecting only noise control filter W2 out of noise control filter W2 and auxiliary filter H2. "selecting response characteristic filter RF2" may mean selecting only auxiliary filter H2 out of noise control filter W2 and auxiliary filter H2. "selecting response characteristic filter RF2" may mean selecting both noise control filter W2 and auxiliary filter H2.

[0244] In the above embodiment, the selector 190 selects, from the plurality of response characteristic filters RF1, the response characteristic filter RF1 that corresponds to the corresponding value that is closest to the value of the function among the plurality of corresponding values, as the control filter CF1.

[0245] However, the selector 190 may select B response characteristic filters RF1 from the plurality of response characteristic filters RF1. Here, B is a natural number equal to or greater than 2. The B response characteristic filters RF1 correspond to B corresponding values ​​selected from the plurality of corresponding values ​​in order of closestness to the function value. The processor 180 may implement ANC using a control filter CF1 based on the B response characteristic filters RF1. Specifically, the control filter CF1 may be configured to have filter coefficients obtained by interpolating the filter coefficients of the B response characteristic filters RF1. The interpolation of the B filter coefficients is, for example, a calculation of dividing the sum of the values ​​of the B filter coefficients by B.

[0246] Machine learning techniques can also be applied to the above-described embodiments. In one configuration example, the control device 110 has a trained machine learning model. The machine learning model determines the control filter CF1 and the control filter CF2 based on a reference signal x and a specific signal j. The machine learning model may determine the control filter CF1 and the control filter CF2 based on a relationship, such as the ratio between the reference signal x and the specific signal j. In one configuration example, the machine learning model determines the control filter CF1 and the control filter CF2 based on a function of the reference signal x and the specific signal j. Specifically, the function represents the ratio between the reference signal x and the specific signal j. The ratio between the reference signal x and the specific signal j is specifically the ratio between the level of the reference signal x and the level of the specific signal j. The function and the ratio may be vectors. The function may be a position discrimination function G. In addition, in the ANC method, the machine learning model determines the response characteristic RC1 according to the position of the noise source 200.

[0247] In the learning phase, the control device 110 may construct a trained machine learning model by learning using a plurality of training data. In the execution phase, the machine learning model determines the control filter CF1 and the control filter CF2 based on the reference signal x and the specific signal j.

[0248] The machine learning model is based on at least one selected from the group consisting of, for example, a neural network, a random forest, and a support vector machine. The machine learning model may be based on deep learning. In the above example, the learning is supervised learning. However, the learning may also be unsupervised learning.

[0249] The training data used for learning can be a data set from ANC previously performed by the ANC system 500. In one specific example, while ANC is being performed, the memory 170 stores multiple combinations of functions of the reference signal x and the specific signal j, and the control filters CF1 and CF2. The control device 110 uses the multiple combinations as multiple training data to generate a trained machine learning model.

[0250] The description regarding the first channel is applicable to the second channel. The description regarding the second channel is applicable to the first channel. The description regarding the ANC system 500 is applicable to the ANC method. The description regarding the ANC method is applicable to the ANC system 500.

[0251] [Experimental Example] The present invention will be described in detail using experimental examples. However, the following experimental examples are merely examples of the present invention, and the present invention is not limited to the following experimental examples. In the following description, "height" refers to the distance from the floor 85 in the upward direction D1.

[0252] A first experimental arrangement 600 shown in Figures 18A and 18B, a second experimental arrangement 700 shown in Figures 19A and 19B, and a third experimental arrangement 800 shown in Figures 20A and 20B were constructed. In the first experimental arrangement 600, the second experimental arrangement 700, and the third experimental arrangement 800, two film speakers 10 were fabricated and arranged as shown in Figure 21. In addition, a reference position 71 was set as shown in Figures 22 and 23. A detailed explanation will be given below.

[0253] 1-1. First Experimental Configuration 600 (FIGS. 18A and 18B) In the first experimental arrangement 600, a structure 80 is placed on a floor 85. The structure 80 defines a space 86 therein. The structure 80 includes a first wall 80A, a second wall 80B, and a third wall 80C. The first support 90A is a plate-shaped door and is attached to the first wall 80A via an attachment (not shown) so as to be rotatable about its axis of rotation in the vertical direction D1. The second support 90B is a plate-shaped door and is attached to the second wall 80B via an attachment (not shown) so as to be rotatable about its axis of rotation in the vertical direction D1. A gap 93 exists between the first support 90A and the second support 90B. The first wall 80A and the second wall 80B face each other in the left-right direction D2 and are connected via the third wall 80C. In the left-right direction D2, the space 86 is located between the first wall 80A and the second wall 80B. In the front-to-rear direction D3, the space 86 is located between the combination of the first support 90A and the second support 90B and the third wall 80C. A desk 95, a first reference microphone 130A, a second reference microphone 130B, and a noise source 200 are disposed inside the space 86. A first error microphone 140A, a second error microphone 140B, and a specific microphone 150 are disposed outside the space 86. The first experimental arrangement 600 further includes a control device 110, a first amplifier 111A, and a second amplifier 111B, which are not shown in FIG. 18A. Some elements, such as the legs of the desk 95, are not shown in FIG. 18A.

[0254] Each of the first wall 80A and the second wall 80B extends in the vertical direction D1 and the front-to-back direction D3, and has a dimension of 190 cm in the vertical direction D1 and 135 cm in the front-to-back direction D3. The third wall 80C extends in the vertical direction D1 and the left-to-right direction D2, and has a dimension of 190 cm in the vertical direction D1 and 90 cm in the left-to-right direction D2.

[0255] The first vertical direction DvA is the same as the up-down direction D1. The first horizontal direction DhA is the same as the left-right direction D2. The first thickness direction DtA is the same as the front-rear direction D3. The second vertical direction DvB is the same as the up-down direction D1. The second horizontal direction DhB is the same as the left-right direction D2. The second thickness direction DtB is the same as the front-rear direction D3. Specifically, each of the first support 90A and the second support 90B is fixed in an orientation extending in the up-down direction D1 and the left-right direction D2 so that the height of their respective centers is 135 cm. In this orientation, each of the first support 90A and the second support 90B has a dimension of 60 cm in the up-down direction D1 and a dimension of 44 cm in the left-right direction D2. A gap 93 of 2 cm is formed between the first support 90A and the second support 90B in the left-right direction D2. In the vertical direction D1, an upper opening 87 is formed above the first support body 90A and the second support body 90B, and a lower opening 88 is formed below them. The upper opening 87 includes a front opening and a ceiling opening.

[0256] Within the space 86, a desk 95 has a top 95t that extends in the left-right direction D2 and the front-back direction D3 at a height of 70 cm so as to contact the first wall 80A, the second wall 80B, and the third wall 80C. The top 95t has a dimension of 90 cm in the left-right direction D2 and a dimension of 45 cm in the front-back direction D3. The first reference microphone 130A is disposed at a height of 120 cm, 12 cm away from the first wall 80A in the left-right direction D2, 50 cm away from the third wall 80C in the front-back direction D3, and 80 cm away from the first support 90A in the front-back direction D3. The second reference microphone 130B is disposed at a height of 120 cm, 12 cm away from the second wall 80B in the left-right direction D2, 50 cm away from the third wall 80C in the front-back direction D3, and 80 cm away from the second support 90B in the front-back direction D3.

[0257] Outside the space 86, the specific microphone 150 is disposed at a height of 135 cm and 10 cm away from the first support 90A and the second support 90B in the front-to-back direction D3. In the left-to-right direction D2, the center position of the first support 90A and the position of the specific microphone 150 coincide with each other.

[0258] Outside the space 86, the first control point 74A and the second control point 74B are set at a height of 135 cm, 100 cm away from the first support 90A and the second support 90B in the front-to-back direction D3, and 46 cm away from each other in the left-to-right direction D2. In the left-to-right direction D2, the center position of the first support 90A coincides with the position of the first control point 74A, and the center position of the second support 90B coincides with the position of the second control point 74B. A first error microphone 140A is located at the first control point 74A. A second error microphone 140B is located at the second control point 74B.

[0259] In the control stages of Experimental Examples 1-A, 2-A, 3-A, 3-D, 4-A, and 4-D described below, the first error microphone 140A and the second error microphone 140B are not necessary to achieve ANC. However, in these experiments, to evaluate the noise reduction effect, the first error microphone 140A was left at the first control point 74A after the tuning stage and used as the first measurement microphone 550A. In addition, the second error microphone 140B was left at the second control point 74B after the tuning stage and used as the second measurement microphone 550B.

[0260] The first reference microphone 130A and the second reference microphone 130B are ME52W manufactured by Olympus Corporation. The first error microphone 140A and the second error microphone 140B are Lch and Rch of CS-10EM manufactured by Roland Corporation. The specific microphone 150 is C-2 manufactured by BEHRINGER. The noise source 200 is SC-A17 manufactured by D&M Holdings Inc.

[0261] Two film speakers 10 having the configuration of the first configuration example shown in FIGS. 13A and 13B were fabricated. Specifically, in fabricating each film speaker 10, a 0.16 mm-thick adhesive sheet (double-sided tape) was used as the first bonding layer 51, in which both sides of a nonwoven fabric were impregnated with an acrylic adhesive. The intervening layer 40 was a 3 mm-thick closed-cell foam obtained by foaming a mixture containing ethylene propylene rubber and butyl rubber at an expansion ratio of approximately 10 times. The second bonding layer 52 was a 0.15 mm-thick adhesive sheet (double-sided tape) in which a nonwoven fabric substrate was coated on both sides with an adhesive containing a solventless acrylic resin. The piezoelectric film 35 was a polyvinylidene fluoride film (total thickness 33 μm) with copper electrodes (containing nickel) vapor-deposited on both sides. In each of the film speakers 10 produced, the first bonding layer 51, the intervening layer 40, the second bonding layer 52, and the piezoelectric film 35 have dimensions of 50 cm in height and 35 cm in width when viewed in a plane, and have an undivided, unframed, plate-like shape with overlapping contours when viewed in a plane.

[0262] 21, one of the film speakers 10 was attached as a first speaker 10A to one of the two main surfaces of the first support body 90A that faces the outside of the space 86. The other of the film speakers 10 was attached as a second speaker 10B to one of the two main surfaces of the second support body 90B that faces the outside of the space 86.

[0263] The longitudinal direction DlA is the same as the first vertical direction DvA, i.e., the same as the up-down direction D1. The short-side direction DsA is the same as the first horizontal direction DhA, i.e., the same as the left-right direction D2. The longitudinal direction DlB is the same as the second vertical direction DvB, i.e., the same as the up-down direction D1. The short-side direction DsB is the same as the second horizontal direction DhB, i.e., the same as the left-right direction D2.

[0264] The dimension LxA of the first radiation surface 15A in the longitudinal direction DlA is 50 cm. The dimension LyA of the first radiation surface 15A in the lateral direction DsA is 35 cm. The dimension LxB of the second radiation surface 15B in the longitudinal direction DlB is 50 cm. The dimension LyB of the second radiation surface 15B in the lateral direction DsB is 35 cm.

[0265] The first top margin M1 is 5 cm. The first bottom margin M2 is 5 cm. The first left margin M3 is 2.5 cm. The first right margin M4 is 2.5 cm. The second top margin M5 is 5 cm. The second bottom margin M6 is 5 cm. The second left margin M7 is 2.5 cm. The second right margin M8 is 2.5 cm.

[0266] 1-2. Second Experimental Configuration 700 (FIGS. 19A and 19B) The specific microphone 150 is not placed in the second experimental arrangement 700. Otherwise, the second experimental arrangement 700 is the same as the first experimental arrangement 600.

[0267] 1-3. Third Experimental Configuration 800 (FIGS. 20A and 20B) In the third experimental arrangement 800, the first error microphone 140A and the second error microphone 140B were arranged at positions different from those in the second experimental arrangement 700. Specifically, the first error microphone 140A and the second error microphone 140B were arranged outside the space 86 at a height of 135 cm, 10 cm away from the first support 90A and the second support 90B in the front-to-back direction D3, and 46 cm away from each other in the left-to-right direction D2. In the left-to-right direction D2, the center position of the first support 90A coincides with the position of the first error microphone 140A, and the center position of the second support 90B coincides with the position of the second error microphone 140B. In the third experimental arrangement 800, a first measurement microphone 550A different from the first error microphone 140A was arranged at the first control point 74A, and a second measurement microphone 550B different from the second error microphone 140B was arranged at the second control point 74B. The first measurement microphone 550A and the second error microphone 140B are each a C-2 manufactured by BEHRINGER, Inc. Otherwise, the third experimental setup 800 is the same as the second experimental setup 700.

[0268] [2. Reference position 71 (Figures 22 and 23)] As shown in Figure 22, 25 reference positions 71 were set. The 25 reference positions 71 were arranged in an array of 5 rows and 5 columns at a height of 120 cm. The rows extended in the left-right direction D2, and the columns extended in the front-back direction D3. In this array, the pitch in the row direction was 12.5 cm. The pitch in the column direction was 10 cm. In Figure 22, a target range 560 for noise reduction is indicated by a dotted line. The target range 560 corresponds to the area in which the 25 reference positions 71 are distributed.

[0269] Here, the 25 reference positions 71 are each given a subscript, and are designated as reference position 711, reference position 712, reference position 713, reference position 714, reference position 715, reference position 716, reference position 717, reference position 718, reference position 719, and reference position 71 10 , reference position 71 11 , reference position 71 12 , reference position 71 13 , reference position 71 14 , reference position 71 15 , reference position 71 16 , reference position 71 17 , reference position 71 18 , reference position 71 19 , reference position 71 20 , reference position 71 21 , reference position 71 22 , reference position 71 23 , reference position 71 24 and reference position 71 25 In FIG. 23, the subscript of the reference position 71 is written in parentheses. Of the 25 reference positions 71, the reference position 71 located at the center is 13 is located exactly in the middle between the first wall 80A and the second wall 80B in the left-right direction D2, 50 cm away from the third wall 80C in the front-rear direction D3, and 80 cm away from the first support 90A in the front-rear direction D3. In the column direction (front-rear direction D3), the reference position 711, the reference position 712, the reference position 713, the reference position 714, and the reference position 715 are arranged in this order, In the column direction (front-rear direction D3), reference positions 716, 717, 718, 719, and 71 10 are arranged in this order, Reference position 71 in the column direction (front-back direction D3) 11 , reference position 71 12 , reference position 71 13 , reference position 71 14 and reference position 71 15 are arranged in this order, Reference position 71 in the column direction (front-back direction D3) 16 , reference position 71 17 , reference position 71 18 , reference position 71 19 and reference position 71 20 are arranged in this order, Reference position 71 in the column direction (front-back direction D3) 21 , reference position 71 22 , reference position 71 23 , reference position 71 24 and reference position 71 25 are arranged in this order, ·In the row direction (horizontal direction D2), reference position 711, reference position 716, reference position 71 11 , reference position 71 16 and reference position 71 21 are arranged in this order, ·In the row direction (horizontal direction D2), reference position 712, reference position 717, reference position 71 12 , reference position 71 17 and reference position 71 22 are arranged in this order, ·In the row direction (horizontal direction D2), reference position 713, reference position 718, reference position 71 13 , reference position 71 18 and reference position 71 23 are arranged in this order, ·In the row direction (horizontal direction D2), reference position 714, reference position 719, reference position 71 14 , reference position 71 19 and reference position 71 24 are arranged in this order, ·In the row direction (horizontal direction D2), reference position 715, reference position 71 10 , reference position 71 15, reference position 71 20 and reference position 71 25 are listed in this order.

[0270] [3. Experimental Example] The experimental examples shown in Table 1 below were carried out. Each experimental example will be described in detail below. [Table 1]

[0271] (Experimental Example 1-A) In Experimental Example 1-A, a first experimental arrangement 600 was adopted. In Experimental Example 1-A, a tuning stage and a control stage were performed according to the embodiment, which will be described in detail below.

[0272] (Tuning Stage) In the tuning stage, noise source 200 was placed at reference position 711. In this state, noise was generated from noise source 200. This noise contained frequency components between 100 Hz and 700 Hz substantially equally. While the noise was being generated, sound was detected by first reference microphone 130A and second reference microphone 130B, and a composite signal was generated by adding together the detection signals of the two reference microphones 130A and 130B. This composite signal was treated as reference signal x. Sound was detected by specific microphone 150, and a specific signal j was generated. Sound was detected by first error microphone 140A, and a first error signal e1 was generated. Sound was detected by second error microphone 140B, and a second error signal e2 was generated.

[0273] The processor 180 of the control device 110 generated a control signal y1 based on the noise control filter W1 and the reference signal x. The first amplifier 111A amplified the control signal y1. The first speaker 10A emitted a first sound wave based on the amplified control signal y1. The processor 180 generated a control signal y2 based on the noise control filter W2 and the reference signal x. The second amplifier 111B amplified the control signal y2. The second speaker 10B emitted a second sound wave based on the amplified control signal y2.

[0274] The processor 180 uses an adaptive algorithm to identify the filter coefficients of the noise control filter W1 so that the first error signal e1 is reduced. The processor 180 uses an adaptive algorithm to identify the filter coefficients of the noise control filter W2 so that the second error signal e2 is reduced. The position estimation unit 160 of the control device 110 calculates a position discriminant function G based on the reference signal x and the specific signal j. * The memory 170 stores the position discriminant function G * The noise control filter W1 having the identified filter coefficients and the noise control filter W2 having the identified filter coefficients are stored in a state where they are associated with each other.

[0275] The position of the noise source 200 is reference position 712, reference position 713, reference position 714, reference position 715, reference position 716, reference position 717, reference position 718, reference position 719, reference position 71 10 , reference position 71 11 , reference position 71 12 , reference position 71 13 , reference position 71 14 , reference position 71 15 , reference position 71 16 , reference position 71 17 , reference position 71 18 , reference position 71 19 , reference position 71 20 , reference position 71 21 , reference position 71 22 , reference position 71 23 , reference position 7124 and reference position 71 25 The tuning stage was carried out in the same manner as the above tuning stage in which the noise source 200 was placed at the reference position 711. In this way, the positions of the 25 reference positions 711 to 71 25 For the location discriminant function G * The noise control filter W1 having the identified filter coefficients and the noise control filter W2 having the identified filter coefficients are stored in the memory 170 in a state where they are associated with each other.

[0276] The tuning stage was specifically performed as described above with reference to Figures 7A to 7C. Based on the NLMS algorithm, the position discriminant function G * The position discriminant function G * When generating the specific signal j, the correction unit 164 corrects the specific signal j.

[0277] (Control Stage) In the control stage, noise is generated from the noise source 200, and the position of the noise source 200 is changed to a reference position 711, a reference position 712, a reference position 713, a reference position 714, a reference position 715, a reference position 716, a reference position 717, a reference position 718, a reference position 719, a reference position 720, a reference position 721, a reference position 722, a reference position 723, a reference position 724, a reference position 725, a reference position 726, a reference position 727, a reference position 728, a reference position 729, a reference position 730, a reference position 731, a reference position 732, a reference position 733, a reference position 734, a reference position 735, a reference position 736, a reference position 737, a reference position 738, 10 , reference position 71 11 , reference position 71 12 , reference position 71 13 , reference position 71 14 , reference position 71 15 , reference position 71 16 , reference position 71 17 , reference position 71 18 , reference position 71 19 , reference position 71 20 , reference position 71 21 , reference position 71 22 , reference position 71 23 , reference position 71 24 and reference position 71 25The reference positions 71 were changed in this order. In Figure 23, the movement route of noise source 200 in the control stage is indicated by a dotted line. As with the tuning stage, the noise contained frequency components of 100 Hz or more and 700 Hz or less substantially equally. The time for which noise source 200 was kept stationary at each reference position 71 was set to one minute.

[0278] In the control stage, the position estimation unit 160 of the control device 110 estimates the current time step n c The reference signal x(n c ) and the current time step n c A specific signal j(n c ) based on the current time step n of the noise source 200 c The position discriminant function G(n c The error calculation unit 191 of the control device 110 generates the 25 position discriminant functions G obtained in the tuning stage. * For each of these, the location discriminant function G * and the location discriminant function G(n c The minimum value discriminator 192 of the control device 110 discriminated the smallest of the 25 calculated closenesses. The filter selector 193 of the control device 110 selected the position discriminant function G that yielded the smallest closeness. * Noise control filter W1 corresponding to * and noise control filter W2 * was selected.

[0279] The processor 180 selects the control filter CF1 from the selected noise control filter W1. * The processor 180 sets the set control filter CF1 and the current time step n c The reference signal x(n c ) based on the current time step n c The control signal y1(n c ) was generated by the first amplifier 111A. c ) is amplified by the first speaker 10A. c ) and emitted the first sound wave.

[0280] The processor 180 then connects the control filter CF2 to the selected noise control filter W2. * The processor 180 sets the set control filter CF2 and the current time step n c The reference signal x(n c ) based on the current time step n c Control signal y2(n c ) was generated by the second amplifier 111B. c ) is amplified by the second speaker 10B. c ) and emitted a second sound wave.

[0281] In the control stage, the position discriminant function G * and the location discriminant function G(n c ) and determine the minimum proximity, and noise control filter W1 * and noise control filter W2 * , setting of the control filter CF1 and the control filter CF2, and the control signal y1(n c ) and control signal y2(n c ) and the emission of the first and second sound waves were repeated while moving to the noise source 200. In this way, the noise control filter W1 * and noise control filter W2 * was selected and used to control the first and second sound waves, thereby maintaining a state in which the first and second sound waves are formed according to the position of the noise source 200.

[0282] The control stage was specifically performed as described above with reference to Figures 8A to 8C. The position discriminant function G(n c ), the correction unit 164 generates the specific signal j(n c ) was corrected. The position discrimination function G * and the location discriminant function G(n c The proximity of the control filter CF1 to the selected noise control filter W1 was evaluated based on NMSD.* The filter coefficients of the control filter CF1 are substituted into the control filter CF2. The control filter CF2 is set by substituting the filter coefficients of the selected noise control filter W2. * This was done by substituting the filter coefficients of the control filter CF1 and the control filter CF2. In the control stage, the filter coefficients of the control filter CF1 and the control filter CF2 were not updated by an adaptive algorithm.

[0283] (Experimental Example 1-B) In Experimental Example 1-B, the second experimental arrangement 700 was adopted. In Experimental Example 1-B, the tuning stage of Experimental Example 1-A was not performed. In the control stage of Experimental Example 1-B, noise was generated from the noise source 200, and the position of the noise source 200 was moved to the reference positions 711-712. 25 In the control stage of Experimental Example 1-B, unlike the control stage of Experimental Example 1-A, the filter coefficients of the control filters CF1 and CF2 were updated by an adaptive algorithm.

[0284] (Experimental Example 1-C) In experimental example 1-C, a third experimental arrangement 800 was adopted. In experimental example 1-C, virtual sensing was performed by using auxiliary filters H1 and H2 in addition to noise control filters W1 and W2. This will be explained in detail below.

[0285] In Experimental Example 1-C, the tuning stage of Experimental Example 1-A was not performed. In the tuning stage of Experimental Example 1-C, the noise source 200 was positioned at the reference position 71. 13 In the control stage of Experimental Example 1-C, unlike the control stage of Experimental Example 1-A, in addition to the noise control filter W1 and the noise control filter W2, the reference position 71 was identified as described above while the noise source 200 was moving. 13In other words, in the control stage of Experimental Example 1-C, the noise source 200 was positioned at the reference position 71 while the noise source 200 was moving. 13 ANC was performed using a filter optimized for the state in which it was placed in the control stage of Experimental Example 1-C. In addition, in the control stage of Experimental Example 1-C, the filter coefficients of the noise control filters W1 and W2, which were created taking into account the auxiliary filters H1 and H2, were updated using an adaptive algorithm.

[0286] (Experimental Example 2-A) In Experimental Example 2-A, an experiment similar to Experimental Example 1-A was carried out.

[0287] (Experimental Example 2-B) In Experimental Example 2-B, an experiment similar to Experimental Example 1-B was carried out.

[0288] (Experimental Example 2-C) In Experimental Example 2-C, an experiment similar to Experimental Example 1-C was carried out.

[0289] (Experimental Example 3-A) In Experimental Example 3-A, an experiment similar to Experimental Example 1-A was carried out.

[0290] (Experimental Example 3-B) In Experimental Example 3-B, an experiment similar to Experimental Example 1-B was carried out.

[0291] (Experimental Example 3-D) In Experimental Example 3-D, the first experimental arrangement 600 was adopted, as in Experimental Example 1-A. In the control stage of Experimental Example 3-D, noise was generated from the noise source 200, and the position of the noise source 200 was moved to the reference positions 711 to 71 25 In the control stage of Experimental Example 3-D, unlike the control stage of Experimental Example 1-A, the control filter CF1 is moved to the reference position 71 obtained in Experimental Example 1-A while the noise source 200 is moving. 13 Noise control filter W1 corresponding to * and the control filter CF2 was fixed at the reference position 71 obtained in Experimental Example 1-A.13 Noise control filter W2 corresponding to * In the control stage of experimental example 3-D, the filter coefficients of the control filters CF1 and CF2 were not updated by the adaptive algorithm. That is, in the control stage of experimental example 3-D, the noise source 200 was fixed at the reference position 71 while the noise source 200 was moving. 13 ANC was performed using a filter optimized for the positioning of the microphone.

[0292] (Experimental Example 4-A) In Experimental Example 4-A, an experiment similar to Experimental Example 1-A was carried out.

[0293] (Experimental Example 4-B) In Experimental Example 4-B, an experiment similar to Experimental Example 1-B was carried out.

[0294] (Experimental Example 4-D) In Experimental Example 4-D, an experiment similar to Experimental Example 3-D was carried out.

[0295] FIG. 24A is a graph showing the change over time in the silencing effect of Experimental Examples 1-A, 1-B, and 1-C. FIG. 24B is a rough manual tracing of the graph of FIG. 24A. FIG. 25A is a graph showing the change over time in the silencing effect of Experimental Examples 2-A, 2-B, and 2-C. FIG. 25B is a rough manual tracing of the graph of FIG. 25A. FIG. 26A is a graph showing the change over time in the silencing effect of Experimental Examples 3-A, 3-B, and 3-D. FIG. 26B is a rough manual tracing of the graph of FIG. 26A. FIG. 27A is a graph showing the change over time in the silencing effect of Experimental Examples 4-A, 4-B, and 4-D. FIG. 27B is a rough manual tracing of the graph of FIG. 27A.

[0296] 24A to 27B, the horizontal axis represents time (unit: minutes). The numbers in parentheses along the horizontal axis represent the positions of the noise source 200 at the 25 reference positions 711 to 71 25 The subscript of the reference position 71 indicates where the reference position is located.

[0297] 24A to 27B, the vertical axis represents the silencing effect (unit: dB). The silencing effects shown in Figures 24A to 27B are based on the first silencing effect and the second silencing effect.

[0298] Specifically, the overall level obtained by subtracting the first silencing level from the first noise level was evaluated as the first silencing effect grasped by the first measurement microphone 550A. The first noise level indicates the sound pressure level measured by the first measurement microphone 550A when each experimental example was modified so that no sound was emitted from the first speaker 10A and the second speaker 10B. The first silencing level indicates the sound pressure level measured by the first measurement microphone 550A in each experimental example. More specifically, the first silencing level was subtracted from the first noise level for N discrete frequencies in the frequency band from 10 Hz to 1000 Hz. When the difference in sound pressure levels at the n-th discrete frequency among the N discrete frequencies is denoted by Ln, the sum L of the difference levels obtained by the following Equation 3 was calculated. (Equation 3) L = 10 × Log 10 (Σ10 (Ln / 10) )(1≦n≦N) This sum L is the overall level. This sum L was evaluated as the first silencing effect sensed by the first measuring microphone 550A.

[0299] The overall level obtained by subtracting the second noise level from the second noise level was evaluated as the second noise reduction effect measured by the second measurement microphone 550B. The second noise level indicates the sound pressure level measured by the second measurement microphone 550B when each experimental example was modified so that sound was not emitted from the first speaker 10A and the second speaker 10B. The second noise reduction level indicates the sound pressure level measured by the second measurement microphone 550B in each experimental example. More specifically, the second noise reduction level was subtracted from the second noise level for N discrete frequencies in the frequency band from 10 Hz to 1000 Hz. When the difference in sound pressure levels at the nth discrete frequency among the N discrete frequencies is denoted by Ln, the sum L of the difference levels obtained by Equation 3 above was calculated. This sum L is the overall level. This sum L was evaluated as the second silencing effect detected by the second measuring microphone 550B.

[0300] The silencing effects shown in FIGS. 24A to 27B are average values ​​of the first silencing effect and the second silencing effect.

[0301] 24A to 27B, it can be seen that in Experimental Examples 1-B, 2-B, 3-B, and 4-B, it takes about one minute for the silencing effect to reach its full potential after noise source 200 is moved. In contrast, it can be seen that in Experimental Examples 1-A, 2-A, 3-A, and 4-A, a high silencing effect is achieved immediately after noise source 200 is moved.

[0302] 24A to 27B, in Experimental Examples 1-C, 2-C, 3-D, and 4-D, the noise source 200 is located at the reference position 71 13 In contrast, in Experimental Examples 1-A, 2-A, 3-A, and 4-A, it is understood that a high silencing effect is exerted regardless of the position at which noise source 200 is disposed.

[0303] 28 is a graph showing the change over time in the position corresponding to the noise control filter in the control stage of Experimental Example 1-A. Specifically, the "position corresponding to the noise control filter" is the position discriminant function G corresponding to the selected noise control filter W1 and noise control filter W2. * The horizontal axis represents time (unit: minutes). The vertical axis represents the position corresponding to the noise control filter. The numbers in parentheses along the horizontal axis represent the positions where the noise source 200 is located at the 25 reference positions 711 to 71 25 The subscripts of the reference position 71 indicate which of the 25 reference positions 711 to 711 corresponds to the noise control filter. 25 28, when noise source 200 is placed at reference position 711, the position corresponding to the noise control filter is reference position 711, when noise source 200 is placed at reference position 712, the position corresponding to the noise control filter is reference position 712, and so on. 24 When the noise control filter is located at the reference position 71, the position corresponding to the noise control filter is the reference position 71. 24 and the noise source 200 is at the reference position 71 25 When the noise control filter is located at the reference position 71, the position corresponding to the noise control filter is the reference position 71. 25 28, it can be seen that the position of noise source 200 is correctly estimated. Although not shown, similar graphs were obtained in Experimental Example 2-A, Experimental Example 3-A, and Experimental Example 4-A.

[0304] (Addendum) The present disclosure provides the following techniques.

[0305] (Technology 1) a reference microphone unit that detects sound and generates a reference signal, the reference microphone unit including one or more reference microphones; a specific microphone unit that detects sound and generates a specific signal, the specific microphone unit including one or more specific microphones; a first speaker that emits a first sound wave; a control device that determines a control filter that defines a response characteristic of the first sound wave to the reference signal based on the reference signal and the specific signal, and realizes active noise control by radiating the first sound wave from the first speaker based on the reference signal and the control filter, Active noise control system. Here, the reference signal and the specific signal may be vectors.

[0306] (Technology 2) the control device determines the control filter based on a function of the reference signal and the specific signal. The active noise control system according to technique 1. Here, the function may be a vector.

[0307] (Technology 3) the control device determines the control filter based on a ratio of the reference signal and the specific signal. 3. The active noise control system according to claim 1 or 2. Here, the ratio may be a ratio between the level of the reference signal and the level of the specific signal. The ratio may be a vector.

[0308] (Technology 4) The determination of the control filter is not a process of converging the filter coefficient of the control filter by an adaptive algorithm. 4. An active noise control system according to any one of claims 1 to 3. The above expression "determination of the control filter is not a process of converging the filter coefficients of the control filter by an adaptive algorithm" will now be explained. This expression encompasses a case where, in a case where a tuning stage and a control stage are present, the process of converging the filter coefficients by an adaptive algorithm is performed in the tuning stage, but the process of converging the filter coefficients of the control filter by an adaptive algorithm is not performed in the control stage.

[0309] (Technology 5) The control filter associates the reference signal with the first sound wave in a one-to-one correspondence. 5. An active noise control system according to any one of claims 1 to 4.

[0310] (Technology 6) The control device a memory that stores a plurality of response characteristic filters that define the response characteristics different from one another; a selector that selects A response characteristic filters from the plurality of response characteristic filters based on the reference signal and the specific signal, where A is a natural number equal to or greater than 1; a processor for implementing the active noise control using the control filter based on the A response characteristic filters. 6. An active noise control system according to any one of claims 1 to 5.

[0311] (Technology 7) the plurality of response characteristic filters are associated one-to-one with a plurality of corresponding values; each of the plurality of response characteristic filters is capable of realizing the active noise control when a function of the reference signal and the specific signal takes the corresponding value corresponding to the response characteristic filter; The A response characteristic filters correspond to A corresponding values ​​selected from the plurality of corresponding values ​​in order from the value closest to the function value. The active noise control system described in technology 6. Here, the corresponding value may be a vector.

[0312] (Technology 8) The control device has a trained machine learning model, The machine learning model determines the control filter based on the reference signal and the specific signal. 6. An active noise control system according to any one of claims 1 to 5.

[0313] (Technology 9) a second speaker that emits a second sound wave; the control device realizes the active noise control by causing the first speaker to emit the first sound wave and the second speaker to emit the second sound wave; the first speaker has a first radiation surface that radiates the first sound wave; the second speaker has a second radiation surface that radiates the second sound wave; At least a portion of the first radiation surface and at least a portion of the second radiation surface are located at different positions from each other in a predetermined direction. 9. An active noise control system according to any one of claims 1 to 8.

[0314] (Technology 10) the one or more reference microphones include a first reference microphone; the one or more specific microphones include a first specific microphone; the first reference microphone and the first specific microphone are arranged in a direction different from the predetermined direction; The active noise control system described in technology 9.

[0315] (Technology 11) The first wall and a second wall opposite the first wall; a space located between the first wall and the second wall, In the active noise control, the control device reduces noise from inside the space outside the space by causing the first speaker to radiate the first sound waves away from the space. 11. An active noise control system according to any one of claims 1 to 10.

[0316] (Technology 12) the control device includes a correction unit that corrects the specific signal in accordance with the reference signal by processing using a correction filter; the control device determines the control filter based on the reference signal and the corrected specific signal. 12. An active noise control system according to any one of claims 1 to 11.

[0317] (Technology 13) Detecting a sound and generating a reference signal; determining a response characteristic of the first sound wave to the reference signal in response to a position of a noise source; and realizing active noise control by radiating the first sound wave based on the reference signal and the response characteristic. Active noise control method.

[0318] (Technology 14) selecting the response characteristic from a plurality of different reference characteristics based on the location of the noise source; The active noise control method according to technology 13.

[0319] (Technology 15) the plurality of reference characteristics are in one-to-one correspondence with a plurality of reference positions; each of the plurality of reference characteristics is capable of realizing the active noise control when the noise source is located at the reference position corresponding to the reference characteristic; selecting, from the plurality of reference characteristics, a reference characteristic corresponding to a reference position among the plurality of reference positions that is closest to the position of the noise source as the response characteristic; The active noise control method according to technology 14.

[0320] (Technology A) selecting A response characteristic filters from a plurality of response characteristic filters that define the different response characteristics based on the position of the noise source, where A is a natural number equal to or greater than 1; and implementing the active noise control using the control filter based on the A response characteristic filters. 16. An active noise control method according to any one of techniques 13 to 15.

[0321] (Technology B) the plurality of response characteristic filters are associated one-to-one with a plurality of reference positions; each of the plurality of response characteristic filters is capable of realizing the active noise control when the noise source is located at the reference position corresponding to that response characteristic filter; the A response characteristic filters correspond to A reference positions selected from the plurality of reference positions in order from the position closest to the value of the noise source; The active noise control method described in Technology A.

[0322] (Technology 16) determining the response characteristics according to the position of the noise source using a trained machine learning model; The active noise control method according to technology 13.

[0323] (Technology 17) The direction of movement of the noise source that causes the transformation of the response characteristic includes a first direction and a second direction different from the first direction. The active noise control method according to any one of techniques 13, 14, 15, A, B and 16.

[0324] (Technology 18) and realizing the active noise control by radiating the first sound wave from a first radiation surface of a first speaker and radiating the second sound wave from a second radiation surface of a second speaker, wherein at least a portion of the first radiation surface and at least a portion of the second radiation surface are located at different positions from each other in a predetermined direction; the direction of movement of the noise source that causes the change in response characteristic includes a direction different from the predetermined direction; An active noise control method according to any one of techniques 13, 14, 15, A, B, 16 and 17.

[0325] (Technology C) The determination of the response characteristics is not a process of converging the filter coefficients of a control filter that realizes the response characteristics by an adaptive algorithm. An active noise control method according to any one of techniques 13, 14, 15, A, B, 16, 17 and 18.

[0326] (Technology D) The response characteristic provides a one-to-one correspondence between the reference signal and the first sound wave. The active noise control method according to any one of techniques 13, 14, 15, A, B, 16, 17, 18 and C.

[0327] (Technology E) correcting the specific signal according to the reference signal by processing using a correction filter; determining a control filter that defines the response characteristics based on the reference signal and the corrected specific signal; The active noise control method according to any one of techniques 13, 14, 15, A, B, 16, 17, 18, C and D. [Industrial Applicability]

[0328] According to the present invention, it is possible to realize ANC that is robust against the location of a noise source. The ANC of the present invention can provide a quiet and comfortable environment for people. For example, the ANC of the present invention can provide such an environment around a personal booth. [Explanation of symbols]

[0329] 10, 10A, 10B, 310 Speaker 15,15A,15B Radiation surface 15jA, 15kA, 15lA, 15mA, 15jB, 15kB, 15lB, 15mB, 80fA, 80lA, 80mA, 80fB, 80lB, 80mB, 80lC, 80mC, 90jA, 90kA, 90lA, 90mA, 90jB, 90kB, 90lB, 90mB End 17,117 Fixed surface 30 Piezoelectric 35 Piezoelectric film 38 Main Surface 40,140 Intervening layer 51,52 Bonding layer 61,62 electrode 71,711,712,713,714,715,716,717,718,719,71 10 ,71 11 ,71 12 ,71 13 ,71 14 ,71 15 ,71 16 ,71 17 ,71 18 ,71 19 ,71 20 ,71 21 ,71 22 ,71 23 ,71 24 ,71 25 Reference position 72 parameter sets 73 Prior estimates 74A, 74B control points 80 Structures 80A, 80B, 80C Wall 85 beds 86 Space 87,88 Opening 95 Desk 95t top plate 110 Control device 111A, 111B Amplifier 130, 130A, 130B, 140A, 140B, 150, 150A, 150B, 550A, 550B Microphones 135A,135B,162 Update section 160 Position estimation part 161 Adaptive Filter 163 Subtraction section 164 Correction Unit 164a Convolution section 164b Subtraction section 170 memory 171 Filter holder 172 Location information storage unit 180 processors 190 Selector 191 Error calculation section 192 Minimum value discrimination unit 200 Noise Source 200w0,200w1,200w2 wave front 500 ANC System 560 target range 600,700,800 Experimental placement D1,D2,D3,DvA,DhB,DtA,DtB,DlA,DlB,DsA,DsB direction O Reference axis

Claims

1. a reference microphone unit that detects sound and generates a reference signal, the reference microphone unit including one or more reference microphones; a specific microphone unit that detects sound and generates a specific signal, the specific microphone unit including one or more specific microphones; a first speaker that radiates a first sound wave; A control device that determines a control filter that defines a response characteristic of the first sound wave to the reference signal based on the reference signal and the specific signal, and realizes active noise control by radiating the first sound wave from the first speaker based on the reference signal and the control filter. Active noise control system.

2. The control device determines the control filter based on a function of the reference signal and the specific signal.

10. The active noise control system of claim 1.

3. The control device determines the control filter based on a ratio of the reference signal and the specific signal.

10. The active noise control system of claim 1.

4. The determination of the control filter is not a process of converging the filter coefficient of the control filter by an adaptive algorithm.

10. The active noise control system of claim 1.

5. The control filter associates the reference signal with the first sound wave in a one-to-one correspondence.

10. The active noise control system of claim 1.

6. The control device includes: A memory that stores a plurality of response characteristic filters that define the response characteristics different from each other; a selector that selects A response characteristic filters from the plurality of response characteristic filters based on the reference signal and the specific signal, where A is a natural number equal to or greater than 1; and a processor for implementing the active noise control using the control filter based on the A response characteristic filters.

10. The active noise control system of claim 1.

7. the plurality of response characteristic filters are associated one-to-one with a plurality of corresponding values; each of the plurality of response characteristic filters is capable of realizing the active noise control when a function of the reference signal and the specific signal takes the corresponding value corresponding to the response characteristic filter; the A response characteristic filters correspond to A corresponding values ​​selected from the plurality of corresponding values ​​in order from closest to the value of the function; 7. An active noise control system according to claim 6.

8. The control device has a trained machine learning model, The machine learning model determines the control filter based on the reference signal and the specific signal.

10. The active noise control system of claim 1.

9. a second speaker that emits a second sound wave; the control device realizes the active noise control by causing the first speaker to radiate the first sound wave and the second speaker to radiate the second sound wave; the first speaker has a first radiation surface that radiates the first sound wave; the second speaker has a second radiation surface that radiates the second sound wave; At least a portion of the first radiation surface and at least a portion of the second radiation surface are located at different positions from each other in a predetermined direction.

10. The active noise control system of claim 1.

10. the one or more reference microphones include a first reference microphone; The one or more specific microphones include a first specific microphone, The first reference microphone and the first specific microphone are arranged in a direction different from the predetermined direction.

10. The active noise control system of claim 9.

11. The first wall, a second wall opposite the first wall; and a space located between the first wall and the second wall, In the active noise control, the control device causes the first speaker to radiate the first sound waves away from the space, thereby reducing noise from inside the space outside the space.

10. The active noise control system of claim 1.

12. the control device includes a correction unit that corrects the specific signal in response to the reference signal by processing using a correction filter; The control device determines the control filter based on the reference signal and the corrected specific signal.

10. The active noise control system of claim 1.

13. Detecting a sound and generating a reference signal; determining a response characteristic of a first sound wave to the reference signal in response to a position of a noise source; and emitting the first sound wave based on the reference signal and the response characteristic to realize active noise control. Active noise control method.

14. selecting the response characteristic from a plurality of different reference characteristics based on a location of the noise source.

14. The method of active noise control according to claim 13.

15. the plurality of reference characteristics being in one-to-one correspondence with a plurality of reference locations; Each of the plurality of reference characteristics is capable of realizing the active noise control when the noise source is located at the reference position corresponding to the reference characteristic, selecting, from the plurality of reference characteristics, a reference characteristic corresponding to a reference position among the plurality of reference positions that is closest to a position of the noise source as the response characteristic; 15. The method of active noise control according to claim 14.

16. determining the response characteristics according to the location of the noise source by a trained machine learning model; 14. The method of active noise control according to claim 13.

17. The direction of movement of the noise source that causes the change in the response characteristic includes a first direction and a second direction different from the first direction.

14. The method of active noise control according to claim 13.

18. and achieving the active noise control by radiating the first sound wave from a first radiating surface of a first speaker and radiating the second sound wave from a second radiating surface of a second speaker, wherein at least a portion of the first radiating surface and at least a portion of the second radiating surface are at different positions from each other in a predetermined direction; The direction of movement of the noise source that causes the change in the response characteristic includes a direction different from the predetermined direction.

14. The method of active noise control according to claim 13.

Images