Setting method, manufacturing method of active noise reduction device, program, and information terminal

Abstract

This provides a setting method that can improve the stability of noise reduction performance. [Solution] The setting method includes step S11 of acquiring multiple sets of measured acoustic transfer function data from the speaker position to the microphone position; step S12 of identifying a frequency band in which the measurement variability of the acquired multiple sets of measured data is greater than a threshold; and step S13 of storing setting information in the storage unit 16 of the active noise reduction device 10 so as to reflect in the active noise reduction device 10 a setting that does not output a cancellation sound even if noise corresponding to the identified frequency band is detected.

Description

Technical Field

[0001] The present disclosure relates to an active noise reduction device that actively reduces noise.

Background Art

[0002] Patent Document 1 discloses a technique related to the estimation of the secondary path transfer function in active noise control.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The present disclosure provides a setting method capable of enhancing the stability of noise reduction performance.

Means for Solving the Problems

[0005] A setting method according to an aspect of the present disclosure is a setting method for an active noise reduction device executed by a computer. The active noise reduction device reduces noise detected when detecting noise generated by the rotation of a power source included in a mobile device in a space provided with a speaker and a microphone in the mobile device by outputting a cancellation sound from the speaker. The setting method includes: a step of acquiring measurement data for a plurality of times of the acoustic transfer function from the position of the speaker to the position of the microphone; a step of specifying a frequency band in which the measurement variation of the acquired measurement data for the plurality of times is larger than a threshold value; and a step of storing, in a storage unit of the active noise reduction device, setting information for reflecting, in the active noise reduction device, a setting of not outputting the cancellation sound even when detecting noise corresponding to the specified frequency band.

Effects of the Invention

[0006] A setting method relating to one aspect of this disclosure can improve the stability of noise reduction performance. [Brief explanation of the drawing]

[0007] [Figure 1] Figure 1 is a block diagram showing the functional configuration of an active noise reduction device corresponding to the SAN algorithm. [Figure 2] Figure 2 shows the relationship between the noise signal and the cancellation signal in the SAN algorithm. [Figure 3] Figure 3 is a block diagram showing the functional configuration of an active noise reduction device corresponding to the SAN Filtered-x LMS algorithm. [Figure 4] Figure 4 shows the relationship between noise and cancellation sound in the SAN Filtered-x LMS algorithm. [Figure 5] Figure 5 is a schematic diagram of a vehicle equipped with an active noise reduction device according to an embodiment. [Figure 6] Figure 6 is a block diagram showing the functional configuration of an active noise reduction device according to an embodiment. [Figure 7] Figure 7 is a block diagram showing the functional configuration of the setting system according to the embodiment. [Figure 8] Figure 8 is a flowchart of Operation Example 1 of the setting system according to the embodiment. [Figure 9] Figure 9 shows an example of identifying exclusion bands. [Figure 10] Figure 10 is a flowchart of operation example 2 of the setting system according to the embodiment. [Figure 11] Figure 11 is a flowchart showing the details of the update process. [Figure 12] Figure 12 is a diagram illustrating the genetic algorithm (method for generating the third transfer function). [Figure 13] Figure 13 is a diagram illustrating the brute-force algorithm. [Figure 14]Figure 14 is a flowchart of operation example 3 of the setting system according to the embodiment. [Figure 15] Figure 15 is a flowchart showing the manufacturing method of an active noise reduction device according to an embodiment. [Modes for carrying out the invention]

[0008] The embodiments will be described in detail below with reference to the drawings. Note that the embodiments described below are all comprehensive or specific examples. The numerical values, shapes, materials, components, arrangement and connection configurations of components, steps, and the order of steps shown in the following embodiments are examples only and are not intended to limit the disclosure. Furthermore, components in the following embodiments that are not described in an independent claim will be described as optional components.

[0009] Furthermore, each figure is a schematic diagram and not necessarily a strictly accurate representation. Note that in each figure, substantially identical components are denoted by the same reference numerals, and redundant explanations may be omitted or simplified.

[0010] (Embodiment) [Noise signal reduction method using SAN algorithm] The following embodiment describes an active noise reduction device with improved noise reduction performance based on the SAN Filtered-x LMS algorithm. SAN stands for Single frequency Adaptive Notch filter, and LMS stands for Least Mean Square.

[0011] Before explaining the active noise reduction device according to the embodiment, a noise signal reduction method using the SAN algorithm will be explained. FIG. 1 is a block diagram showing the functional configuration of an active noise reduction device corresponding to the SAN algorithm. FIG. 2 is a diagram showing the relationship between a noise signal (sinusoidal wave signal of noise) and a canceling signal in the SAN algorithm. In the following noise signal reduction method using the SAN algorithm, the noise signal will be described assuming it is a sinusoidal wave signal of a single frequency.

[0012] In FIGS. 1 and 2, n is an integer of 0 or more, indicating the sampling number in the discrete time system. When the frequency of the noise signal to be reduced is f0 [Hz], the normalized angular frequency ω0 [rad] is expressed as in the following [Equation 1].

[0013] ω0 = 2πf0T s = 2πf0 / f s [Equation 1]

[0014] In [Equation 1], T s [sec] is the sampling period, and f s [Hz] is the sampling frequency. By using the normalized angular frequency ω0, nT representing the discrete time is represented by n. s

[0015] The sinusoidal wave signal n d (n) of the noise is expressed as in the following [Equation 2] using the normalized angular frequency ω0, amplitude R, and phase θ [rad].

[0016] n d (n) = Rsin(ω0n + θ) [Equation 2]

[0017] n d (n) is used to generate a canceling signal. Since the canceling signal y(n) has the same amplitude as n d (n) and is in the opposite phase, it is expressed as in the following [Equation 3].

[0018] y(n) = Rsin{ω0n + (θ - π)} =A(n)sin(ω0n)+B(n)cos(ω0n) [Formula 3]

[0019] A(n) and B(n) are the filter coefficients of the adaptive filter. The amplitude R of the cancellation signal y(n) is given by A(n). 2 +B(n) 2 It is expressed as the square root of θ, and the phase (θ-π) is expressed as the arctangent of B(n) / A(n). Therefore, changing the magnitudes of the filter coefficients A(n) and B(n) of the adaptive filter changes the amplitude of the cancellation signal, and changing the ratio of the filter coefficients A(n) and B(n) of the adaptive filter changes the phase of the cancellation signal.

[0020] Here, the filter coefficients A(n) and B(n) of the adaptive filter are optimized by the LMS algorithm to minimize e(n). e(n) is the error signal resulting from the interference between the noise signal and the cancellation signal. This reduces the noise signal.

[0021] [Noise reduction method using the SAN Filtered-x LMS algorithm] Next, we will explain a noise reduction method using the SAN Filtered-x LMS algorithm. Figure 3 is a block diagram showing the functional configuration of an active noise reduction device corresponding to the SAN Filtered-x LMS algorithm. Figure 4 is a diagram showing the relationship between noise and cancellation sound in the SAN Filtered-x LMS algorithm. In the following explanation of the noise reduction method using the SAN Filtered-x LMS algorithm, the noise will be assumed to be engine noise. Engine noise is a noise that is instantaneously close to a single-frequency sine wave.

[0022] The cancellation signal propagates through the speaker, the vehicle cabin, and the microphone, and is input to the active noise reduction device. This transmission path is represented by the acoustic transfer function C. m (z) is used to represent this, where z represents the z-transform. The SAN Filtered-x LMS algorithm is based on the above SAN algorithm and further incorporates the acoustic transfer function Cm This algorithm takes (z) into consideration.

[0023] In Figures 3 and 4, the simulated transfer function C m ^ (z) is the acoustic transfer function C m This is a transfer function (filter) that simulates (z). m (n) is the engine noise at the microphone position with frequency f0 [Hz]. m (n) is C m This is the discrete-time n impulse response of (z). m (n)*y(n) represents the cancellation sound at the microphone position, and * signifies the convolution operator. Note that when actually reducing engine noise, the convolution operation is a continuous-time integral, but in the following explanation, it will be described as a discrete-time sum-of-products operation.

[0024] In the noise reduction method based on the SAN Filtered-x LMS algorithm, the following processes (1) to (5) are repeatedly performed until the filter coefficients A(n) and B(n) converge to their optimal values.

[0025] (1) Based on the signal indicating the engine rotation frequency, the engine noise n m The frequency f0 [Hz] of (n) is detected.

[0026] (2) A sine wave x having a frequency of f0 [Hz] s (n) and cosine wave x c (n) is generated, multiplied by coefficients A(n) and B(n), and added together to generate the cancellation signal y(n) shown in [Equation 4].

[0027] y(n)=A(n)x s (n) + B(n)x c (n) [Formula 4]

[0028] (3) Based on the cancellation signal y(n), a cancellation sound is output from the speaker. At the microphone position, the cancellation sound c m(n)*y(n) and engine booming sound n m The residual sound (error signal) e(n) resulting from the interference of (n) is detected by a microphone.

[0029] (4)C m ^ (z) sine wave x s (n) and cosine wave x c By filtering each of (n), the sine wave r s (n) and cosine wave r c Generate (n).

[0030] (5) The filter coefficients A(n) and B(n) are updated based on the LMS update equations shown in [Equation 5] and [Equation 6]. μ is the step size parameter that determines the amount (update rate) of updates to the filter coefficients A(n) and B(n) per sample.

[0031] A(n+1) = A(n) - μr s (n)e(n)[Formula 5]

[0032] B(n+1) = B(n) - μr c (n)e(n)[Formula 6]

[0033] Here, let's add some information about engine boom noise. Engine boom noise is noise generated in the passenger compartment when vibrations from the engine's intake, compression, combustion, and exhaust processes propagate through the vehicle's chassis and other components. For example, in the case of a 4-cylinder, 4-stroke engine, two rotations of the shaft result in explosions in all four cylinders, meaning two explosions per rotation. This generates noise with a frequency component twice the engine's rotation frequency. This noise is sometimes called the second-order boom noise (second-order component) of the engine's rotation, and because the noise level of the second-order component is higher than other components, it tends to be a problem. In addition to the second-order component, harmonic components can also be problematic.

[0034] In the case of a 6-cylinder engine, the noise level of the third-order component is high, while in the case of a 3-cylinder engine, the noise level of the 1.5-order component is high. In other words, when the number of cylinders in an engine is reduced through downsizing, the frequency of the dominant engine noise becomes lower.

[0035] [Configuration of the active noise reduction device] Next, the configuration of the active noise reduction device according to the embodiment will be described. Figure 5 is a schematic diagram of a vehicle equipped with the active noise reduction device according to the embodiment. Figure 6 is a block diagram showing the functional configuration of the active noise reduction device according to the embodiment.

[0036] As shown in Figure 5, the active noise reduction device 10 is mounted on the vehicle 50 and reduces noise in the cabin space 51. The vehicle 50 may be a gasoline vehicle or a hybrid vehicle. Hybrid vehicles here include series, parallel, and split hybrid vehicles. Hybrid vehicles here also include plug-in hybrid vehicles.

[0037] A speaker 52 and a microphone 53 are installed in space 51. Note that in Figures 5 and 6, for the sake of simplicity, only one set of speaker 52 and microphone 53 is shown, but in reality, multiple sets of speaker 52 and microphone 53 are installed in space 51, and multiple sets of speaker 52 and microphone 53 are used to reduce noise.

[0038] The vehicle 50 also includes an engine 54, an engine control unit 55, and an ECU (Electronic Control Unit) 56.

[0039] The engine 54 is a power source for the vehicle 50 and a drive unit that is a source of noise in the space 51. The engine 54 is located in a space separate from the space 51, for example. Specifically, the engine 54 is installed in a space formed inside the hood of the vehicle 50.

[0040] The engine control unit 55 controls (drives) the engine 54 based on the accelerator operation of the vehicle driver 50. The engine control unit 55 also outputs a pulse signal (engine pulse signal) corresponding to the rotational speed (frequency) of the engine 54 to the active noise reduction device 10. The frequency of the pulse signal is, for example, proportional to the rotational speed (frequency) of the engine 54. Specifically, the pulse signal is an analog signal such as a so-called tachometer pulse.

[0041] The ECU 56 is a computer that electronically controls the vehicle 50. The ECU 56 outputs a digital signal indicating the rotational speed (frequency) of the engine 54 to the active noise reduction device 10.

[0042] Furthermore, the ECU56 and the active noise reduction device 10 communicate via CAN (Controller Area Network).

[0043] The active noise reduction device 10 is an active noise reduction device that reduces noise at the location where the microphone 53 is installed by using a cancellation sound output from the speaker 52. The active noise reduction device 10 is implemented by, for example, a microprocessor such as a microcontroller or a DSP (Digital Signal Processor), and a memory unit 16. Specifically, the memory unit 16 is a semiconductor memory or the like.

[0044] As shown in Figure 6, the active noise reduction device 10 specifically comprises a frequency detection unit 11, a reference signal generation unit 12, an adaptive filter 13, a correction unit 14, an update unit 15, and a storage unit 16. The reference signal generation unit 12 includes a sine wave generation unit 12a and a cosine wave generation unit 12b, and the adaptive filter 13 includes an adaptive filter 13a, an adaptive filter 13b, and an adder 13c. The correction unit 14 includes a correction unit 14a and a correction unit 14b, and the update unit 15 includes an update unit 15a and a update unit 15b. The functions of these components are realized, for example, by a microprocessor such as a DSP executing a computer program stored in the storage unit 16.

[0045] The frequency detection unit 11 acquires a signal indicating the rotational speed of the engine 54 (an analog signal output by the engine control unit 55, or a digital signal output by the ECU 56), and based on the acquired signal, detects (calculates) the (instantaneous) frequency of the engine noise. The relationship between the frequency f0 [Hz] of the engine noise, the rotational speed RPM [rpm] of the engine 54, and the order ORD of the engine noise is expressed by the following [Equation 7]. In other words, [Equation 7] is an equation for detecting the first frequency corresponding to the engine rotational speed.

[0046] f0=(RPM)(ORD) / 60[Formula 7]

[0047] The sine wave generation unit 12a generates a sine wave of the frequency detected by the frequency detection unit 11, and outputs it to the reference signal x S Output as (n). n is a non-negative integer representing the sampling number in the discrete-time system. Reference signal x S (n) is output to the adaptive filter 13a, the correction unit 14a, and the update unit 15a.

[0048] The cosine wave generation unit 12b generates a cosine wave of the frequency detected by the frequency detection unit 11, and outputs it to the reference signal x C Output as (n). Reference signal x C (n) is output to the adaptive filter 13b, the correction unit 14b, and the update unit 15b.

[0049] The adaptive filter 13a receives the reference signal x output from the sine wave generator 12a. S Multiply (n) by the filter coefficient A(n). The filter coefficient A(n) is sequentially updated by the update unit 15a. The reference signal x multiplied by the filter coefficient A(n) S (n) is the cancellation signal A(n)x S (n) is output to the adder 13c.

[0050] The adaptive filter 13b receives the reference signal x output from the cosine wave generator 12b. CMultiply (n) by the filter coefficient B(n). The filter coefficient B(n) is sequentially updated by the update unit 15b. The reference signal x multiplied by the filter coefficient B(n) C (n) is the cancellation signal B(n)x C (n) is output to the adder 13c.

[0051] The summing unit 13c receives the cancellation signal A(n)x output from the adaptive filter 13a. S (n) and the cancellation signal B(n)x output from the adaptive filter 13b. C The cancel signal y(n) is generated by adding (n). The adder 13c outputs the generated cancel signal y(n) to the speaker 52.

[0052] The correction unit 14a processes the reference signal x S (n) simulates the transfer function C m The corrected reference signal r is corrected (filtered) using ^(z). S Generate (n). The corrected reference signal r generated S (n) is output to the update unit 15a.

[0053] Note that the simulated transfer function C m ^(z) is the acoustic transfer function C from the position of speaker 52 to the position of microphone 53. m This is a simulated transfer function of (z). Simulated transfer function C m ^(z) specifically represents the gain and phase (phase lag) for each frequency. Simulated transfer function C m ^(z) is, for example, measured in advance in space for each frequency and stored in the memory unit 16 of the active noise reduction device 10. In other words, the memory unit 16 stores the frequency and the gain and phase for correcting the signal of that frequency.

[0054] The correction unit 14b processes the reference signal x C (n) simulates the transfer function C m The corrected reference signal r is corrected (filtered) using ^(z). C Generate (n). The corrected reference signal r generated C(n) is output to the update unit 15b.

[0055] The update unit 15a receives the corrected reference signal r from the correction unit 14a. S Based on (n) and the error signal e(n) output by the microphone 53, the filter coefficient A(n) is calculated and the calculated filter coefficient A(n) is output to the adaptive filter 13a. The update unit 15a also sequentially updates the filter coefficient A(n) using the above [Equation 5]. In other words, the update unit 15a updates the filter coefficient A(n) using an update formula that includes the step size parameter μ (a parameter related to the update speed).

[0056] The update unit 15b receives the corrected reference signal r from the correction unit 14b. C Based on (n) and the error signal e(n) output by the microphone 53, the filter coefficient B(n) is calculated and the calculated filter coefficient B(n) is output to the adaptive filter 13b. The update unit 15b also sequentially updates the filter coefficient B(n) using the above [Equation 6]. In other words, the update unit 15b updates the filter coefficient B(n) using an update formula that includes the step size parameter μ.

[0057] [Configuration of the setting system] As described above, the memory unit 16 of the active noise reduction device 10 stores a pre-measured simulated transfer function. The actual acoustic transfer function inside the vehicle 50 will change depending on the arrangement of people and luggage inside the vehicle 50, even if the same active noise reduction device 10 is installed in the same vehicle 50 as an industrial product. Also, if the acoustic characteristics of the speakers are different, the overall sound input / output characteristics will change, so the noise reduction performance may change if the vehicle 50 is different. For this reason, the difference between the actual acoustic transfer function and the simulated transfer function stored in the memory unit 16 may become large, and it may not be possible to sufficiently reduce the noise with the cancellation sound, or the noise may increase with the cancellation sound (in other words, the cancellation sound itself becomes noise).

[0058] Therefore, the inventors have found a configuration for the active noise reduction device 10 that excludes frequency bands where changes in the acoustic transfer function are likely to occur from the frequency bands targeted for noise reduction. When the active noise reduction device 10 with this configuration is set up so that the frequency corresponding to the rotational speed of the engine 54 detected by the frequency detection unit 11 falls within a frequency band where changes in the acoustic transfer function are likely to occur, the output of the cancellation sound is stopped. This makes it possible to suppress the increase in noise caused by the cancellation sound.

[0059] The configuration of the setting system for making such settings will be described below. Figure 7 is a block diagram showing the functional configuration of the setting system.

[0060] As shown in Figure 7, the setting system 30 comprises an active noise reduction device 10 and an information terminal 20. Since the configuration of the active noise reduction device 10 has already been described, a detailed explanation is omitted.

[0061] The information terminal 20 is an information terminal used to configure various settings of the active noise reduction device 10, and is specifically a personal computer or tablet terminal. The information terminal 20 comprises a communication unit 21, an information processing unit 22, and a storage unit 23.

[0062] The communication unit 21 is a communication circuit for the information terminal 20 to communicate with the active noise reduction device 10 via a local communication network. The communication unit 21 may be a wired communication circuit or a wireless communication circuit. Although not shown in the figures, the active noise reduction device 10 also has a communication circuit corresponding to the communication unit 21.

[0063] The information processing unit 22 performs information processing for setting various settings of the active noise reduction device 10. The information processing unit 22 is implemented by, for example, a microcomputer, but may also be implemented by a processor. The functions of the information processing unit 22 are realized, for example, by the microcomputer or processor constituting the information processing unit 22 executing a computer program stored in the storage unit 23.

[0064] The memory unit 23 is a storage device that stores information necessary for information processing performed by the information processing unit 22, as well as computer programs executed by the information processing unit 22. The computer programs stored in the memory unit 23 include dedicated application programs (hereinafter also referred to as dedicated apps) for the setting system 30. The memory unit 23 is implemented by, for example, semiconductor memory, but may also be implemented by an HDD (Hard Disk Drive) or the like.

[0065] The configuration system 30 may be implemented as a client-server system, and in addition to the active noise reduction device 10 and the information terminal 20, it may also include a server device (cloud server). In this case, some or all of the processing performed by the information terminal 20 in the following embodiment may be performed by the server device. For example, the information terminal 20 may be used as a user interface with the worker, and the actual information processing may be performed by the server device.

[0066] [Example of the configuration system in operation 1] Next, we will describe an example of operation 1 for setting the active noise reduction device 10 to exclude frequency bands where changes in the acoustic transfer function are likely to occur from the frequency bands targeted for noise reduction. Figure 8 is a flowchart of the setting system 30's operation example 1.

[0067] The operator measures the acoustic transfer function from the speaker position to the microphone position multiple times in a vehicle 50 equipped with the active noise reduction device 10, either under the same measurement environment or under different measurement environments. A measuring instrument such as an FRA (Frequency Response Analyzer) is used for the measurement. The frequency band targeted for noise reduction is, for example, the frequency band from FsHz to FeHz (where Fs and Fe are positive integers), and in one measurement of the acoustic transfer function, for example, the gain and phase are measured in 1Hz increments within the frequency band from FsHz to FeHz. In other words, in one measurement of the acoustic transfer function, the gain and phase of Fe-Fs+1 pairs are measured.

[0068] The operator inputs multiple sets of measured acoustic transfer function data into an information terminal 20 running a dedicated application. The information processing unit 22 of the information terminal 20 acquires multiple sets of measured acoustic transfer function data (S11). The measured data is input into the information terminal 20 using, for example, a USB (Universal Serial Bus) memory, but the method of inputting the measured data into the information terminal 20 is not particularly limited.

[0069] The information processing unit 22 identifies frequency bands (hereinafter also referred to as exclusion bands) in which the measurement variability of the acquired measured data is greater than a threshold (S12). The exclusion band can be said to be a frequency band in which the acoustic transfer function is likely to change. For example, when identifying the exclusion band based on gain, the set of frequencies in which the measurement variability (dispersion) of the gain measured multiple times is greater than a threshold is considered the exclusion band. Figure 9 shows an example of the identification of the exclusion band, where the vertical axis of Figure 9 shows the measurement variability of the gain and the horizontal axis shows the frequency. The value of the threshold can be determined empirically or experimentally as appropriate by the designer of the active noise reduction device 10.

[0070] Next, the information processing unit 22 transmits setting information to the active noise reduction device 10 using the communication unit 21, thereby storing the setting information in the storage unit 16 of the active noise reduction device 10 (S13). This setting information is for reflecting in the active noise reduction device 10 the setting that it will not output a cancellation sound even if it detects noise having a frequency corresponding to the exclusion band identified in step S12. In other words, the active noise reduction device 10, with the setting information stored in the storage unit 16, stops outputting the cancellation sound when the frequency corresponding to the rotational speed of the engine 54 detected by the frequency detection unit 11 falls within the exclusion band where changes in the actual acoustic transfer function are likely to occur.

[0071] In this way, the active noise reduction device 10, after setting up in step S13, can suppress the increase in noise caused by the cancellation sound by not outputting a cancellation sound for noise that has a frequency corresponding to an exclusion band where the actual acoustic transfer function is likely to change when noise occurs.

[0072] [Example of the configuration system in operation 2] As described above, the memory unit 16 of the active noise reduction device 10 stores the measured simulated transfer function. However, there is room for further consideration regarding how to generate the simulated transfer function (method of generating the simulated transfer function). The inventor has found a method for generating the simulated transfer function based on a genetic algorithm. Below, the operation of generating the simulated transfer function based on the genetic algorithm and storing it in the memory unit 16 (operation example 2) will be described. Figure 10 is a flowchart of operation example 2 of the setting system 30.

[0073] The operator inputs multiple sets of measured acoustic transfer function data into the information terminal 20 running a dedicated application. The information processing unit 22 of the information terminal 20 acquires multiple sets of measured acoustic transfer function data (S21). The process in step S21 is the same as the process in step S11.

[0074] The information processing unit 22 generates a first simulated transfer function based on the acquired multiple sets of measured data (S22). For example, the information processing unit 22 generates the first simulated transfer function by averaging the acquired multiple sets of measured data.

[0075] The first simulated transfer function is composed of N first parameters corresponding to N distinct frequency values. The following explanation describes the case where the first parameter is gain, but the first parameter may also be phase. When the gain and phase are measured in 1Hz increments in the frequency band FsHz to FeHz, N = Fe - Fs + 1, and the frequency values ​​are integers between Fs and Fe. Note that N can be any natural number greater than or equal to 2.

[0076] The information processing unit 22 updates the first simulated transfer function based on a genetic algorithm (S23). Details of the update process in step S23 will be described later. The information processing unit 22 determines whether the number of updates has reached a predetermined number (S24). If the information processing unit 22 determines that the number of updates has not reached a predetermined number (No in S24), it performs the update process (S23) and the determination process (S24) again. If the information processing unit 22 determines that the number of updates has reached a predetermined number (Yes in S24), it stores the first simulated transfer function, which has been updated a predetermined number of times, as the final simulated transfer function in the storage unit 16 of the active noise reduction device 10 (S25). Specifically, the information processing unit 22 stores the final simulated transfer function in the storage unit 16 of the active noise reduction device 10 by transmitting the final simulated transfer function to the active noise reduction device 10 using the communication unit 21.

[0077] The predetermined number of times is a fixed number determined empirically or experimentally by the designer of the active noise reduction device 10, but the information processing unit 22 may repeat the update until the noise reduction performance of the first simulated transfer function converges (no further improvement is observed). In this case, the predetermined number of times is the number of times necessary for the noise reduction performance of the first simulated transfer function to converge.

[0078] Here, we will explain the details of the update process in step S23. Figure 11 is a flowchart showing the details of the update process.

[0079] During the update process, the information processing unit 22 simulates the noise reduction effect of the active noise reduction device 10 (for example, the frequency characteristics of the noise level) assuming that the first simulated transfer function before the update is used (S23a). The simulation is performed by executing a computer program for simulation that is pre-stored in the storage unit 23 of the information terminal 20.

[0080] Next, the information processing unit 22 generates a second simulated transfer function composed of N second parameters by randomly changing each of the N first parameters within a predetermined numerical range (S23b). The predetermined numerical range is, for example, ±5 dB, but can be determined empirically or experimentally by the designer of the active noise reduction device 10.

[0081] Next, the information processing unit 22 simulates the noise reduction effect of the active noise reduction device 10, assuming that the generated second simulated transfer function is used (S23c).

[0082] Next, the information processing unit 22 generates a third simulated transfer function composed of N third parameters based on the results of the processing in steps S23a to S23c, and sets the generated third simulated transfer function as the updated first simulated transfer function (S23d). Figure 12 is a diagram illustrating the method for generating the third transfer function.

[0083] As shown in Figure 12, the first simulated transfer function is G A1 , G A2 , G A3 ··G AN It is composed of N first parameters (gains), and the second simulated transfer function is G B1 , G B2 , G B3 ··G BN Assume it is composed of N first parameters (gains). The numerical values ​​represent frequency values.

[0084] The information processing unit 22 generates a third simulated transfer function by selecting the parameter that results in a lower simulated noise level (the parameter enclosed in an ellipse in Figure 12) from among the first and second parameters corresponding to the same frequency value, based on the simulation results of steps S23a and S23c, as the third parameter. In other words, the information processing unit 22 generates the third simulated transfer function by selectively selecting the best parameters for each frequency value.

[0085] In this way, the setting system 30 generates a first simulated transfer function that is expected to yield a high noise reduction effect by updating the first simulated transfer function a predetermined number of times based on a genetic algorithm, and can store this simulated transfer function used to generate cancellation sound in the memory unit 16 of the active noise reduction device 10.

[0086] In step S23b, while it was explained that the generation of the second simulated transfer function involves randomly changing each of the N first parameters within a predetermined numerical range, step S23b may also involve randomly changing the first parameters beyond the predetermined range with a relatively low predetermined probability, such as a few percent. In other words, the first parameters may be mutated. This can prevent the first parameters from getting stuck in a so-called local minimum and failing to approach the optimal solution.

[0087] Furthermore, in the generation of the third simulated transfer function in step S23d, the parameters were selected under the condition that the noise level in the simulation would be low. However, in addition to the noise level, or instead of the noise level, the condition could also be that the sound is perceived as good. In other words, in step S23d, the parameters should be selected under the condition that a high noise reduction effect can be obtained. Note that perceived sound is that the line showing the frequency characteristics of the noise level is smooth and has few localized irregularities.

[0088] [Example of the configuration system in operation 3] Next, we will describe the operation (operation example 3) of generating a simulated acoustic transfer function based on a brute-force algorithm, which differs from the genetic algorithm, and storing it in the memory unit 16. First, we will explain the overview of the brute-force algorithm. Figure 13 is a diagram illustrating the brute-force algorithm.

[0089] In the following, the final simulated transfer function will be referred to as the second simulated transfer function, and the provisional simulated transfer function used to determine the second simulated transfer function will be referred to as the first simulated transfer function. The case where the N first parameters (or second parameters) constituting the first simulated transfer function (or second simulated transfer function) are gains will be explained, but the first parameter (or second parameter) may also be phase.

[0090] In a brute-force algorithm, the N first parameters can take any of M predetermined values ​​(in other words, a set of candidate values), where M is a natural number greater than or equal to 2. Therefore, the information processing unit 22 of the information terminal 20 provisionally sets all N first parameters to one of the M possible values ​​(Figure 13(a)), and simulates the frequency characteristics of the error signal level (Figure 13(b)), performing this process for all M possible values. In other words, the information processing unit 22 generates a first simulated transfer function composed of the N first parameters with the provisionally set values, and simulates the noise reduction effect of the active noise reduction device 10 assuming the generated first simulated transfer function is used, performing this simulation M times, changing the provisionally set values ​​in each of the M ways.

[0091] The information processing unit 22 then selects the value among the M possible values ​​for each of the N first parameters that results in the smallest signal level of the error signal (the value that yields the highest noise reduction effect) as the final second parameter, thereby generating a second simulated transfer function that is considered to yield the highest noise reduction effect (Figure 13(c)). In other words, by selecting the first parameter that yields the highest noise reduction effect among the M possible first parameters corresponding to the same frequency value as the second parameter, a second simulated transfer function composed of N second parameters is generated.

[0092] The following describes Operation Example 3, which uses this brute-force algorithm. Figure 14 is a flowchart of Operation Example 3 of the setting system 30. Note that Operation Example 3 below is triggered, for example, when an operator performs a predetermined operation on the information terminal 20.

[0093] The information processing unit 22 of the information terminal 20 generates a first simulated transfer function by provisionally setting all N first parameters to one of M possible values ​​(S31). The information processing unit 22 then simulates the noise reduction effect of the active noise reduction device 10, assuming that the generated first simulated transfer function is used (S32).

[0094] Next, the information processing unit 22 determines whether all M types of first simulated transfer functions have been generated (simulated) (S33). If the information processing unit 22 determines that not all M types of first simulated transfer functions have been generated yet (No in S33), it generates a first simulated transfer function by provisionally setting all N first parameters to values ​​from the M types that have not yet been tried (S31), and then simulates the noise reduction effect of the active noise reduction device 10 assuming that this first simulated transfer function is used (S32). For example, the information processing unit 22 simulates the frequency characteristics of the signal level of the error signal.

[0095] On the other hand, if the information processing unit 22 determines that it has already generated all M of the first simulated transfer functions (Yes in S33), the information processing unit 22 generates a second simulated transfer function by adopting the value among the M values ​​that yields the highest noise reduction effect as the second parameter for each of the N first parameters (S34). Specifically, the information processing unit 22 generates a second simulated transfer function by adopting the value among the M values ​​that results in the smallest signal level of the error signal in the simulation as the second parameter for each of the N first parameters (S34).

[0096] Then, the information processing unit 22 stores the second simulated transfer function in the memory unit 16 of the active noise reduction device 10 (S35). Specifically, the information processing unit 22 stores the second simulated transfer function in the memory unit 16 of the active noise reduction device 10 by transmitting the second simulated transfer function to the active noise reduction device 10 using the communication unit 21.

[0097] In this way, the setting system 30 generates a second simulated transfer function that is expected to yield a high noise reduction effect based on a brute-force algorithm, and can store it in the memory unit 16 of the active noise reduction device 10 as a simulated transfer function used to generate the cancellation sound.

[0098] Furthermore, in the generation of the second simulated transfer function in step S34, the parameters were selected under the condition that the signal level of the error signal in the simulation is minimized. However, in addition to the signal level of the error signal, or instead of the signal level of the error signal, the condition that the sound is best may also be used. In other words, in step S34, the parameters should be selected under the condition that the greatest noise reduction effect is obtained.

[0099] [Differentiation] In the above embodiment, an example of reducing engine noise (noise correlated with the rotation of the engine 54) was described. However, the active noise reduction device 10 may also reduce noise correlated with, for example, the rotation of the propeller shaft. Furthermore, the active noise reduction device 10 may also reduce noise generated by the operation of power sources other than the engine 54.

[0100] Furthermore, while the above embodiment described an example of reducing one order component (e.g., a second-order component) of noise such as engine noise, the active noise reduction device 10 may reduce multiple order components of noise (e.g., a second-order component and a fourth-order component, etc.) simultaneously (in parallel).

[0101] Furthermore, in the above embodiment, three methods can be considered for generating the simulated transfer function: a conventional method that uses measured data (or multiple sets of measured data) as is, a method based on a genetic algorithm, and a method based on a brute-force algorithm. Here, the phase of the simulated transfer function may be generated using the method that uses measured data as is, while the gain may be generated using the method based on a genetic algorithm or a brute-force algorithm, thus employing different generation methods for the phase and gain.

[0102] Furthermore, in operation examples 1 to 3 of the above embodiment of the setting system, the setting method for the active noise reduction device 10 was described. Here, if the above setting method is performed as part of the manufacturing process of the active noise reduction device 10, the above setting method can also be considered as a method for manufacturing the active noise reduction device. Figure 15 is a flowchart of the method for manufacturing the active noise reduction device 10.

[0103] As shown in Figure 15, in the method for manufacturing the active noise reduction device 10, the active noise reduction device 10 is assembled (S41), setting information is stored in the storage unit 16 of the assembled active noise reduction device 10 based on operation example 1 (S42), and a simulated transfer function used to generate cancellation sound is stored based on operation example 2 or operation example 3 (S43). Thus, the invention derived from the disclosures of this specification includes a method for manufacturing the active noise reduction device 10.

[0104] [Effects, etc.] The inventions derived from the disclosures in this specification include, for example, the following. The inventions derived from the disclosures in this specification will be described below, along with the effects obtained by such inventions.

[0105] Invention 1 is a setting method for an active noise reduction device 10, which is executed by a computer such as an information terminal 20. The active noise reduction device 10 reduces noise by outputting a cancellation sound from the speaker 52 when it detects noise generated by the rotation of an engine 54 in a space 51 in a vehicle 50 (mobile device) where a speaker 52 and a microphone 53 are provided. The setting method includes: step S11, acquiring multiple sets of measured acoustic transfer function data from the position of the speaker 52 to the position of the microphone 53; step S12, identifying a frequency band in which the measurement variability of the acquired multiple sets of measured data is greater than a threshold; and step S13, storing setting information in the storage unit 16 of the active noise reduction device 10 to reflect a setting in the active noise reduction device 10 that does not output a cancellation sound even if noise corresponding to the identified frequency band is detected. The vehicle 50 is an example of a mobile device, and the engine 54 is an example of a power source.

[0106] With this setting method, the active noise reduction device 10 can suppress the increase in noise caused by the cancellation sound by not outputting a cancellation sound for noise that has a frequency corresponding to an exclusion band where the actual acoustic transfer function is prone to change. In other words, the setting method can suppress the increase in noise caused by the cancellation sound output to reduce noise. To put it another way, the setting method can improve the stability of the noise reduction performance.

[0107] Invention 2 is a setting method of Invention 1, further comprising the steps of: S22 generating a first simulated transfer function composed of N first parameters corresponding to N different frequency values ​​(N is a natural number of 2 or more) based on multiple sets of measured data obtained; and S25 storing the first simulated transfer function, which has been updated a predetermined number of times based on a genetic algorithm, in the storage unit 16 of the active noise reduction device 10.

[0108] This configuration method allows the first simulated transfer function, generated based on a genetic algorithm, to be stored in the memory unit 16 of the active noise reduction device 10.

[0109] Invention 3 is a setting method of Invention 2, in which, in updating the first simulated transfer function, the noise reduction effect of the active noise reduction device 10 assuming the first simulated transfer function is used is simulated, a second simulated transfer function consisting of N second parameters is generated by randomly changing each of the N first parameters within a predetermined numerical range, the noise reduction effect of the active noise reduction device 10 assuming the generated second simulated transfer function is used is simulated, and the parameter that can obtain a high noise reduction effect from among the first and second parameters corresponding to the same frequency value is set as the third parameter, thereby obtaining a third simulated transfer function consisting of N third parameters corresponding to N different frequency values, and this is set as the updated first simulated transfer function.

[0110] This configuration method allows a first simulated transfer function, which is generated based on a genetic algorithm and is expected to yield a high noise reduction effect, to be stored in the memory unit 16 of the active noise reduction device 10.

[0111] Invention 4 is a setting method of Invention 3, in which, during a predetermined number of updates of the first simulated transfer function, some of the N first parameters may be changed beyond a predetermined numerical range.

[0112] This configuration method can prevent the first parameter from getting stuck in a so-called local minima and failing to approach the optimal solution.

[0113] Invention 5 is a setting method of Invention 1, further comprising the steps of generating a first simulated transfer function composed of N first parameters, each having a provisionally set value corresponding to N different frequency values ​​(where N is a natural number of 2 or more), and simulating the noise reduction effect of the active noise reduction device 10 assuming the generated first simulated transfer function is used, by changing the provisionally set value in M ​​ways (where M is a natural number of 2 or more) and performing this simulation M times; and storing a second simulated transfer function in the storage unit 16 of the active noise reduction device 10, which is obtained by setting the first parameter that yields the highest noise reduction effect among the M ways of first parameters corresponding to the same frequency value as the second parameter, and which consists of N second parameters corresponding to N different frequency values.

[0114] This setting method allows a second simulated transfer function, which is expected to yield a high noise reduction effect, to be stored in the memory unit 16 of the active noise reduction device 10.

[0115] Invention 6 is a method for manufacturing an active noise reduction device 10, comprising the steps of: S41, assembling the active noise reduction device 10; and S42, storing setting information in the storage unit 16 of the assembled active noise reduction device 10 by performing the setting method described in Invention 1.

[0116] This method of manufacturing the active noise reduction device 10 makes it possible to manufacture an active noise reduction device 10 in which the noise is suppressed from increasing due to the cancellation sound output to reduce the noise. In other words, this method of manufacturing the active noise reduction device 10 makes it possible to manufacture an active noise reduction device with improved stability of noise reduction performance.

[0117] Invention 7 is a program that causes a computer to execute one of the setting methods from Inventions 1 to 5.

[0118] Such a program allows the computer to prevent the noise from increasing due to the cancellation sound output used to reduce noise. In other words, the computer can improve the stability of its noise reduction performance.

[0119] Invention 8 is an information terminal 20 for setting up an active noise reduction device 10, wherein the active noise reduction device 10 reduces the detected noise by outputting a cancellation sound from the speaker 52 when it detects noise generated by the rotation of the engine 54 of the vehicle 50 in a space 51 in the vehicle 50 where a speaker 52 and a microphone 53 are provided, and the information terminal 20 includes an information processing unit 22 that acquires multiple sets of measured acoustic transfer function data from the position of the speaker 52 to the position of the microphone 53, and the information processing unit 22 identifies a frequency band in which the measurement variability of the acquired multiple sets of measured data is greater than a threshold, and stores setting information in the storage unit 16 of the active noise reduction device 10 to reflect a setting in the active noise reduction device 10 that does not output a cancellation sound even if noise corresponding to the identified frequency band is detected.

[0120] Such an information terminal 20 can suppress the increase in noise caused by the cancellation sound output to reduce noise. In other words, the information terminal 20 can improve the stability of its noise reduction performance.

[0121] (Other embodiments) Although embodiments have been described above, this disclosure is not limited to the embodiments described above.

[0122] For example, the active noise reduction device according to the above embodiment may be mounted on a mobile device other than a vehicle. The mobile device may be, for example, an aircraft or a ship. Furthermore, this disclosure may be implemented as such a mobile device other than a vehicle.

[0123] Furthermore, the configuration of the active noise reduction device according to the above embodiment is merely an example. For instance, the active noise reduction device may include components such as a D / A converter, a low-pass filter (LPF), a high-pass filter (HPF), a power amplifier, or an A / D converter.

[0124] Furthermore, the processing performed by the active noise reduction device according to the above embodiment is merely an example. For instance, some of the processing described in the above embodiment may be implemented by analog signal processing instead of digital signal processing.

[0125] Furthermore, for example, in the above embodiment, a process performed by a specific processing unit may be performed by another processing unit. Also, the sequence of processes in the operation of the active noise reduction device described in the above embodiment is just an example. The sequence of processes may be changed, and the processes may be executed in parallel. Similarly, the sequence of processes in the operation of the setting system described in the above embodiment is just an example. The sequence of processes may be changed, and the processes may be executed in parallel.

[0126] Furthermore, the general or specific embodiments of this disclosure may be implemented in systems, apparatus, methods, integrated circuits, computer programs, or non-temporary recording media such as computer-readable CD-ROMs. They may also be implemented in any combination of systems, apparatus, methods, integrated circuits, computer programs, and computer-readable non-temporary recording media.

[0127] For example, the present disclosure may be implemented as an information terminal or setting system of the above embodiment. The present disclosure may be implemented as a setting method, a method for manufacturing an active noise reduction device, or a method for generating a simulated transfer function as described in the above embodiment. The present disclosure may be implemented as a program (program product) for causing a computer to execute these methods, or as a computer-readable non-temporary recording medium on which such a program is stored.

[0128] Furthermore, this disclosure also includes forms obtained by applying various modifications to each embodiment that a person skilled in the art could conceive, or forms realized by arbitrarily combining the components and functions of each embodiment without departing from the spirit of this disclosure. [Industrial applicability]

[0129] The setting method disclosed herein can improve the noise reduction performance of an active noise reduction device. [Explanation of Symbols]

[0130] 10 Active noise reduction device 11 Frequency detection unit 12 Reference signal generation section 12a Sine wave generator 12b Cosine wave generator 13, 13a, 13b Adaptive filters 13c Addition section 14, 14a, 14b Correction section 15, 15a, 15b update section 16, 23 Storage section 20 Information terminals 21 Communications Department 22 Information Processing Unit 30 Configuration System 50 Vehicles (Mobile Devices) 51 Space 52 speakers 53 Microphone 54 Engine 55 Engine Control Unit 56 ECU

Claims

1. A method for setting up an active noise reduction device, which is performed by a computer, The active noise reduction device reduces the detected noise by outputting a cancellation sound from the speaker when it detects noise generated by the rotation of the power source of the mobile device in the space within the mobile device where the speaker and microphone are provided. The aforementioned setting method is The steps include: acquiring multiple sets of measured data of the acoustic transfer function from the speaker position to the microphone position; The steps include identifying a frequency band in which the measurement variability of the multiple sets of acquired actual measurement data is greater than a threshold, The process includes the step of storing setting information in the memory of the active noise reduction device to reflect in the active noise reduction device a setting that the cancellation sound will not be output even if noise corresponding to the specified frequency band is detected. How to set it up.

2. moreover, The steps include generating a first simulated transfer function based on the multiple measured data obtained, consisting of N first parameters corresponding to N distinct frequency values ​​(where N is a natural number greater than or equal to 2), and The process includes the step of storing the first simulated transfer function, which has been updated a predetermined number of times based on a genetic algorithm, in the memory unit of the active noise reduction device. The setting method according to claim 1.

3. In updating the first simulated transfer function, The noise reduction effect of the active noise reduction device was simulated assuming the first simulated transfer function was used. By randomly changing each of the N first parameters within a predetermined numerical range, a second simulated transfer function composed of N second parameters is generated. The noise reduction effect of the active noise reduction device is simulated assuming that the generated second simulated transfer function is used. The updated first simulated transfer function is defined as a third simulated transfer function consisting of N third parameters corresponding to N different frequency values, obtained by selecting a third parameter from among the first and second parameters corresponding to the same frequency value that provides a high noise reduction effect, and using these N third parameters. The setting method described in claim 2.

4. In updating the first simulated transfer function a predetermined number of times, some of the N first parameters may be changed beyond the predetermined numerical range. The setting method described in claim 3.

5. moreover, The steps include generating a first simulated transfer function composed of N first parameters, each having a provisionally set value corresponding to N distinct frequency values ​​(where N is a natural number greater than or equal to 2), and simulating the noise reduction effect of the active noise reduction device assuming the generated first simulated transfer function is used, by changing the provisionally set value in M ​​different ways (where M is a natural number greater than or equal to 2) and performing this simulation M times; The process includes the step of storing in the memory unit of the active noise reduction device a second simulated transfer function, which is obtained by selecting the first parameter that provides the highest noise reduction effect from among M ways of the first parameter corresponding to the same frequency value as the second parameter, and which consists of N second parameters corresponding to N different frequency values. The setting method according to claim 1.

6. Steps for assembling the active noise reduction device, The method includes the step of storing the setting information in the storage unit of the assembled active noise reduction device by performing the setting method described in claim 1. A method for manufacturing an active noise reduction device.

7. A program for causing the computer to execute the setting method described in any one of claims 1 to 5.

8. An information terminal for configuring settings related to an active noise reduction device, The active noise reduction device reduces the detected noise by outputting a cancellation sound from the speaker when it detects noise generated by the rotation of the power source of the mobile device in the space within the mobile device where the speaker and microphone are provided. The aforementioned information terminal is The system includes an information processing unit that acquires multiple sets of measured acoustic transfer function data from the speaker position to the microphone position. The aforementioned information processing unit, Identify the frequency band in which the measurement variability of the multiple measured data points obtained is greater than the threshold. Setting information for reflecting the setting that the cancellation sound will not be output even if noise corresponding to the specified frequency band is detected is stored in the memory of the active noise reduction device. Information terminal.

Images