Active vibration noise reduction device

Abstract

To provide an active vibration noise reduction device that can reduce the noise of a plurality of control points (microphones) to stably reduce noise in a wide range.SOLUTION: An active vibration noise reduction device 100 (200) comprises: speakers 20, 21 that output cancellation sounds y11, y12, y21, y22 for cancelling noises d1, d2; microphones 30, 31 that generate error signals e1, e2 from the noises d1, d2 and the cancellation sounds y11, y12, y21, y22; and a control filter W that generates a control signal u1 for controlling the cancellation sounds y11, y12, y21, y22 on the basis of the error signals e1, e2. The microphones 30, 31 are provided in plurality. The control filter W is adaptively updated with the square sum of the sound pressure of the error signals e1, e2 generated by the plurality of microphones 30, 31 as an evaluation function.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an active vibration noise reduction device. [Background technology]

[0002] Conventionally, active noise reduction devices have been considered for noise (e.g., road noise) generated inside a vehicle cabin, which reduce noise by generating a canceling sound that is in the opposite phase to the noise and causing the generated canceling sound to interfere with the noise.

[0003] For example, the abstract of the noise control device disclosed in Patent Document 1 states that "the noise control device is provided with a plurality of microphones that detect residual sounds and output the residual sounds as error signals to a plurality of adaptive filters, a plurality of actuators that reproduce compensation signals from the plurality of adaptive filters and cancel noise to form residual sounds, and a plurality of error signal mixing means that divide the plurality of microphones into a plurality of groups, mix the error signals from the microphones in each group to form a mixed error signal, and output each mixed error signal to each adaptive filter." [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 6-59683 Summary of the Invention [Problem to be solved by the invention]

[0005] For example, when considering applying the noise control device described in Patent Document 1 to a vehicle, acoustic modes exist within the vehicle cabin. Therefore, even if signals acquired by microphones are added together as an error signal, there are cases where the added frequency signal is reinforced and cases where it cancels out. As a result, in the vehicle cabin, when the error signal is reinforced, it can be controlled, but when the error signal is canceled out, it cannot be controlled.

[0006] The present invention has been made in consideration of the above circumstances, and has as its object to provide an active noise reduction device that can stably reduce noise over a wide range by reducing noise from multiple control points (microphones). [Means for solving the problem]

[0007] That is, in order to solve the above-mentioned problems of the present invention, an active vibration noise reduction device of the present invention comprises a speaker that outputs a canceling sound to cancel out a noise, a microphone that generates an error signal from the noise and the canceling sound, and a control filter that generates a control signal for controlling the canceling sound based on the error signal, wherein a plurality of microphones are provided, and the control filter is adaptively updated using the sum of squares of the sound pressures of the error signals of the plurality of microphones as an evaluation function. [Effects of the Invention]

[0008] According to the present invention, noise at a plurality of control points (microphones) is reduced, thereby making it possible to stably reduce noise over a wide range. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a block diagram showing a schematic configuration of an active vibration noise reduction device according to an embodiment of the present invention. [Figure 2] FIG. 1 is an explanatory diagram showing an LMS algorithm for calculating filter coefficients that minimize an evaluation function. [Figure 3] This is a conceptual diagram showing that learning multiple channels does not work well. [Figure 4] FIG. 10 is an explanatory diagram showing the concept of setting different initial values ​​to the control filter unit. DETAILED DESCRIPTION OF THE INVENTION

[0010] The following describes in detail embodiments of the present invention. Note that the embodiments described below are examples for realizing the present invention, and should be appropriately modified or changed depending on the configuration of the device to which the present invention is applied and various conditions. The present invention is not limited to the following embodiments. In addition, in each drawing, the same components are given the same reference numerals, and their description will be omitted as appropriate.

[0011] In this specification, the "^" (hat) next to various symbols indicates an identified value or an estimated value. In the figures, the "^" is placed above the various symbols, but in the text, it is placed after the various symbols.

[0012] <Present Embodiment> [Outline of active vibration noise reduction device] Fig. 1 is a block diagram showing the schematic configuration of an active vibration noise reduction device according to this embodiment. The active vibration noise reduction devices 100 and 200 shown in Fig. 1 constitute an ANC device (Active Noise Control Device) for reducing noise generated inside the vehicle cabin.

[0013] Various noises occur inside the vehicle while it is moving, such as tire noise, wind noise, engine noise, etc. ANC devices are installed in vehicles to cancel out noise generated by the transmission of vibrations from the power unit (engine, motor, etc.) and the inflow of exhaust noise, resulting in a quieter vehicle and creating a comfortable, high-quality interior space.

[0014] That is, the active vibration noise reduction devices 100, 200 generate canceling sounds y11, y12, y21, y22 that are in opposite phase to the noises d1, d2 generated by the noise sources, and reduce the noises d1, d2 by having the generated canceling sounds y11, y12, y21, y22 interfere with the noises d1, d2. The noises d1, d2 correspond to, for example, road noise caused by wheel vibrations due to forces from the road surface. Note that road noise is one example of the noises d1, d2, and the noises d1, d2 may also be noises other than road noise, such as drivetrain noise caused by vibrations of a drive source such as an internal combustion engine or an electric motor.

[0015] 1, the active vibration noise reduction apparatus 100 according to this embodiment is configured to include a noise control section 110, a speaker 20, microphones 30 and 31, and a sound field learning section 140. The active vibration noise reduction apparatus 200 has the same configuration as the active vibration noise reduction apparatus 100, with the noise control section 210 corresponding to the noise control section 110 and the sound field learning section 240 corresponding to the sound field learning section 140. The speaker 21 is connected to the active vibration noise reduction apparatus 200.

[0016] 1 indicates a noise transfer path, and indicates the transfer function of the primary path from the noise source to the microphone 30. Also, transfer function P2 in FIG. 1 indicates a noise transfer path, and indicates the transfer function of the primary path from the noise source to the microphone 31.

[0017] 1 indicates the transfer function of the secondary path from the speaker 20 to the microphone 30, and transfer function C12 indicates the transfer function of the secondary path from the speaker 20 to the microphone 31. Also, transfer function C21 in FIG. 1 indicates the transfer function of the secondary path from the speaker 21 to the microphone 30, and transfer function C22 indicates the transfer function of the secondary path from the speaker 21 to the microphone 31.

[0018] The speaker 20 outputs canceling sounds y11 and y12 to cancel out the noises d1 and d2. The speaker 20 is provided, for example, in front of the driver's seat or in a door on the side of the passenger seat. The speaker 21 outputs canceling sounds y21 and y22 for canceling out the noises d1 and d2. The speaker 21 is provided, for example, in front of the passenger seat or in a door on the side of the passenger seat.

[0019] The microphones 30 and 31 generate error signals e1 and e2 from the noises d1 and d2 and the canceling sounds y11, y12, y21, and y22. The microphone 30 is provided, for example, on the headrest of the driver's seat. The microphone 30 generates the error signal e1 based on the canceling sound y11 output by the speaker 20, the canceling sound y21 output by the speaker 21, and the noise d1 at the position of the microphone 30.

[0020] On the other hand, the microphone 31 is provided, for example, on the headrest of the passenger seat. The microphone 31 generates an error signal e2 based on the canceling sound y12 output by the speaker 20, the canceling sound y22 output by the speaker 21, and the noise d2 at the position of the microphone 31.

[0021] The noise control unit 110 and the sound field learning unit 140 are configured, for example, by a computer having an arithmetic processing unit (a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit)) and a storage device (memory such as a ROM (Read Only Memory) or RAM (Random Access Memory)). In other words, the components of the active vibration noise reduction device 100 other than the speaker 20 and the microphones 30, 31 may be configured, for example, as a single piece of hardware, or may be configured as a unit consisting of multiple pieces of hardware.

[0022] A reference signal r corresponding to the noises d1 and d2 is input to the noise control unit 110. The reference signal r is input to the noise control unit 110, for example, from a reference microphone (not shown) that generates the reference signal r from the noises d1 and d2. The noise control unit 110 is configured to include a control filter unit 111, a first secondary path filter unit 112, a second secondary path filter unit 113, and a control update unit 114.

[0023] The control filter unit 111 generates a control signal u1 for controlling the canceling sounds y11 and y12 from the reference signal r. The control signal u1 controls the canceling sounds y11 and y12, thereby canceling out the noises d1 and d2. The control filter unit 111 is also composed of a control filter W. The control filter W is, for example, an FIR (Finite Impulse Response) filter. An FIR filter is a type of digital filter, and is a filter whose impulse response has a finite duration. In other words, an FIR filter is a filter whose output signal (impulse response) converges within a finite time when an impulse signal is input.

[0024] The control filter unit 111 performs filtering on the reference signal r using the control filter W to generate a control signal u1 for controlling the speaker 20. The control filter unit 111 inputs the generated control signal u1 to the speaker 20. The speaker 20 generates cancellation sounds y11 and y12 in accordance with the control signal u1 generated by the control filter unit 111. The control filter unit 111 also inputs the generated control signal u1 to the sound field learning unit 140.

[0025] The first secondary path filter unit 112 is configured by a secondary path filter C^11 that indicates an estimated value of a transfer function C11 from the speaker 20 to the microphone 30. The secondary path filter C^11 is a filter that indicates an estimated value of the transfer function C11 of the secondary path. The secondary path filter C^11 is configured by, for example, an FIR filter.

[0026] The first secondary path filter unit 112 corrects the reference signal r by filtering the reference signal r using the secondary path filter C^11. The first secondary path filter unit 112 inputs the corrected reference signal r to the control update unit 114.

[0027] The second secondary path filter unit 113 is configured by a secondary path filter C^12 that indicates an estimated value of a transfer function C12 from the speaker 20 to the microphone 31. The secondary path filter C^12 is a filter that indicates an estimated value of the transfer function C12 of the secondary path. The secondary path filter C^12 is configured by, for example, an FIR filter.

[0028] The second secondary path filter unit 113 corrects the reference signal r by filtering the reference signal r using the secondary path filter C^12. The second secondary path filter unit 113 inputs the corrected reference signal r to the control update unit 114.

[0029] The control update unit 114 adaptively updates the control filter W of the control filter unit 111 using an adaptive algorithm such as an LMS algorithm (Least Mean Square Algorithm). Specifically, the control update unit 114 adaptively updates the filter coefficients of the control filter W so that the error signals e1 and e2 output from the microphones 30 and 31 are minimized. The control update unit 114 adds up the error signals e1 and e2 and adaptively updates the sum so that the sum is minimized. Note that the filter coefficients of the control filter W are adaptively updated, whereby the control filter W is adaptively updated.

[0030] The sound field learning unit 140 is configured to include a first canceling estimated signal generation unit 141, a first secondary path update unit 142, a second canceling estimated signal generation unit 143, and a second secondary path update unit 144. The sound field learning unit 140 is also configured to include a first noise estimation signal generation unit 145, a primary path update unit 146, a second noise estimation signal generation unit 147, and a primary path update unit 148. The sound field learning unit 140 is also configured to include virtual error signal generation units 149 and 150.

[0031] The first cancelling estimation signal generation unit 141 is configured with a secondary path filter C^11. The secondary path filter C^11 of the first cancelling estimation signal generation unit 141 is a filter that indicates an estimated value of the transfer function C11 of the secondary path, similar to the secondary path filter C^11 of the first secondary path filter unit 112. The secondary path filter C^11 of the first cancelling estimation signal generation unit 141 is configured with, for example, an FIR filter.

[0032] The first cancellation estimation signal generation unit 141 generates a cancellation estimation signal y^11 indicating an estimate of the cancellation y11 by filtering the control signal u1 input from the control filter unit 111 of the noise control unit 110 with the secondary path filter C^11. The first cancellation estimation signal generation unit 141 inputs the generated cancellation estimation signal y^11 to the virtual error signal generation unit 149.

[0033] The first secondary path update unit 142 uses an adaptive algorithm such as an LMS algorithm to adaptively update the secondary path filter C^11 of the first cancellation estimation signal generation unit 141. Specifically, the first secondary path update unit 142 adaptively updates the secondary path filter C^11 so that the virtual error signal ev1 input from the virtual error signal generation unit 149 is minimized.

[0034] The second cancellation estimation signal generation unit 143 is configured with a secondary path filter C^12. The secondary path filter C^12 of the second cancellation estimation signal generation unit 143 is a filter that indicates an estimated value of the transfer function C12 of the secondary path, similar to the secondary path filter C^12 of the second secondary path filter unit 113. The secondary path filter C^12 of the second cancellation estimation signal generation unit 143 is configured with, for example, an FIR filter.

[0035] The second cancellation estimation signal generation unit 143 generates a cancellation estimation signal y^12 indicating an estimate of the cancellation y12 by filtering the control signal u1 input from the control filter unit 111 of the noise control unit 110 with the secondary path filter C^12. The second cancellation estimation signal generation unit 143 inputs the generated cancellation estimation signal y^12 to the virtual error signal generation unit 150.

[0036] The second secondary path update unit 144 uses an adaptive algorithm such as an LMS algorithm to adaptively update the secondary path filter C^12 of the second cancellation estimation signal generation unit 143. Specifically, the second secondary path update unit 144 adaptively updates the secondary path filter C^12 so that the virtual error signal ev2 input from the virtual error signal generation unit 150 is minimized.

[0037] The first noise estimation signal generator 145 is configured by a primary path filter P^1. The primary path filter P^1 is a filter that indicates an estimated value of a transfer function P1 of the primary path. The primary path filter P^1 is configured by, for example, an FIR filter.

[0038] The first noise estimation signal generator 145 filters the reference signal r using the primary path filter P^1 to generate a noise estimation signal d^1 indicating an estimate of the noise d1. The first noise estimation signal generator 145 inputs the generated noise estimation signal d^1 to the virtual error signal generator 149.

[0039] The first primary path updater 146 uses an adaptive algorithm such as an LMS algorithm to adaptively update the primary path filter P^1 of the first noise estimation signal generator 145. Specifically, the first primary path updater 146 adaptively updates the primary path filter P^1 so that the virtual error signal ev1 input from the virtual error signal generator 149 is minimized.

[0040] The second noise estimation signal generator 147 is configured by a primary path filter P^2. The primary path filter P^2 is a filter that indicates an estimated value of a transfer function P2 of the primary path. The primary path filter P^2 is configured by, for example, an FIR filter.

[0041] The second noise estimation signal generator 147 generates a noise estimation signal d^2 indicating an estimate of the noise d2 by filtering the reference signal r with the primary path filter P^2. The second noise estimation signal generator 147 inputs the generated noise estimation signal d^2 to the virtual error signal generator 150.

[0042] The second primary path update unit 148 uses an adaptive algorithm such as an LMS algorithm to adaptively update the primary path filter P^2 of the second noise estimation signal generation unit 147. Specifically, the second primary path update unit 148 adaptively updates the primary path filter P^2 so that the virtual error signal ev2 input from the virtual error signal generation unit 150 is minimized.

[0043] The virtual error signal generator 149 is composed of an adder. The virtual error signal generator 149 generates a virtual error signal ev1 by adding together the error signal e1 input from the microphone 30, the cancellation estimation signal y^11 input from the first cancellation estimation signal generator 141, the noise estimation signal d^1 input from the first noise estimation signal generator 145, and the cancellation estimation signal y^21. The virtual error signal generator 149 inputs the generated virtual error signal ev1 to the first secondary path update unit 142 and the primary path update unit 146. The cancellation estimation signal y^21 is a cancellation estimation signal indicating an estimated value of the cancellation y21, which is generated in the sound field learning unit 240 of the active vibration noise reduction device 200 in the same manner as the first cancellation estimation signal generator 141.

[0044] The virtual error signal generation unit 150 is composed of an adder. The virtual error signal generation unit 150 generates a virtual error signal ev2 by adding together the error signal e2 input from the microphone 31, the cancellation estimation signal y^12 input from the second cancellation estimation signal generation unit 143, the noise estimation signal d^2 input from the second noise estimation signal generation unit 147, and the cancellation estimation signal y^22. The virtual error signal generation unit 150 inputs the generated virtual error signal ev2 to the second secondary path update unit 144 and the primary path update unit 148. The cancellation estimation signal y^22 is a cancellation estimation signal indicating an estimated value of the cancellation y22, which is generated in the sound field learning unit 240 of the active vibration noise reduction device 200 in the same manner as the second cancellation estimation signal generation unit 143.

[0045] [Renewal of active noise and vibration reduction devices] Next, a description will be given of the update processing of the active vibration noise reduction apparatus 100 according to this embodiment. The update processing of the active vibration noise reduction apparatus 100 will be described with reference to FIGS.

[0046] 2 is an explanatory diagram showing an LMS algorithm for calculating filter coefficients that minimize an evaluation function. In this embodiment, the control filter W is updated by updating the filter coefficients.

[0047] In the algorithm shown in Figure 2, the evaluation function J (e.g., 2 ) is minimized, the minimum value is searched for along the negative direction of the gradient of the evaluation function J. When the evaluation function J is minimized, the update amount ΔW becomes 0. Also, Figure 2 shows that when the evaluation function J is at its minimum, the noises d1 and d2 and the error signals e1 and e2 are minimized.

[0048] In this embodiment, the control filter W of the control filter unit 111 is adaptively updated using the sum of squares of the sound pressures of the error signals e1 and e2 of the microphones 30 and 31 as the evaluation function J.

[0049] In acoustic power control, if the sum of squares of sound pressures at the positions of the microphones 30 and 31 (also called control points) is taken as an evaluation function J, the evaluation function J is calculated by the following equations (1) to (4).

[0050]

number

[0051]

number

[0052]

number

[0053]

number

[0054] In addition, this embodiment is provided with secondary path filters C^11, C^12 that indicate estimated values ​​of transfer functions from each speaker 20, 21 to multiple microphones 30, 31, and the control filter W of the control filter unit 111 is adaptively updated based on the multiple secondary path filters C^11, C^12.

[0055] Therefore, based on the evaluation function J, the control filter W is adaptively updated by the update equations of Equations (5) and (6).

[0056]

number

[0057]

number

[0058] Here, the secondary path filters C^11 and C^12 that constitute the control filter W are updated by the following equations (7) to (11).

[0059]

number

[0060]

number

[0061]

number

[0062]

number

[0063]

number

[0064] In order to calculate the value that minimizes the evaluation function J in FIG. 2 using the above-mentioned equations (1) to (4), the following equation (12) is calculated by partially differentiating equation (1) with respect to the control filter W, which is an unknown.

[0065]

number

[0066] From equation (12), equation (5), which is an update equation for updating the control filter W along the negative gradient of the evaluation function J, can be derived.

[0067] As explained above, the active vibration noise reduction device 100 according to this embodiment is configured to include the speaker 20, the microphones 30, 31, and the control filter W. A plurality of microphones 30, 31 are provided, and the control filter W is adaptively updated using the sum of squares of the sound pressures of the error signals e1, e2 of the plurality of microphones 30, 31 as the evaluation function J, as shown in equation (1).

[0068] With this configuration, the active vibration noise reduction device 100 adaptively updates the filter coefficients of the control filter W so that multiple error signals e1, e2 are simultaneously reduced, as shown in FIG. 2, and is therefore able to reduce a wide range of noises d1, d2 without being affected by frequency.

[0069] Therefore, the active vibration noise reduction device 100 according to this embodiment can perform simultaneous control at a plurality of control points (microphones 30, 31) even in a closed space such as the interior of a vehicle where acoustic modes exist.

[0070] Furthermore, even if the user adjusts the seat position, the control filter W is updated in accordance with the change in the seat position (microphones 30, 31), so the noises d1 and d2 can be maintained in a reduced state. In this case, the active vibration noise reduction device 100 also updates the secondary path filters C^11 and C^12 by the sound field learning section 140 in accordance with the change in the seat position (microphones 30, 31), so the noise can be maintained in a reduced state.

[0071] The active vibration noise reduction device 100 is also configured to include secondary path filters C^11 and C^12 that indicate estimated values ​​of transfer functions C11 and C12 from the speaker 20 to the multiple microphones 30 and 31. For example, the control filter W is adaptively updated based on the multiple secondary path filters C^11 and C^12, as shown in equation (6).

[0072] According to this configuration, the control filter W of the control filter unit 111 can be adaptively updated more reliably by adaptively updating the multiple secondary path filters C^11 and C^12 as shown in equation (6).

[0073] In addition, each of the multiple secondary path filters C^11, C^12 may be adaptively updated based on virtual error signals ev1, ev2 (virtual error signals) calculated from the error signals e1, e2 of the multiple microphones 30, 31 corresponding to the cancellation sounds y11, y12, y21, y22.

[0074] According to this configuration, the secondary path filters C^11 and C^12 are adaptively updated by calculating virtual error signals ev1 and ev2 calculated from the error signals e1 and e2 as shown in equations (7) and (8). As a result, the active vibration noise reduction device 100 can enhance the noise reduction effect by adaptively updating the secondary path filters C^11 and C^12 of the first canceling estimated signal generator 141 and the second canceling estimated signal generator 143.

[0075] Furthermore, the active vibration noise reduction device 100 is provided with a speaker 20, and the active vibration noise reduction device 200 is provided with a speaker 21. In this way, secondary path filters C^11, C^12, C^21, C^22 may be provided between each of the multiple speakers 20, 21 and each of the multiple microphones 30, 31. In this case, each of the secondary path filters C^11, C^12, C^21, C^22 is adaptively updated based on the secondary path filters C^11, C^12, C^21, C^22 of the multiple microphones 30, 31 corresponding to the multiple speakers 20, 21.

[0076] According to this configuration, the active vibration noise reduction device 100 can adaptively update each of the secondary path filters C^11, C^12, C^21, C^22 based on the secondary path filters C^11, C^12, C^21, C^22 of the multiple microphones 30, 31 corresponding to the multiple speakers 20, 21, as shown in equations (8) and (11), thereby more accurately updating each of the secondary path filters C^11, C^12, C^21, C^22.

[0077] It is desirable that the secondary path filters C^11, C^12, C^21, and C^22 are adaptively updated based on, for example, the secondary path filters C^11, C^12, C^21, and C^22 of the multiple microphones 30 and 31 corresponding to all of the speakers 20 and 21. By adaptively updating the secondary path filters C^11, C^12, C^21, and C^22 based on the secondary path filters C^11, C^12, C^21, and C^22 of the multiple microphones 30 and 31 corresponding to all of the speakers 20 and 21, the accuracy of sound reduction can be further improved and noise can be reduced over a wide range.

[0078] As described above, the active vibration noise reduction device 100 is provided with the speaker 20, and the active vibration noise reduction device 200 is provided with the speaker 21. In this case, the active vibration noise reduction devices 100, 200 are each provided with a control filter W for each speaker 20, 21, and the initial values ​​of each control filter W may be different.

[0079] According to this configuration, each control update unit 114 sets an initial value for each control filter W so as to shift the initial output timing.

[0080] Here, for example, in the case of reducing road noise inside the vehicle cabin using vehicle body vibrations detected by an acceleration sensor as in Patent Documents 1 and 2, if the same initial value is set among multiple control filters W, the control signal of one active vibration noise reduction device will be the same as the control signal of another active vibration noise reduction device.

[0081] Therefore, all of the secondary path filters C^11, C^12, C^21, and C^22 have the same value, the update formula of equation (5) also has the same value, and the update values ​​of the control filters W all have the same value.

[0082] That is, as shown using equations (13) to (18) below, when the same initial value is set for control filter W1 (used in equations such as equation (17)) of control filter section 111 of active vibration noise reduction apparatus 100 and control filter W2 (used in equations such as equation (18)) of the control filter section of active vibration noise reduction apparatus 200, the control signal u1 of active vibration noise reduction apparatus 100 and the control signal u2 of active vibration noise reduction apparatus 200 will initially have the same value. Therefore, even if control filters W1 and W2 of active vibration noise reduction apparatuses 100 and 200 are updated, the secondary path filters C^11, C^12, C^21, and C^22 will each have the same value.

[0083]

number

[0084]

number

[0085]

number

[0086]

number

[0087]

number

[0088]

number

[0089] Therefore, as shown in equations (19) to (24), even if the control filter W1 of the active vibration noise reduction device 100 and the control filter W2 of the active vibration noise reduction device 200 are updated by rc11, rc12, rc21, and rc22 shown in equation (6) calculated using the secondary path filters C^11, C^12, C^21, and C^22, which have the same values, the values ​​will ultimately remain the same.

[0090]

number

[0091]

number

[0092]

number

[0093]

number

[0094]

number

[0095]

number

[0096] As a result, in the next control cycle, the control signal u1 of the active vibration noise reduction apparatus 100 and the control signal u2 of the active vibration noise reduction apparatus 200 remain the same value. In other words, the two speakers 20 and 21 output the same canceling sounds y11, y12, y21, and y22.

[0097] In this way, even if the control filter W of the active vibration noise reduction device 100 and the control filter W of the active vibration noise reduction device 200 are updated, the filter coefficients will end up being the same values, and the same canceling sound will be output from the two speakers 20, 21.

[0098] 3 is a conceptual diagram showing that learning of a large number of channels (multiple speakers 20, 21) does not go well. As shown in FIG. 3, the control signal u1 generated by the control filter W of the control filter unit 111 has the same value as the control signal u2 generated by the control filter W of the control filter unit 211. In this case, the secondary path filter C^1 of the secondary path filter unit 115 has the same value as the secondary path filter C^2 of the secondary path filter unit 215.

[0099] As a result, the control filter W of the control filter unit 111 has the same effect as driving two speakers 20 in parallel with only one control signal u1, and essentially becomes one control channel.

[0100] Therefore, it is conceivable that the active vibration noise reduction devices 100, 200 create uncorrelated control sounds (canceling sounds) and adjust the output timing by setting the initial values ​​of the control filters W of the control filter sections 111, 211 to different values.

[0101] 4 is an explanatory diagram showing the concept of setting different initial values ​​to the control filter units. As shown in FIG. 4, initial values ​​are set to the control filter units 111 and 211 at different times.

[0102] The control update unit 114 (FIG. 1) sets an initial value for the control filter unit 111 at a predetermined timing t, and the control update unit 214 (FIG. 3) sets an initial value for the control filter unit 211 at the predetermined timing t with a delay of ΔT.

[0103] As a result, the outputs of control filter unit 111 and control filter unit 211 become signals that are shifted by a time ΔT, as shown in Fig. 4. As a result, road noise becomes broadband noise, so by shifting the timing of the initial value, it is possible to create an uncorrelated output.

[0104] For example, an initial value α is set for the i-th element of the i-th control filter unit 111, and the other elements are set to zero.

[0105]

number

[0106] Similarly, the initial value α is set for the j-th element of the i-th control filter unit 211, which is different from the i-th element, and the other elements are set to 0.

[0107]

number

[0108] According to equations (25) and (26), the control filter W of the first control filter unit 111 first outputs a control signal u1, then the control filter W of the second control filter unit 211 outputs a control signal u2, and finally the Sth control filter unit outputs a control signal. In this way, by shifting the output timing of each control channel, the control signals u1 and u2 can be made different.

[0109] 3 can be trained, so that the control filter unit 111 and the control filter unit 211 converge to the correct values. Therefore, the active vibration noise reduction devices 100 and 200 can correctly train the secondary path filters C^1 and C^2 of the secondary path filter units 115 and 215 even in the case of a large number of channels.

[0110] In this way, the control update units 114, 214 set the initial value α for the control filter W of the control filter units 111, 211 so as to shift the initial output timing by ΔT, thereby preventing the same control sound (canceling sound) from being output from the two speakers 20, 21.

[0111] In addition, the active vibration noise reduction device 100 may be provided with primary path filters P^1 and P^2, which are transfer functions P1 and P2 from the noise source to the multiple microphones 30 and 31, respectively, and the common primary path filters P^1 and P^2 may be adaptively updated for control of the multiple speakers 20 and 21.

[0112] According to this configuration, by using equation (9), the multiple outputs (noises d1, d2) reaching the multiple microphones 30, 31 from the noise source are made common, thereby reducing the amount of calculation in the primary path filter units P^1, P^2.

[0113] In addition, the control filter W of the active vibration noise reduction device 100 generates a control signal u1 for controlling the cancellation sounds y11, y12 from the reference signal r, and may also be updated for each of the multiple microphones 30, 31 by the sum of values ​​based on update amounts normalized by the reference signal r and the secondary path filters C^11, C^12 corresponding to the microphones 30, 31.

[0114] The active vibration noise reduction device 100 requires a large number of filters to be learned during control using sound field learning acoustic power control. In this case, noise reduction effects cannot be achieved unless all filters (primary path filters P^1, P^2, secondary path filters C^11, C^12) converge. Therefore, it is desirable for the active vibration noise reduction device 100 to improve the learning speed of all filters.

[0115] Therefore, the calculation formula for normalizing the update amount of each filter coefficient based on the input signal will be explained using formulas (27) to (29).

[0116]

number

[0117]

number

[0118]

number

[0119] With this configuration, the active vibration noise reduction device 100 can improve the convergence speed by normalizing the update amount of each filter coefficient by the norm of the signal vector based on the input signal, as shown in equations (27) to (29). Furthermore, the active vibration noise reduction device 100 normalizes each of the terms in the summation calculation while updating the control filter W. In particular, since the learning amount for each of the microphones 30 and 31 is normalized as shown in equation (27), the convergence speed can be improved.

[0120] In this way, the active vibration noise reduction device 100 can improve the convergence speed of the entire control by normalizing the update amount of each of the multiple microphones 30, 31.

[0121] In the description of this embodiment, FIR filters are used for the control filter W, the primary path filters P^1 and P^2, and the secondary path filters C^11, C^12, C^21, and C^22, but this embodiment is not limited to FIR filters, and other filters (for example, single-frequency adaptive notch filters) can be applied as appropriate. [Explanation of symbols]

[0122] 110,210 Noise control section 111 Control filter section 112 First secondary path filter unit 113 Secondary path filter unit 114 Control Update Unit 20,21 Speaker 30,31 Mike 140,240 Sound field learning section 141 First cancellation noise estimation signal generation unit 143 Second cancellation noise estimation signal generation unit 142 First secondary route update unit 144 Secondary Route Update Unit 145 First noise estimation signal generation unit 147 Second noise estimation signal generator 146 First Primary Route Update Unit 148 Second Primary Route Update Unit 149,150 Virtual error signal generator 100,200 Active vibration and noise reduction device C^11,C^12,C^21,C^22 secondary path filters P^1,P^2 primary path filter

Claims

1. a speaker that outputs a canceling sound to cancel out noise; a microphone for generating an error signal from the noise and the cancellation sound; a control filter that generates a control signal for controlling the cancellation based on the error signal; The microphone is provided in plurality, The control filter The sum of squares of the sound pressures of the error signals of the plurality of microphones is adaptively updated as an evaluation function. An active vibration noise reduction device characterized by:

2. a secondary path filter that provides an estimate of a transfer function from the speaker to the plurality of microphones; The control filter adaptively updated based on a plurality of said secondary path filters; 2. An active vibration noise reduction device according to claim 1.

3. Each of the plurality of secondary path filters comprises: adaptively updated based on a virtual error signal calculated from the microphone error signal corresponding to the cancellation sound; 3. An active vibration noise reduction device according to claim 2.

4. The speaker is provided in plurality, a secondary path filter is provided between each of the plurality of speakers and each of the plurality of microphones; Each of the secondary path filters comprises: adaptively updated based on the secondary path filters of the plurality of speakers and the corresponding microphones; 4. An active vibration noise reduction device according to claim 3.

5. The control filter is provided for each of the speakers, The initial values ​​of the control filters are different from each other.

4. An active vibration noise reduction device according to claim 3.

6. a first-order path filter that indicates an estimate of a transfer function from a noise source to the plurality of microphones; adaptively updating the common primary path filter for controlling the plurality of speakers; 5. An active vibration noise reduction device according to claim 4.

7. The control filter A control signal for controlling the cancellation is generated from a reference signal, and the control signal for each of the plurality of microphones is updated with a sum of a value based on the reference signal and an update amount normalized by the secondary path filter corresponding to the microphone.

3. An active vibration noise reduction device according to claim 2.

Images