JP2002374588A - Device and method for reducing acoustic noise - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an acoustic noise reduction device that can effectively reduce acoustic noise from an acoustic energy wave generating means independently of dispersion in the characteristic of the acoustic energy wave generating means and dispersion in vibration constants due to the characteristic and accuracy of members associated with the acoustic energy wave generating means. SOLUTION: The acoustic noise reduction device is provided with an acoustic energy wave generating means 23 that is driven by a drive signal from a drive signal source and generates an acoustic energy wave, an adaptive filter 33 that uses a signal on the basis of the drive signal for a reference input, a delay means 34 that receives a signal on the basis of the drive signal and has a prescribed transfer function, a least square mean arithmetic means 35 that uses the output of the delay means 34 for a reference input, an error microphone 30 that picks up the acoustic energy wave and gives a signal on the basis of an output residue signal to the least square mean arithmetic means 35, and a subtractor means 27 that subtracts an adaptive output signal form the adaptive filter 33 from the drive signal. The subtractor means 27 reduces the acoustic noise from the acoustic energy wave generating means 23.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an acoustic noise reducing apparatus and method.

[0002]

2. Description of the Related Art In recent years, the number of home electronic devices having a built-in motor has been increasing more and more. In addition, due to demands for higher functions and higher access speeds of home electronic devices, the built-in motor is rotated at a high speed. Tend to be. For this reason, noise and noise generated from the motor have become a significant problem.

[0003] In addition, since home electronic equipment incorporating such a motor is sometimes used in a relatively quiet place, the noise and noise of the motor must be reduced as much as possible. Conventionally, mechanical measures such as structural sound insulation by the case of home electronic equipment, improvement of the bearing of the motor, and suppression of vibration around the drive unit have already been performed. There has been a limit in reducing noise and noise generated from home electronic devices that incorporate the same.

By the way, US Pat. No. 5,638,267 proposes a method of suppressing such noise and vibrations and disturbances of a motor driving part which cause the noise mainly by a feedforward control in a driving circuit. ing. In a hard disk drive (hereinafter abbreviated as HDD), a head attached to an actuator is a voice coil motor (hereinafter abbreviated as VCM).
Therefore, noise generated when a seek operation is performed is suppressed mainly by controlling a drive signal from a digital servo circuit including a digital signal processor (hereinafter abbreviated as DSP).

More specifically, the noise at the time of this seek is VC
This is due to the vibration of the drive part generated when M suddenly accelerates and decelerates the actuator to the data read / write position on the disk. As a countermeasure, an effective drive waveform for suppressing the vibration is obtained in advance. During operation, the drive waveform is deformed by feedforward control, or a seek operation is performed in a time change pattern in which a rapid change in acceleration is suppressed so as not to generate vibration.

[0006]

In such a conventional acoustic noise reduction method, the motor is not controlled so that the target noise is always minimized. And
There is a drawback that it is difficult to reduce noise if there are variations in motor characteristics.

In view of the above, the present invention controls the acoustic energy wave generating means so as to minimize the target noise at all times, thereby providing a member related to the acoustic energy wave generating means, for example, an actuator part. We propose an acoustic noise reduction device and method that can effectively reduce acoustic noise from acoustic energy wave generation means even if there are variations in vibration constants due to characteristics and accuracy, and variations in characteristics of acoustic energy wave generation means. What you want to do.

[0008]

According to a first aspect of the present invention, an acoustic energy wave generating means for generating an acoustic energy wave driven by a drive signal from a drive signal source, and a signal based on the drive signal as a reference input. An adaptive filter, a delay unit having a predetermined transfer function to which a signal based on the drive signal is input, a least-mean-square calculating unit having an output of the delay unit as a reference input, and collecting an acoustic energy wave; An error microphone for inputting a signal based on the output residual signal to the least mean square calculating means, and a subtraction means for subtracting an adaptive output signal from the adaptive filter from the drive signal; This is an acoustic noise reduction device configured to reduce acoustic noise from the means.

According to a second aspect of the present invention, in the acoustic noise reduction device of the first aspect, limiter means for inputting an output residual signal from an error microphone is provided, and the output signal from the limiter means is subjected to least square mean. This is an acoustic noise reduction device to be input to the calculation means.

According to a third and a fourth invention, in the acoustic noise reduction device according to the first and second inventions, respectively, a level detecting means to which an output residual signal from an error microphone is inputted;
Acoustic noise reduction comprising: level comparing means for comparing a detection output from the level detecting means with a threshold level; and switch means inserted between the adaptive filter and the subtracting means, the on / off of which is controlled by the comparison result of the level comparing means. Device.

According to a fifth, sixth, seventh and eighth aspect of the present invention, in the acoustic noise reducing apparatus according to the first, second, third and fourth aspects, acoustic noise is generated by the acoustic energy wave generating means. When the step gain coefficient in the least mean square calculation means is set to a predetermined value, and when no acoustic noise is generated from the acoustic energy wave generation means,
This is an acoustic noise reduction device in which the step gain coefficient in the least mean square calculation means is set to zero so that the coefficient of the adaptive filter is not updated.

According to a ninth aspect of the present invention, the acoustic energy wave generating means is driven by a drive signal from a drive signal source to generate an acoustic energy wave, and a signal based on the drive signal is input to the adaptive filter as a reference input, and A signal based on the signal is input to delay means having a predetermined transfer function, and the output of the delay means is input to the least mean square calculation means as a reference input, and the acoustic energy wave is collected by an error microphone,
A signal based on the output residual signal is input to the least mean square calculation means, an adaptive output signal from the adaptive filter is input to the subtraction means, and is subtracted from the drive signal to obtain an acoustic noise from the acoustic energy wave generation means. This is a method of reducing acoustic noise.

According to a tenth aspect, in the acoustic noise reduction method according to the ninth aspect, the output residual signal from the error microphone is input to the limiter means, and the output signal from the limiter means is input to the least mean square calculating means. This is a method for reducing acoustic noise.

[0014] The eleventh and twelfth inventions respectively correspond to the ninth invention.
In the acoustic noise reduction method according to the tenth aspect, an output residual signal from an error microphone is input to a level detection unit, and a detection output from the level detection unit is input to a level comparison unit, and is compared with a threshold level. The acoustic noise reduction method is such that the on / off of a switch inserted between the adaptive filter and the subtractor is controlled based on the comparison result of the level comparator.

According to a thirteenth, fourteenth, fifteenth, and sixteenth aspect, in the ninth, tenth, eleventh, and twelfth acoustic noise reduction methods, acoustic noise is generated by the acoustic energy wave generating means. When the step gain coefficient in the least mean square calculation means is set to a predetermined value, and when no acoustic noise is generated from the acoustic energy wave generation means, the step gain coefficient in the least mean square calculation means is set to zero. This is an acoustic noise reduction method in which the coefficient of the adaptive filter is not updated. .

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the acoustic noise reduction method according to the present invention, the above-described noise is obtained as an acoustic signal by a microphone, and the LMS (Least
Controls the adaptive coefficient of an adaptive filter that has a mean square (least mean square method) operation algorithm and transforms the drive signal from the digital servo circuit with the output of the adaptive filter so that this acoustic noise is minimized. is there.

Further, the acoustic noise reduction method according to the present invention comprises:
Unlike the conventional method of reducing noise by feedforward control, the noise is reduced by feedback control, so the motor is controlled so that the target noise is always minimized, and the variation in vibration constant due to the characteristics and accuracy of the actuator part In addition, even if there are variations in the motor characteristics, it is possible to reduce noise.

Further, in the acoustic noise reduction method according to the present invention, the present invention can be applied to, for example, a hard disk drive, and detects a surrounding environmental sound level input to a microphone.
For example, if the ambient sound level is high,
An automatic selection mode is provided in which the FF is used to perform a high-speed seek operation, and when the surroundings are quiet, the reduction function is turned on to perform a normal seek operation.

Further, the acoustic noise reduction method according to the present invention can be applied to, for example, a camera-integrated VTR, and can be applied to a device having a recording device including driving parts therein, such as a rotary drum motor or a cap. Stan motor,
The noise generated from the zoom / focus motor of the lens, etc. is acquired as an acoustic signal by a built-in microphone, and each motor drive signal is similarly transformed by adaptive processing,
Acoustic noise can be reduced.

Hereinafter, an acoustic noise reduction apparatus and method according to an embodiment of the present invention will be described.
LMS (Least Mean Square) via an FIR filter used in the acoustic noise reduction apparatus and method according to the present invention.
) (Least Square Mean Method) An operation algorithm circuit will be described with reference to FIG.

In FIG. 1, a main input S from an input terminal 1 is a desired signal, a noise N from an input terminal 2 is a noise signal generated from a noise source, and a reference input X from an input terminal 3 is This is a signal correlated with the noise N.

Further, the plant (Plant) circuit 7 is a circuit having an arbitrary transfer function. For example, it can be considered that the plant circuit 7 is a circuit which models a transmission path of an acoustic signal radiated into a certain space. This plant circuit 7 is a control target in adaptive control. LM via FIR filter
The S operation algorithm circuit inputs the plant (P
lant) The transfer function of the circuit 7 is changed to FIR (Finit Impulse Res
ponse) (finite impulse response) filter 4
It is characterized in that it is input to an LMS (least mean square) calculation unit 5.

First, a main input S from the input terminal 1 is input to one + terminal of the adder 8, and a noise signal N is input to the other + terminal from the input terminal 2. The output of the adder 8 is input to the + terminal of another adder 9, and the adaptive filter output Y from the adaptive filter 6 input to the − terminal of the adder 9 is subtracted from the output of the adder 8. Is done. The output of the adder 9 is supplied to the plant circuit 7, and the output of the plant circuit 7 is output from the output terminal 10.

The output of the plant circuit 7 is the residual signal E
Is input to the LMS calculation unit 5. The reference input X from the input terminal 3 is input to the LMS operation unit 5 via the FIR filter 4. The reference input X from the input terminal 3 is also input to the adaptive filter 6.

The adaptive filter 6 generally has the configuration shown in FIG.
Is composed of an FIR filter as described later,
The filter coefficient is adaptively controlled by the operation result of the LMS operation unit 5, so that the adder 9 subtracts the noise N from the main input S from the input terminal 1 and outputs a desired signal. Is done.

Further, a specific configuration of the LMS adaptive filter 6 of FIG. 1 will be described with reference to FIG. First, FIG.
1 corresponds to the reference input X in FIG. 1 and is input to the adaptive filter 6 surrounded by the broken line and to the LMS calculation unit 5 via the FIR filter 4.

The adaptive filter 6 is generally composed of an FIR digital filter having several hundred taps.
The adaptive filter coefficients W ₀ , W ₁ ,
W ₂ , ‥‥‥‥, and W _m are adaptively updated according to the LMS algorithm.

Here, the FIR of (m + 1) taps
Shows a filter. 11₁, 11_Two, ‥‥‥‥, 1
1_mIndicates a delay unit with a unit sampling time, and X₀, X
₁, X_Two, ‥‥‥‥, X_mIs subject to each delay
Signal, and 12₀, 12 ₁12_Two, ‥‥‥‥, 12
_mIndicates a multiplier for coefficient multiplication, and W₀, W₁, W_Two,
‥‥‥‥, W_mIs the multiplier 12₀, 12₁, 12_Two, ‥
‥‥‥, 12_mShows the multiplication coefficient of.

Each of the multipliers 12 ₀ , 12 ₁ , 1
The outputs from 2 ₂ , ‥‥‥‥, and 12 _m are added by an adder 13 and output as an adaptive filter output Y. Therefore, the adaptive filter output Y is represented by the following equation.

[0030]

(Equation 1)

Here, Y corresponds to the output Y of the adaptive filter shown in FIG. Further, the LMS calculation unit 5 calculates the respective adaptive filter coefficients W ₀ , W ₁ , W ₂ , ‥‥‥‥, and W _m from the reference input X and the residual signal E in accordance with the following equation for each sampling. Update.

[0032]

## EQU2 ## W _{k + 1} = W _k + 2μ · E _k · X _k

In the equation (2), each suffix k represents the passage of time. For example, k is a unit sampling number, and W is the (k + 1) th sampling.
_{If k + 1} is the current adaptive filter coefficient, W _k represents the k-th sampling, that is, the adaptive filter coefficient one sampling past. Μ is a step gain, or
It is called a step size and is a parameter that determines the convergence speed in the LMS algorithm. The larger the value, the faster the convergence, but the accuracy after convergence decreases. Conversely, the smaller the value, the slower the convergence, but the accuracy after convergence. Rises,
It is optimized and set according to the adaptive system conditions used. Also, an E _k and X _k are referred to as respectively k sampling th residual signal E input X, therefore, in the number 2 in the formula (k + 1) sampling th W _{K + 1} to k sampling th W _k, A process of sequentially calculating and updating from E _k and X _k is performed.

Then, the LM through the FIR filter shown in FIG.
In the S operation algorithm circuit, a time delay by the plant circuit 7 exists between the adaptive filter output Y output from the adaptive filter 6 and the residual signal E. The processing is performed by the FIR filter 4 and input to the LMS operation unit 5 to satisfy the sequential operation within the same sampling. However, the processing timing of the adaptive filter 6 and the processing timing of the LMS operation unit 5 also have a time delay. Minute shift. However, the adaptive filter is generally a nonlinear, time-varying system. If the noise N of the target changes in a sufficiently long time with respect to the sampling time, and the adaptive operation proceeds slowly in accordance with this, It can be treated as a linear, time-invariant system, and the equation of Equation 2 can be applied.

Therefore, the LMS operation unit 5 in FIG. 1 always reduces the adaptive filter coefficient W in the adaptive filter 6 to minimize the noise N component which is a signal having a high correlation with the reference input X included in the residual signal E. As described above, the output is updated by the equation (2), so that an output from which noise has been reduced is always obtained at the output terminal 10.

Next, an overall configuration of an example of the acoustic noise reduction device and the reduction method according to the embodiment of the present invention will be described with reference to FIG. FIG. 3 shows a case where the present invention is applied to a reduction in seek operation noise during data read / write in an HDD (hard disk drive).
This is a resonance sound caused by vibration generated in a head and an actuator when a CM (voice coil motor) crosses a magnetic field.

To suppress this unnecessary vibration, the VCM
In the present invention, the vibration frequency component is removed from the control signal input to the control signal, and this is optimized by an adaptive filter based on an LMM operation algorithm via an FIR filter.

First, in FIG. 3, a magnetic head 20, a hard disk 21, a spindle motor 22, a VCM 23
And the like, the spindle motor 22
Is driven by a control system (not shown) so that the hard disk 21 maintains a predetermined number of revolutions, and predetermined data is stored on the hard disk 21 by an actuator portion including the magnetic head 20, the VCM 23, and the like. Read / write (read / write) is performed at a predetermined position.

In recent years, digital servo control by a DSP (digital signal processor), a microcomputer, or the like, which has become mainstream, has been adopted for servo control of the actuator portion.
In order to position the head 20, a sector servo system, which has become mainstream in recent years, is adopted, in which servo information written in a time-division multiplexed manner with data on the hard disk 21 is sampled and used for control. .

First, the servo information read by the magnetic head 20 is decoded by a read channel 24 composed of an equalizer, a Viterbi decoder and the like. The data is sent to a memory and subjected to a predetermined process. The servo information is input to a digital servo control circuit 25 including a DSP or the like. On the other hand, target position information is input to the digital servo control circuit 25 from an input terminal 26. The target position information is generated by a microcomputer or the like (not shown) and read / write as a final purpose at the time of seeking. Location information.

In the digital servo control circuit 25, the control described later with reference to FIG. 4 is performed, and the control output is input to the + terminal of the adder 27, and the control output is applied to the-terminal of the adder 27. The adaptive filter output Y input to the terminal is subtracted. The output of the adder 27 is supplied to a D / A converter 28, which performs digital-to-analog conversion. The analog output is supplied to a VCM 23 via a VCM driver 29, and the VCM 23 causes the magnetic head 20 to be stored on the hard disk 21. It is driven to a predetermined head position.

Here, the seek operation sound as described above is generated by this series of operations. In the present invention, this acoustic noise is converted into an electric signal by the error microphone 30 installed near the hard disk. The signal from the error microphone 30 is amplified by an amplifier (AMP) 31 and then supplied to an A / D converter 32 to be analog-converted.
Digitally converted. The residual signal E, which is a digital output from the A / D converter 32, is passed through a limiter 36 to an LMS
The data is input to the calculation unit 35.

Here, the limiter 36 is connected to the error microphone 3
0 is used to limit a signal or the like based on excessive sound input together with the seek operation sound. The threshold level to be limited is set near the maximum level of the seek operation sound to be generated. Only the operation sound signal is input.

Further, the sampling timing signal of the servo signal from the digital servo control circuit 25 is input to the LMS operation section 35 as a reference input X by the FIR filter 34.
, And the sampling timing signal is also input to the adaptive filter 33. Here, the adaptive filter 33, the LMS operation unit 35, the FIR filter 34
Has the same configuration as the adaptive filter 6, the LMS operation unit 5, and the FIR filter 4 in FIGS. 1 and 2, and the adaptive coefficient of the adaptive filter 33 is calculated according to the equation (2).
And the output Y of the adaptive filter is added to the adder 27.
Is input to

Next, a specific configuration of the digital servo control circuit 25 in FIG. 3 will be described with reference to FIG. First, the servo information is input to the input terminal 40, the current detection position information is calculated by the position calculation unit 42, and the calculated detection position information is input to the + terminal of the adder 43 and one of the speed monitor 44. Input terminal.

The input terminal 4 is connected to the negative terminal of the adder 43.
The target position information from 1 is input and subtracted from the detected position information input to the + terminal of the adder 43, and the difference between the current detected position and the target position is output from the adder 43. This difference signal is output to the speed control circuit 4
5, input to the feedforward control circuit 46 and the following control circuit 47. First, the speed control circuit 45
Is controlled according to the speed profile for the remaining seek distance to the target position. The feedforward control circuit 46 is a control for accelerating by a predetermined feedforward amount based on the difference signal from the adder 43. The following control circuit 47 includes the magnetic head 20
This is tracking control once the target track position on the hard disk 21 is reached.

The control signal from the feed forward control circuit 46 is input to an adder 49, and the speed control circuit 45
Is added to the output of the adder 48, which is a control signal from the control unit, and the added output is input to one fixed terminal of the changeover switch (SW) 50 and to the other input terminal of the speed monitor 44. The speed monitor 44 observes a difference between the actual speed based on the position detected by the position calculator 42 and the output speed, and inputs the amount of the difference to the minus terminal of the adder 48 to perform speed correction. Performs feedback control.

The control output from the following control circuit 47 is input to the other fixed terminal of the changeover switch (SW) 50. The switching output of the changeover switch (SW) 50 is output from the output terminal 51 as a control output.

Therefore, in FIG. 4, first, at the start of the seek operation, the changeover switch (SW) 50 selects the output of the adder 49. When the maximum acceleration is performed by the feedforward control and the speed approaches the target position, the speed is increased. It reaches the target position while decelerating by control, and then the changeover switch (SW) 50
Is switched to the following control circuit 47 side and repeats the operation of performing the following control. A series of these controls are performed by digital processing using a DSP or the like, and the sampling frequency is recorded per rotation of the hard disk. It is determined by the number of servo sectors and the rotation frequency, and is generally about 3 to 10 kHz.

As described above, in the acoustic noise reduction device of FIG. 3, since the seek operation is controlled at each sample timing of the digital servo control as shown in FIG. 4, the seek sound generated at that time is also sampled. The noise frequency is generated repeatedly at each timing, and the noise frequency includes many harmonics of the sampling frequency. Therefore, by inputting the timing pulse depending on the sampling frequency to the reference input X, the adaptive filter output Y is generated so as to minimize the harmonic noise of the sampling frequency remaining in the residual signal E. Then, the FIR filter 34
Is made to have a delay corresponding to the time from the adaptive filter 33 to the error microphone 30.
Adaptive processing is performed by the LMS operation algorithm via the R filter.

Next, an actual operation example of the HDD will be described with reference to FIG. First, FIG.
This is a time change of the VCM drive current from the M driver 29, and shows a waveform from a seek operation to a follow-up operation to a target track through a settling operation.
Generally, it has a time of about 10 mS from the start to the end of the seek operation of the full stroke seek.

The settling operation is an operation of pulling a target track. First, at the start of the seek operation, the maximum current is supplied to the VCM 23, and the magnetic head 20 is accelerated to a target speed. When the magnetic head 20 gradually approaches the target position, a reverse current is supplied to the VCM 23 according to a predetermined speed profile. To reach the target track while decelerating. FIG. 5A also shows the time change of the head acceleration.

Next, FIG. 5B shows a time change of the speed of the magnetic head 20, which is obtained by integrating the acceleration of FIG. 5A with time. Further, FIG. 5C shows a time change of the position of the magnetic head 20, which is obtained by integrating the speed of FIG. 5B with time.

Here, what is to be reduced by the acoustic noise reduction device is the acoustic noise generated during the seek operation, and this acoustic noise is the rapidly rising acceleration portion of the VCM drive current waveform shown in FIG. Occurs in the deceleration part where the vehicle suddenly falls. In this part, since the drive loop contains many harmonics, unnecessary vibration or resonance that causes acoustic noise in the actuator is generated.

Therefore, as shown in FIG. 6, when the adaptive control shown in FIG. 6B is used, the acoustic noise due to unnecessary vibration of the target portion is minimized with respect to the VCM drive current waveform without the adaptive control shown in FIG. 6A. As described above, that is, a pseudo noise waveform of the frequency of the unnecessary vibration is generated by the adaptive filter 33, and the current waveform is generated by subtracting the pseudo noise waveform so as to suppress the unnecessary vibration.

Next, referring to FIG. 7, the overall configuration of another example of the acoustic noise reduction device and the reduction method according to the embodiment of the present invention will be described. FIG. 7 shows a configuration in which an automatic ON / OFF control unit for acoustic noise reduction is added to the acoustic noise reduction device of FIG. 3. In FIG. 7, portions corresponding to FIG. A duplicate description is omitted.

First, similarly to FIG. 3, the acoustic noise obtained by the error microphone 30 is converted to an amplifier (AMP) 31 and A
The signal is taken into the adaptive processing via the / D converter (ADC) 32. At this time, environmental sounds in the surroundings are taken in at the same time as the acoustic noise. The synthesized signal of the acoustic noise and the environmental sound is input to an environmental sound level detection circuit 60 to be described later, and the limiter 36 limits the environmental sound signal whose level is higher than the acoustic noise, and then inputs the LMS arithmetic unit 35. However, since the environmental sound signal has no correlation with the reference input X, the influence on the adaptive processing is small.

Next, a specific configuration of the environmental sound level detection circuit 60 in FIG. 7 will be described with reference to FIG. First,
The synthesized signal of the acoustic noise and the environmental sound input to the input terminal 70 is supplied to an absolute value processing circuit 72 to be converted into an absolute value, and the absolute value signal is supplied to a detection unit 73. Is calculated, and the average value is input to one terminal of the level comparator 74.

The other terminal of the level comparator 74 has
The threshold level from the input terminal 71 is input, for example, from a microcomputer (not shown). The level comparator 74 compares these two levels. If the environmental sound level is higher than the threshold level, an OFF control signal is output from the output terminal 75. Conversely, the environmental sound level is higher than the threshold level. If it is smaller, the output terminal 75 outputs an ON control signal. Then, the changeover switch (SW) 61 in FIG. 7 is controlled by the ON / OFF control output.
Is turned on, and when it is off, the changeover switch (SW) 61 is controlled to be turned off.

Thus, when the environmental sound is loud, the acoustic noise is difficult to hear, so that the adaptive filter output Y is not subtracted and the adaptive control is turned off. Acoustic noise reduction automatic ON / OFF control such that the control is turned ON can be performed.

Thus, for example, a high-speed seek operation mode in which acoustic noise is generated and a normal seek operation mode in which acoustic noise is reduced are automatically selected. In general, H
In the seek operation in the DD, for example, after the head movement to the target track is completed, there is time until the next seek operation, so that the generated acoustic noise is often intermittent. In addition, it can be said that the frequency components of the acoustic noise generated from the actuator unit in each seek operation have a high correlation.

Therefore, in the present invention, adaptive processing may be provided with a learning effect for such intermittently generated noise to accelerate convergence. In this case,
In the above equation 2, the adaptive operation is performed with the step gain μ set to a predetermined value only during the seek operation, and μ is set to zero during the following operation other than the seek operation.
The updating of the coefficient of the adaptive filter is stopped, and the adaptive processing is executed while the adaptive filter coefficient at the previous seek is stored.

As described above, according to the acoustic noise reducing apparatus and method according to the present invention, the noise component included in the energy wave generated from the acoustic energy wave generating means based on the driving signal from the driving signal source is converted into the driving signal. The driving signal to the acoustic energy wave generating means is controlled so as to be minimized by adaptive processing using a signal having a high correlation as a reference input.

The present invention is not limited to an HDD, but includes a drive device controlled by a servo signal, for example,
The present invention is also applicable to a helical scan type magnetic recording / reproducing device, an optical disk recording / reproducing device, and the like.

[0065]

According to the first aspect of the present invention, the acoustic energy wave generating means for generating the acoustic energy wave driven by the drive signal from the drive signal source, and the adaptation using the signal based on the drive signal as a reference input. A filter, a delay unit having a predetermined transfer function to which a signal based on the drive signal is input, a least-mean-square calculating unit having an output of the delay unit as a reference input, and a sound energy wave collected, and an output remaining An error microphone for inputting a signal based on the difference signal to the least mean square calculation means, and a subtraction means for subtracting an adaptive output signal from the adaptive filter from the drive signal; Therefore, by controlling the acoustic energy wave generation means so that the target noise is always minimized, the acoustic energy is reduced. Effectively reduce acoustic noise from the acoustic energy wave generator even if there are variations in the vibration constant due to the characteristics and accuracy of the members related to the energy wave generator, for example, the characteristics and accuracy of the actuator part. An acoustic noise reduction device that can be reduced can be obtained.

According to a second aspect, in the acoustic noise reduction device of the first aspect, limiter means for inputting an output residual signal from the error microphone is provided, and the output signal from the limiter means is minimized. Since the signal is input to the root mean square calculating means, in addition to the effect of the first aspect, when an acoustic signal having a higher level than the acoustic energy wave from the acoustic energy wave generating means is mixed into the error microphone, the level of that level is reduced. It is possible to obtain an acoustic noise reduction device capable of realizing a normal and stable acoustic noise reduction effect by eliminating the influence of a high acoustic signal.

According to the third and fourth inventions, in the acoustic noise reduction devices of the first and second inventions, respectively, the level detecting means to which the output residual signal from the error microphone is inputted, and the level detecting means thereof Level comparing means for comparing the detected output with the threshold level, and switch means inserted between the adaptive filter and the subtracting means, the on / off of which is controlled by the comparison result of the level comparing means. In addition to the effects of the second aspect, it is possible to obtain an acoustic noise reduction device capable of controlling whether or not acoustic noise is reduced by the acoustic energy wave generating means depending on the level of environmental sound.

According to the fifth, sixth, seventh and eighth aspects of the present invention, in the acoustic noise reducing apparatuses of the first, second, third and fourth aspects, respectively, the acoustic noise is reduced by the acoustic energy wave generating means. When this occurs, the step gain coefficient in the least mean square calculation means is set to a predetermined value,
When no acoustic noise is generated by the acoustic energy wave generating means, the step gain coefficient in the least mean square calculating means is set to zero so that the coefficient of the adaptive filter is not updated. 3rd, 4th
In addition to the effect of the invention, for example, avoiding adaptive operation every time for noise generated intermittently such as seek sound at the time of seek operation in a hard disk drive, and maintaining the previous adaptive state Accordingly, a learning effect can be provided, and thereby, an acoustic noise reduction device that can increase the pull-in speed of the adaptive processing can be obtained.

According to the ninth aspect, the acoustic energy wave generating means is driven by the drive signal from the drive signal source to generate an acoustic energy wave, and a signal based on the drive signal is input to the adaptive filter as a reference input. A signal based on the drive signal is input to delay means having a predetermined transfer function, the output of the delay means is input as a reference input to the least mean square calculation means, and an acoustic energy wave is collected by an error microphone and output. A signal based on the residual signal is input to the least-mean-square calculating means, an adaptive output signal from the adaptive filter is input to the subtracting means, and is subtracted from the driving signal to reduce acoustic noise from the acoustic energy wave generating means. By controlling the acoustic energy wave generating means so that the target noise is always minimized, the acoustic energy wave Effective reduction of acoustic noise from the acoustic energy wave generating means even if there are variations in vibration constants due to the characteristics and accuracy of the members related to the means, for example, the actuator portion, and variations in the characteristics of the acoustic energy wave generating means. Thus, a method for reducing acoustic noise can be obtained.

According to a tenth aspect, in the acoustic noise reduction method according to the ninth aspect, the output residual signal from the error microphone is input to the limiter means, and the output signal from the limiter means is calculated by the least mean square arithmetic means. In addition to the effect of the ninth aspect, when an acoustic signal having a higher level than the acoustic energy wave from the acoustic energy wave generating means is mixed into the error microphone, the sound signal having the higher level is input. An acoustic noise reduction method capable of realizing a normal and stable acoustic noise reduction effect by eliminating the influence can be obtained.

According to the eleventh and twelfth inventions, in the acoustic noise reduction methods of the ninth and tenth inventions, respectively, the residual signal output from the error microphone is input to the level detecting means, and Is input to the level comparing means, compared with the threshold level, and the on / off of the switch means inserted between the adaptive filter and the subtracting means is controlled by the comparison result of the level comparing means. In addition to the effects of the ninth and tenth aspects, it is possible to obtain an acoustic noise reduction method that can control whether or not acoustic noise is reduced by the acoustic energy wave generating means depending on the level of environmental sound. it can.

According to the thirteenth, fourteenth, fifteenth, and sixteenth aspects, in the ninth, tenth, eleventh, and twelfth aspects of the acoustic noise reduction method, acoustic noise is reduced by the acoustic energy wave generating means. When it occurs, the step gain coefficient in the least mean square calculation means is set to a predetermined value, and when no acoustic noise is generated from the acoustic energy wave generation means, the step gain coefficient in the least mean square calculation means is set to zero. Since the adaptive filter coefficients are not updated, in addition to the effects of the ninth, tenth, eleventh, and twelfth inventions, for example, a sound such as a seek sound generated during a seek operation in a hard disk drive is provided. By avoiding the adaptive operation every time for the noise that occurs intermittently in the Can have a result, it is possible to obtain the acoustic noise reduction method which can quickly pull rate adaptation process.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an LMS operation algorithm circuit via an FIR filter.

FIG. 2 is a block diagram showing a specific circuit of an LMS adaptive filter in the circuit of FIG.

FIG. 3 is a block diagram illustrating an example of an acoustic noise reduction device according to an embodiment of the present invention.

FIG. 4 is a block diagram showing a specific circuit of a digital servo circuit in the circuit of FIG. 3;

FIG. 5 is a characteristic curve diagram showing an example of a seek operation in the HDD of the acoustic noise reduction device of FIG. 3;

FIG. 6 is a characteristic curve diagram showing vibration control by adaptive processing of the acoustic noise reduction device of FIG. 3;

FIG. 7 is a block diagram showing another example of the acoustic noise reduction device according to the embodiment of the present invention.

8 is a block diagram showing a specific circuit of an environmental sound level detection circuit in the acoustic noise reduction device of FIG.

[Explanation of symbols]

20 magnetic head, 21 hard disk, 22 spindle motor, 23 voice coil motor (VCM), 2
4 read channels, 25 digital servo control circuits, 27 adders, 28 D / A converters, 29 VCM
Driver, 30 error microphone, 31 amplifier, 32 A
/ D converter, 33 adaptive filter, 34 FIR filter, 35 LMS (least mean square) calculation unit, 36 limiter.

Claims (16)

[Claims] 1. An acoustic energy wave generating means for generating an acoustic energy wave driven by a drive signal from a drive signal source; an adaptive filter using a signal based on the drive signal as a reference input; A delay unit having a predetermined transfer function to which the base signal is input, a least-mean-square calculating unit using the output of the delay unit as a reference input, and collecting the acoustic energy wave based on the output residual signal. An error microphone for inputting a signal to the least mean square calculation means; and a subtraction means for subtracting an adaptive output signal from the adaptive filter from the drive signal. An acoustic noise reduction device characterized in that acoustic noise is reduced. 2. The acoustic noise reduction device according to claim 1, further comprising limiter means for receiving an output residual signal from said error microphone, wherein said least square mean calculation is performed on an output signal from said limiter means. Acoustic noise reduction device characterized by inputting to means. 3. The acoustic noise reduction device according to claim 1, wherein a level detection means to which an output residual signal from said error microphone is inputted, and a detection output from said level detection means being compared with a threshold level. An acoustic noise reduction device comprising: a level comparison unit; and a switch unit that is inserted between the adaptive filter and the subtraction unit and whose on / off is controlled by a comparison result of the level comparison unit. 4. The acoustic noise reduction device according to claim 2, wherein a level detection unit to which an output residual signal from the error microphone is input, and a detection output from the level detection unit is compared with a threshold level. An acoustic noise reduction device comprising: a level comparison unit; and a switch unit that is inserted between the adaptive filter and the subtraction unit and whose on / off is controlled by a comparison result of the level comparison unit. 5. The acoustic noise reducing device according to claim 1, wherein when acoustic noise is generated by the acoustic energy wave generating means, a step gain coefficient in the least mean square calculating means is set to a predetermined value. When no acoustic noise is generated by the acoustic energy wave generating means,
An acoustic noise reduction device, wherein the step gain coefficient in the least mean square calculation means is set to zero so that the coefficient of the adaptive filter is not updated. 6. The acoustic noise reduction device according to claim 2, wherein when acoustic noise is generated by the acoustic energy wave generating means, a step gain coefficient in the least mean square calculating means is set to a predetermined value. When no acoustic noise is generated by the acoustic energy wave generating means,
An acoustic noise reduction device, wherein the step gain coefficient in the least mean square calculation means is set to zero so that the coefficient of the adaptive filter is not updated. 7. The acoustic noise reduction device according to claim 3, wherein when acoustic noise is generated by the acoustic energy wave generating means, a step gain coefficient in the least mean square calculating means is set to a predetermined value. When no acoustic noise is generated by the acoustic energy wave generating means,
An acoustic noise reduction device, wherein the step gain coefficient in the least mean square calculation means is set to zero so that the coefficient of the adaptive filter is not updated. 8. The acoustic noise reduction device according to claim 4, wherein when acoustic noise is generated by said acoustic energy wave generating means, a step gain coefficient in said least mean square calculating means is set to a predetermined value. When no acoustic noise is generated by the acoustic energy wave generating means,
An acoustic noise reduction device, wherein the step gain coefficient in the least mean square calculation means is set to zero so that the coefficient of the adaptive filter is not updated. 9. An acoustic energy wave generating means is driven by a drive signal from a drive signal source to generate an acoustic energy wave, and a signal based on the drive signal is input to an adaptive filter as a reference input, and Inputting the based signal to delay means having a predetermined transfer function, inputting the output of the delay means to the least mean square calculating means as a reference input, collecting the acoustic energy wave by an error microphone,
A signal based on the output residual signal is input to the least-mean-square calculating means, an adaptive output signal from the adaptive filter is input to a subtracting means, and is subtracted from the driving signal. Acoustic noise reduction method characterized by reducing acoustic noise of the sound. 10. The acoustic noise reduction method according to claim 9, wherein an output residual signal from said error microphone is input to limiter means, and an output signal from said limiter means is input to said least mean square arithmetic means. A method for reducing acoustic noise, characterized in that: 11. The acoustic noise reduction method according to claim 9, wherein an output residual signal from said error microphone is inputted to a level detecting means, and a detection output from said level detecting means is inputted to a level comparing means. An acoustic noise reduction method, wherein the on / off of a switch inserted between the adaptive filter and the subtractor is controlled by a comparison result of the level comparator. 12. The acoustic noise reduction method according to claim 10, wherein an output residual signal from said error microphone is inputted to a level detecting means, and a detection output from said level detecting means is inputted to a level comparing means. An acoustic noise reduction method, wherein the on / off of a switch inserted between the adaptive filter and the subtractor is controlled by a comparison result of the level comparator. 13. The acoustic noise reduction method according to claim 9, wherein when acoustic noise is generated by said acoustic energy wave generating means, a step gain coefficient in said least mean square calculating means is set to a predetermined value. When no acoustic noise is generated by the acoustic energy wave generating means,
An acoustic noise reduction method, characterized in that the step gain coefficient in the least mean square calculation means is set to zero so that the coefficient of the adaptive filter is not updated. 14. The acoustic noise reduction method according to claim 10, wherein when acoustic noise is generated by said acoustic energy wave generating means, a step gain coefficient in said least mean square calculating means is set to a predetermined value. When no acoustic noise is generated by the acoustic energy wave generating means,
An acoustic noise reduction method, characterized in that the step gain coefficient in the least mean square calculation means is set to zero so that the coefficient of the adaptive filter is not updated. 15. The acoustic noise reducing method according to claim 11, wherein when acoustic noise is generated by said acoustic energy wave generating means, a step gain coefficient in said least mean square calculating means is set to a predetermined value. When no acoustic noise is generated by the acoustic energy wave generating means,
An acoustic noise reduction method, characterized in that the step gain coefficient in the least mean square calculation means is set to zero so that the coefficient of the adaptive filter is not updated. 16. The acoustic noise reduction method according to claim 12, wherein when acoustic noise is generated by said acoustic energy wave generating means, a step gain coefficient in said least mean square calculating means is set to a predetermined value. When no acoustic noise is generated by the acoustic energy wave generating means,
An acoustic noise reduction method, characterized in that the step gain coefficient in the least mean square calculation means is set to zero so that the coefficient of the adaptive filter is not updated.