close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JP2008197438

код для вставкиСкачать
Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JP2008197438
An object of the present invention is to commercialize a noise canceling system with a digital
configuration. SOLUTION: A signal obtained by collecting an external sound by a microphone is
converted into a digital signal by an A / D converter composed of a ΔΣ modulator and a
decimation filter and input to a noise canceling filter in a DSP, In the system of a noise canceling
system that generates a signal for noise cancellation and converts it into an analog signal by a D
/ A converter equipped with an interpolation filter and outputs it from the driver, the decimation
filter and the interpolation filter It comprises by minimum phase shift type FIR. [Selected figure]
Figure 5
Signal processing apparatus, signal processing method
[0001]
The present invention relates to a signal processing apparatus and method for performing signal
processing according to a predetermined purpose for an audio signal.
[0002]
There is known a so-called noise canceling system compatible with a headphone device, which is
designed to actively cancel external noises that are heard when playing back audio of content
such as music by using the headphone device. It has become
And as such a noise canceling system, two systems of a feedback system and a feedforward
10-04-2019
1
system are known roughly.
[0003]
For example, Patent Document 1 generates an audio signal obtained by inverting the phase of
noise inside a sound tube collected by a microphone unit provided in the vicinity of an earphone
unit in a sound tube attached to the user's ear, The configuration in which external noise is
reduced by outputting this as sound from the earphone unit 3, that is, the configuration of the
noise canceling system corresponding to the feedback method is described. Further, Patent
Document 2 has, as a basic configuration thereof, a configuration in which characteristics based
on a required transfer function are given to an audio signal obtained by collecting sound by a
microphone attached to a headphone device outer casing, and output from the headphone
device. That is, the configuration of the noise canceling system corresponding to the feed forward
method is described.
[0004]
JP-A-3-214892 JP-A-3-96199
[0005]
By the way, although it can be said about any of the above-mentioned feedback system and
feedforward system, what is put into practical use as noise canceling system of the headphone
apparatus in consumer equipment at present is what was comprised by the analog circuit. It has
become.
In order to effectively obtain the noise cancellation effect of the noise canceling system, for
example, the difference between the external unnecessary sound picked up by the microphone
and the sound outputted from the driver for the cancellation of the unnecessary sound It is
necessary to keep the difference within a certain level. In other words, in the noise canceling
system, it is required that the speed (response speed) from when the external unnecessary sound
is input until the cancellation sound corresponding to this is output is within a certain value.
However, when the noise canceling system is to be configured by digital circuits, the input and
the output are provided with an A / D converter and a D / A converter. In the processing time of
A / D converter and D / A converter widely used at present, the delay is considerably large when
considering adoption as a noise canceling system, and it is difficult to obtain an effective noise
10-04-2019
2
cancellation effect. For example, in the fields of military, industrial, etc., there are A / D
converters and D / A converters with high sampling frequency and low delay, but these are
extremely expensive and should be adopted in consumer equipment Is not realistic. This is the
reason why the noise canceling system is currently configured with an analog circuit instead of a
digital circuit.
[0006]
However, by replacing the analog circuit with a digital circuit, it is easier to change or switch the
characteristics or the operation mode, for example, without changing or replacing the constants
of physical component elements. If it is an audio related system such as a noise canceling system,
there are many advantages such as further improvement of the sound quality. Therefore, an
object of the present invention is, for example, to form a noise canceling system of a headphone
device for consumer use with a digital circuit, but to obtain a practically sufficient noise
canceling effect.
[0007]
Therefore, in consideration of the problems described above, the present invention is configured
as a signal processing device as follows. That is, after the delta sigma modulation processing is
performed on the input analog signal to convert it into the digital signal of the first format by the
predetermined quantization bit of 2 bits or more at the predetermined sampling frequency, the
digital signal of the first format An analog-to-digital conversion unit having a decimation filter for
converting the signal into a digital signal of the second type as a pulse code modulation signal
and outputting the digital signal; and inputting the digital signal after being output from the
analog-to-digital conversion unit Cancellation signal processing means for giving and outputting
a predetermined signal characteristic for canceling a predetermined cancellation target sound,
and a digital signal in the second format outputted from the cancellation signal processing
means. Digital signal that is converted to an analog signal by A log conversion process is
performed, and for the digital-to-analog conversion process, a predetermined sampling frequency
higher than the second format and a digital signal in the second format input than the second
format are used for the digital signal in the second format. Digital-to-analog conversion means
adapted to include an interpolation filter for performing processing of converting into a digital
signal of the third type with a small number of quantization bits, and at least one of a decimation
filter and an interpolation filter One of them is formed by a digital filter of a predetermined type
in which the signal delay is considered to be smaller than that of the linear phase type finite
impulse response system.
10-04-2019
3
[0008]
Depending on the above configuration, an analog signal including a component of the sound to
be canceled is converted to a digital signal by delta sigma modulation for the signal processing
system of the system that cancels (decreases, attenuates) the predetermined cancellation target
sound, and Analog-to-digital conversion means adapted to convert and output a pulse code
modulation (PCM) signal by a decimation filter, and a digital signal from the analog-to-digital
conversion means are input to a PCM signal format A cancellation signal processing means
adapted to give a signal characteristic for canceling a cancellation target sound by corresponding
digital signal processing, and a signal outputted from the cancellation signal processing means
are converted into an analog signal using ΔΣ modulation Digital-to-analog conversion means It
is assumed. In addition, with regard to the above-described decimation filter and the
interpolation filter included in the digital-analog conversion means, the signal delay between
input and output is better than that of a linear phase type finite impulse response (FIR) system. Is
formed by digital filters of a type that is considered to be small. This configuration makes it
possible to reduce the signal processing delay in the analog-to-digital conversion means and the
digital-to-analog conversion means.
[0009]
Then, as described above, the signal processing delay at the analog-to-digital conversion site and
the digital-to-analog conversion site is shortened, so that the system to cancel the cancellation
target sound under the digital system (noise cancel It is possible to satisfy the response speed
required for the signal processing system of the ring system). That is, it is possible to easily
realize the noise canceling system by the digital circuit method. And, by realizing a digital noise
canceling system, it is possible to implement functions that were difficult with analog circuits or
to improve the sound quality, which is useful for users. Is rising.
[0010]
As a best mode for carrying out the present invention (hereinafter referred to as an embodiment),
a headphone device equipped with a noise canceling system will be exemplified. Therefore, prior
to describing the configuration as the present embodiment, the basic concept of the noise
canceling system corresponding to the headphone device will be described.
10-04-2019
4
[0011]
As a basic method of such a noise canceling system compatible with a headphone device, a
method of performing servo control by a feedback method and a feed forward method are
known. First, the feedback method will be described with reference to FIG.
[0012]
Fig. 1 (a) schematically shows a model example of a noise canceling system by the feedback
method on the side of the right ear (R channel in 2-channel stereo by L (left) and R (right) of the
headphone wearer (user). Is shown. In the structure on the R channel side of the headphone
device here, first, the driver 202 is provided in a position corresponding to the right ear of the
user 500 wearing the headphone device in the housing unit 201 corresponding to the right ear. .
The driver 202 has the same meaning as a so-called speaker, and is driven by the amplified
output of the audio signal to emit the audio into space so as to be output.
[0013]
Then, as a feedback method, the microphone 203 is provided at a position near the right ear of
the user 500 in the housing portion 201. Depending on the microphone 203 provided in this
manner, the sound output from the driver 202 and the sound that is going to intrude into the
housing portion 201 from the external noise source 301 and reach the right ear, that is, the right
ear The external noise in the housing 302, which is an external sound to be generated, is picked
up. The noise source 301 leaks out as a sound pressure from, for example, a gap of an ear pad of
the housing as a cause of the generation of the noise 302 in the housing, or the housing of the
headphone device vibrates due to the sound pressure of the noise sound source 301. It can be
mentioned that this is transmitted into the housing part. In order to cancel (attenuate, reduce) the
noise 302 in the housing, such as a signal having an inverse characteristic to the sound signal
component of the external sound, from the sound signal obtained by collecting the sound by the
microphone 203, for example. And a voice signal for cancellation (voice signal for cancellation) is
generated, and this signal is fed back so as to be synthesized with the voice signal (audio source)
of the necessary sound for driving the driver 202. Thus, at the noise cancellation point 400 set at
a position corresponding to the right ear in the housing unit 201, the external sound is canceled
by synthesizing the output sound from the driver 201 and the component of the external sound.
10-04-2019
5
A sound is obtained and the user's right ear will hear this sound. Then, by giving such a
configuration also to the L channel (left ear) side, a noise canceling system as a headphone
device corresponding to normal L, R 2 channel stereo can be obtained.
[0014]
The block diagram of FIG. 1 (b) shows an example of a basic model configuration of the noise
canceling system by the feedback method. In FIG. 1 (b), as in FIG. 1 (a), a configuration
corresponding to only the R channel (right ear) side is shown, and an L channel (left The same
system configuration is provided for the side of the ear. Also, the block shown in this figure
indicates one specific transfer function corresponding to a specific circuit part, circuit system,
etc. in a system of a noise canceling system by a feedback method, and in this case, the transfer
function block Do. The character shown in each transfer function block represents the transfer
function of the transfer function block, and the voice signal (or voice) is transferred to the
transfer function indicated there each time it passes through the transfer function block. Is to be
given. First, the sound collected by the microphone 203 provided in the housing unit 201 is a
transfer function block corresponding to the microphone 203 and a microphone amplifier that
amplifies the electric signal obtained by the microphone 203 and outputs the sound signal. It will
be obtained as an audio signal via 101 (transfer function M). The voice signal having passed
through the transfer function block 101 is input to the synthesizer 103 via a transfer function
block 102 (transfer function -β) corresponding to an FB (FeedBack) filter circuit. The FB filter
circuit is a filter circuit in which the characteristic for generating the above-described canceling
audio signal is set from the audio signal obtained by collecting the sound by the microphone
203, and the transfer function thereof is represented as -β. It is
[0015]
Also, here, the audio signal S of the audio sound source, which is the content such as music, is
assumed to be subjected to equalization by the equalizer, and the synthesizer via the transfer
function block 107 (transfer function E) corresponding to this equalizer It is supposed to be
input to 13.
[0016]
In the synthesizer 103 here, the above two signals are synthesized by addition.
10-04-2019
6
The audio signal synthesized in this manner is amplified by the power amplifier and output as a
drive signal to the driver 202, whereby the driver 202 is output as audio. That is, the voice signal
from the synthesizer 103 passes through the transfer function block 104 (transfer function A)
corresponding to the power amplifier, and further passes through the transfer function block
105 (transfer function D) corresponding to the driver 202 as voice. It is released into space. The
transfer function D of the driver 202 is determined by, for example, the structure of the driver
202.
[0017]
Then, the voice output by the driver 202 passes through the transfer function block 106
(transfer function H) corresponding to the spatial path (spatial transfer function) from the driver
202 to the noise cancel point 400, and the noise cancel point 400 , And will be combined with
the in-housing noise 302 in the space here. Then, as the sound pressure P of the output sound
that is to reach, for example, the right ear from the noise cancellation point 400, the sound of the
noise sound source 301 entering from the outside of the housing unit 201 is canceled.
[0018]
In the noise canceling system model shown in FIG. 1 (b), the sound pressure P of the output
sound is N for the noise 302 in the housing and S for the audio signal of the audio source. The
transfer function shown in the transfer function block, M, −β, E, A, D, H, is represented as
follows. In the equation (1), focusing on the noise 302 in the housing, it can be seen that N is
attenuated by a coefficient represented by 1 / (1 + ADHMβ). However, in order for the system of
the equation shown in (Equation 1) to operate stably without oscillating in the noise reduction
target frequency band, it is necessary that
[0019]
Generally speaking, the absolute value of the product of each transfer function in the feedback
type noise canceling system is expressed by 1 << | ADHMβ |, and combined with the stability
determination of Nyquist in classical control theory. The equation (2) can be interpreted as
follows. Here, in the system of the noise canceling system shown in FIG. 1B, consider the system
represented by (−ADHMβ) obtained by cutting one portion of the loop portion related to N
which is the noise 302 in the housing . We will call this system "open loop" here. As an example,
10-04-2019
7
if the transfer function block 101 corresponding to the microphone and the microphone
amplifier and the transfer function block 102 corresponding to the FB filter circuit are portions
to be disconnected, the above-described open loop can be formed.
[0020]
The above-mentioned open loop is assumed to have, for example, the characteristics shown by
the Bode diagram of FIG. In this Bode line portion, the frequency is shown on the horizontal axis,
and the gain is shown on the lower half and the phase is shown on the upper half on the vertical
axis. When this open loop is targeted, the following two conditions need to be satisfied in order
to satisfy Equation (2) based on the stability determination of Nyquist. Condition 1: Phase 0 deg.
When passing through the point (0 degrees), the gain must be less than 0 dB. Condition 2: When
the gain is 0 dB or more, the phase 0 deg. Do not include the points of
[0021]
If the above two conditions 1 and 2 are not satisfied, the loop is subjected to positive feedback,
which causes oscillation (howling). In FIG. 2, phase margins Pa and Pb corresponding to the
above-mentioned condition 1 and gain margins Ga and Gb corresponding to the condition 2 are
shown. If the margin is small, the possibility of oscillation will increase due to various individual
differences among users who use the headphone device to which the noise canceling system is
applied, and variations in the state when the headphone device is worn. For example, in FIG. The
gain when passing through the point is smaller than 0 db, and according to this, gain margins Ga
and Gb are obtained. However, for example, temporarily phase 0 deg. The gain when passing
through the point is 0 dB or more and the gain margins Ga and Gb disappear, or the phase 0 deg.
If the gain when passing through the point is less than 0 dB, the gain margins Ga and Gb become
close to 0 dB, and oscillation will occur or the possibility of oscillation will increase. Similarly, in
FIG. 2, when the gain is 0 dB or more, the phase 0 deg. The phase margin Pa, Pb is obtained.
However, for example, when the gain is 0 dB or more, the phase 0 deg. It has passed the point of
Alternatively, phase 0 deg. When the phase margins Pa and Pb become smaller, the oscillation or
the possibility of oscillation increases.
[0022]
Next, in the configuration of the noise canceling system of the feedback type shown in FIG. 1 (b),
10-04-2019
8
in addition to the above-mentioned external voice (noise) canceling (reducing) function,
necessary sound (necessary sound) is The case of reproduction output will be described. Here, as
the required sound, for example, an audio signal S of an audio sound source as content such as
music is shown. In addition, as this audio | voice signal S, it is considered besides the thing of
such a musical thing or the content according to this. For example, when the noise canceling
system is applied to a hearing aid or the like, a microphone provided outside the housing to pick
up the surrounding necessary sound (different from the microphone 203 provided in the system
for noise cancellation) It becomes an audio signal obtained by sound collection. When applied to
what is called a so-called headset, it becomes an audio signal such as the other party's speaking
voice received by communication such as telephone communication. That is, the audio signal S
corresponds to general audio that needs to be reproduced and output according to the
application of the headphone device.
[0023]
First, in (Equation 1), the audio signal S of the audio source is focused. Then, it is assumed that
the transfer function E corresponding to the equalizer is set as having a characteristic according
to the equation represented by The transfer characteristic E is substantially the reverse
characteristic (1 + open loop characteristic) with respect to the open loop when viewed on the
frequency axis. Then, substituting the equation of the transfer function E shown by this equation
3 into the equation 1, the sound pressure P of the output sound in the model of the noise
canceling system shown in FIG. Can be represented. Of the transfer functions A, D, H shown in
the term of ADHS in (Eq. 4), first, the transfer function A corresponds to the power amplifier, the
transfer function D corresponds to the driver 202, and the transfer function H is noise from the
driver 202 Since it corresponds to the space transfer function of the path up to the cancellation
point 400, if the position of the microphone 203 in the housing unit 201 is in a position close to
the ear, the sound signal S has a noise cancellation function. It can be seen that the same
characteristics as a normal headphone can be obtained.
[0024]
Next, the noise canceling system by the feedforward method will be described. FIG. 3A shows a
configuration of the side corresponding to the R channel as an example of a noise canceling
system according to the feedforward method, as in FIG. 1A. In the feed-forward method, the
microphone 203 is provided outside the housing portion 201 in such a manner that voices
arriving from the noise source 301 can be picked up. Then, an external voice collected by the
microphone 203, that is, a voice that has arrived from the noise source 301 is collected to obtain
10-04-2019
9
a voice signal, the voice signal is subjected to an appropriate filtering process, and the cancel
voice signal is obtained. Will be generated. Then, this cancellation voice signal is synthesized with
the voice signal of the required sound. That is, the cancellation voice signal electrically simulating
the acoustic characteristic from the position of the microphone 203 to the position of the driver
202 is synthesized with the voice signal of the necessary sound. Then, by causing the driver 202
to output an audio signal in which the cancellation audio signal and the audio signal of the
required sound are synthesized in this manner, the noise obtained from the noise cancellation
point 400 includes the noise source 301 to the housing portion 201. It is made to be able to hear
what the sound which has invaded into the inside was canceled.
[0025]
FIG. 3B shows the configuration of the side corresponding to one channel (R channel) as a basic
model configuration example of the noise canceling system by the feedforward method. First, the
sound collected by the microphone 203 provided outside the housing portion 201 is obtained as
an audio signal via the transfer function block 101 having the transfer function M corresponding
to the microphone 203 and the microphone amplifier. Next, the voice signal that has passed
through the transfer function block 101 is input to the synthesizer 103 via a transfer function
block 102 (transfer function −α) corresponding to an FF (FeedForward) filter circuit. The FF
filter circuit 102 is a filter circuit in which the characteristic for generating the above-described
canceling audio signal is set from the audio signal obtained by collecting the sound by the
microphone 203, and the transfer function thereof is expressed as −α. It is
[0026]
Also, the audio signal S of the audio source here is directly input to the synthesizer 103. The
audio signal synthesized by the synthesizer 103 is amplified by the power amplifier and output
to the driver 202 as a drive signal, so that the driver 202 is output as audio. That is, also in this
case, the audio signal from the synthesizer 103 passes through the transfer function block 104
(transfer function A) corresponding to the power amplifier, and further the transfer function
block 105 (transfer function D) corresponding to the driver 202. It is emitted into space as voice
via. Then, the voice output by the driver 202 passes through the transfer function block 106
(transfer function H) corresponding to the spatial path (spatial transfer function) from the driver
202 to the noise cancel point 400, and the noise cancel point 400 , Where it will be synthesized
in space with the noise 302 in the housing.
10-04-2019
10
[0027]
Also, as shown by a transfer function block 110, the path from the noise source 301 to the noise
cancellation point 400 until the sound emitted from the noise source 301 intrudes into the
housing portion 201 and reaches the noise cancellation point 400. A transfer function (space
transfer function F) corresponding to is given. On the other hand, the microphone 203 picks up a
voice that is supposed to arrive from the noise source 301 which is an external voice, but at this
time, the sound (noise) emitted from the noise source 301 is the microphone 203. By the time it
reaches, as shown as the transfer function block 111, the transfer function (spatial transfer
function G) corresponding to the path from the noise source 301 to the microphone 203 will be
given. As an FF filter circuit corresponding to the transfer function block 102, a transfer function
-α is set in consideration of the above space transfer functions F and G. As a result, the sound
pressure of the noise source 301 entering from the outside of the housing portion 201 is
canceled as the sound pressure P of the output sound that is to reach the right ear, for example,
from the noise cancellation point 400.
[0028]
In the system of the noise canceling system according to the feedforward method shown in FIG.
3B, the sound pressure P of the output sound is N for the noise emitted from the noise source
301 and the audio signal of the audio source. After assuming S, the transfer functions
represented in each transfer function block, M, -α, E, A, D, H, are used to be represented as
follows. Also, ideally, the transfer function F of the path from the noise source 301 to the
cancellation point 400 can be expressed as Next, when the equation shown in (Equation 6) is
substituted into (Equation 5), the first term and the second term on the right side will be offset.
From this result, the sound pressure P of the output sound can be expressed as In this way, it is
indicated that the sound arriving from the noise source 301 is canceled, and only the audio
signal of the audio source is obtained as voice. That is, in theory, in the right ear of the user, the
noise-cancelled voice can be heard. However, in reality, it is difficult to configure a complete FF
filter circuit that can provide a transfer function such that equation (6) completely holds. In
addition, the shape of the ear by a person, the individual difference in how to wear a headphone
device is relatively large, and the change in the relationship between the noise generation
position and the microphone position is a noise particularly in the middle and high frequency
band. It is known to affect the reduction effect. For this reason, with regard to the middle and
high frequencies, active noise reduction processing is avoided, and passive sound insulation is
often performed mainly depending on the structure of the housing of the headphone device and
the like. Moreover, to describe for confirmation, (Equation 6) means imitating the transfer
function of the path from the noise source 301 to the ear with the electric circuit including the
10-04-2019
11
transfer function -α.
[0029]
Further, in the noise canceling system of the feedforward method shown in FIG. 3A, since the
microphone 203 is provided on the outside of the housing, the noise canceling system of the
feedback method of FIG. However, it can be arbitrarily set in the housing portion 201 in
accordance with the listener's ear position. However, normally, the transfer function -α is fixed,
and at the design stage, it is determined that some target characteristic is targeted. On the other
hand, the shape of the ear differs depending on the listener. For this reason, there is also a
possibility that a phenomenon such as a sufficient noise cancellation effect can not be obtained
or noise components are added in a non-reversed phase to cause abnormal noise. From these
facts, it is generally considered that it is difficult to obtain a sufficient amount of noise
attenuation (cancellation amount) although the feedforward method has low possibility of
oscillation and high stability. On the other hand, it is said that the feedback method needs
attention to the stability of the system instead of expecting a large noise attenuation amount. As
described above, the feedback method and the feed forward method each have features.
[0030]
By the way, as a present condition, the noise canceling system of the headphone device which is
practically used in the consumer and is put to practical use is an analog system adopting an
analog circuit. However, if the signal processing system of the noise canceling system is digital
signal processing, it is easy to provide various functions such as changing the characteristics of
the noise canceling system and the operation mode, switching, etc. It is possible, and high sound
quality can be achieved. Thus, the merit of digitizing the noise canceling system is great.
[0031]
Therefore, FIG. 4 shows one configuration example that can be considered properly when it is
assumed that a noise canceling system of a headphone device is constructed using digital devices
known at present. The noise canceling system shown in this figure is configured based on the
feed forward method shown in FIG. Also, although the headphone device (hereinafter simply
referred to as a headphone) 1 shown here corresponds to a two-channel stereo by L (left) and R
(right), the system configuration of this figure is L It corresponds to either channel or R channel.
10-04-2019
12
Also, in this figure, in order to make the explanation simple and easy to understand, the signal
system of the audio sound source to be originally heard is omitted, and only the system for
canceling the external sound (noise source) is shown. .
[0032]
In FIG. 4, first, the microphone 2 is for picking up an external sound including an external sound
(external noise) around the headphone 1 to be canceled. In the case of the feed forward system,
in general, the microphones 2 are generally provided to the outside of the casing corresponding
to one side channel of each of L and R of the headphone 1. Note that in this figure, the
microphone 2 provided corresponding to one of L and R channels is shown. A signal obtained by
collecting an external sound by the microphone 2 is amplified by the amplifier 3 and is input to
the A / D converter unit 4 as an analog audio signal.
[0033]
The A / D converter unit 4 in this case is, for example, one component or device, and PCM (Pulse
Pulse Number) (P is a natural number) having a predetermined sampling frequency and a
quantization bit number a (a is a natural number) Code Modulation) Convert to digital signal in
signal format and output. For this purpose, for example, as shown in the drawing, the ΔΣ
modulator 41 and the decimation filter 42 are provided. The ΔΣ (delta sigma) modulator 41
converts the input analog audio signal into a 1-bit digital signal at a predetermined sampling
frequency. This digital signal is lowered to a predetermined sampling frequency by the
decimation filter 42, and is converted into a PCM signal of a type in which the number of
quantization bits is a predetermined number of bits a larger than 1 and the A / D converter unit
It is assumed that the output of 4. Further, in the device as such an A / D converter unit 4, in
general, the above-mentioned decimation filter 42 is formed by a linear phase FIR (Finite Impulse
Response) system (linear phase FIR) having linear phase characteristics. It is formed. The digital
signal to be processed in this noise canceling system is an audio signal. Therefore, on the premise
of faithful sound reproduction, it is ideally found that no distortion of the waveform occurs. If
linear phase characteristics are given by FIR, the above waveform distortion does not occur. In
addition, in the case of an FIR system, it is possible to easily obtain an accurate linear phase
characteristic, as is well known. For this reason, the digital filter as the decimation filter 42 is
configured by the linear phase FIR. In order to make the digital filter of the FIR system linear
phase type, for example, with respect to tap coefficients, peak values of coefficients are set at the
centers of the number of taps (orders) to be symmetrical. It can be realized by setting.
10-04-2019
13
[0034]
The digital signal output from the A / D converter unit 4 is input to the DSP 5. The DSP 5 in this
case is a part that executes required signal processing for generating an audio signal of at least
sound to be output from the driver 1 a of the headphone 1 by digital signal processing. As
understood from the following description, the audio signal to be output from the driver 1a of
the headphone 1 is an audio signal of a digital audio source and an audio for allowing the
external sound collected by the microphone 2 to be canceled and canceled. The signal (voice
signal for cancellation) is synthesized. The DSP 5 is provided as, for example, one chip or device,
and is formed as digital signal processing corresponding to a predetermined PCM signal format.
[0035]
In this figure, a noise cancellation filter 51 is shown as a signal processing functional block
implemented in the DSP 5. The noise canceling filter 51 is also formed as a digital filter. The
noise cancellation filter 51 corresponds to the FF filter circuit shown in FIG. 3 and is a digital
signal output from the A / D converter unit 4, that is, a digital audio signal corresponding to the
external sound collected by the microphone 11. Enter Then, using this input signal, as a sound to
be output from the driver 1a, an audio signal of a sound having an action of canceling external
sound that reaches the ear of the headphone wearer corresponding to the driver 1a and is heard
(canceling voice Signal). As the simplest canceling audio signal, for example, an audio signal input
to the noise canceling filter 51, that is, an audio signal obtained by picking up an external sound,
has an inverse characteristic and an antiphase. Is a signal that In addition, in practice, a
characteristic (corresponding to the transfer characteristic −α in FIG. 3) in consideration of the
transfer characteristic such as a circuit or space in the system of the noise canceling system is
given.
[0036]
If, for example, a digital signal as an audio sound source is to be input in practice, for example, a
synthesizer is provided in the DSP according to FIG. 3, and the output signal of the noise
canceling filter 51 and the signal of the audio sound source And can be synthesized. Therefore, in
practice, the sampling frequency and the number of quantization bits a of the signal input to and
output from the noise cancellation filter 51 are determined to match the sampling frequency and
the number of quantization bits of the digital signal of the audio source. It is assumed.
10-04-2019
14
[0037]
The digital signal from the noise cancellation filter 51, which is the output of the DSP 5 in this
case, is input to a DAC (D / A converter) / amplifier unit 6. The DAC / amplifier unit 6 is also, for
example, one chip component, which receives the PCM format digital signal converted by the A /
D converter unit 4 described above and converts it into an analog signal. For example, as shown
in the figure, the configuration includes an interpolation filter 61, a noise shaper 62, a PWM
circuit 63, and a power drive unit 64.
[0038]
The digital signal from the DSP 5 (the noise canceling filter 51) input to the DAC amplifier unit 6
is first input to the interpolation filter 61. In the interpolation (oversampling) filter 61, the
sampling frequency is obtained by multiplying the sampling frequency by a factor represented
by a power of 2 by a predetermined factor (2, 4, 8, 16...) Convert and output as you pull up. In
addition, with regard to the number of quantization bits of the output signal, conversion is
performed such that the number of bits b is a predetermined value smaller than a in this case.
The interpolation filter 61 is also formed by a linear phase FIR system for the same reason as the
decimation filter 42 described above.
[0039]
The digital signal output from the interpolation filter 61 is subjected to noise shaping processing
by the noise shaper 62. The signal after this noise shaping has, for example, a sampling
frequency obtained by multiplying the sampling frequency at the input by a predetermined
factor (2, 4, 8, 16...) By a factor of 2 and the number of quantization bits Is converted to a format
according to the number of quantization bits c, which is smaller than a bits at the input and is
close to 1 bit. As is well known, noise shaping is obtained as a result of ΔΣ modulation
processing, and thus the noise shaper 62 can be realized by a ΔΔ modulator. That is, the digital
noise canceling system shown in this figure adopts a configuration to which ΔΣ modulation is
applied for A / D conversion and D / A conversion.
[0040]
10-04-2019
15
The output of the noise shaper 62 is PWM-modulated by a PWM (Pulse Width Modulation)
circuit 63 and converted to a 1-bit string signal, and then input to the power drive unit 64 in the
subsequent stage. The power drive unit 64 is formed of, for example, a switching drive circuit
that switches and amplifies a 1-bit string signal at high voltage, and a low pass filter (LC low pass
filter) for converting the amplified output into an audio signal waveform. An amplified output as
an audio signal is obtained. Here, the output of the power drive unit 64 is used as the output of
the DAC / amplifier unit 6. The amplified output from the DAC / amplifier unit 6 is supplied as a
drive signal to the driver 1a via the capacitor C1 for DC insulation after the filter 7 removes a
predetermined unnecessary band component.
[0041]
The sound output from the driver 1a driven in this manner is a combination of the sound
component of the digital audio source and the sound component of the noise cancellation audio
signal, but the sound of the noise cancellation audio signal is Depending on the component, an
external sound reaching the ear corresponding to the driver 1a from the outside is canceled
(canceled). As a result, ideally, the external sound is canceled and the sound of the digital audio
source is relatively emphasized as the sound that the headphone wearing person listens to by the
ear corresponding to the driver 1a.
[0042]
The configuration shown in FIG. 4 uses, for example, an A / D converter, a DSP, a D / A converter,
etc. which are easily available for consumer use, and in the present situation, for example, as a
digital noise canceling system In the case of trying to create an audio source such as a CD, the
configuration can be properly considered first.
[0043]
However, in the above configuration, it has been found that it is difficult to obtain a sufficient
noise cancellation effect in reality.
This means that the signal processing time (propagation time) possessed by the actual devices as
the A / D converter unit 4 and the DAC / amplifier unit 6, that is, the signal delay between the
10-04-2019
16
input and output (signal delay) is considerably large. Is the reason. Essentially, these devices are
supposed to handle audio signals as audio sources such as ordinary music pieces in a single way,
so even if the signal processing causes delays, this is not a problem It is However, when such a
device is used as it is for a noise cancellation ring system, the delay becomes so large that it can
not be ignored. That is, even in the entire system of the noise canceling system configured using
these devices, there is a large delay in the time (response speed) from the external sound picked
up by the microphone 2 to the sound output by the driver. It will occur. Due to this delay, for
example, it becomes difficult to cancel the external sound due to the sound component for noise
cancellation output from the driver. For example, even if only the A / D converter unit 4 is taken,
if the delay at a sampling frequency of 44.1 KHz is for 40 samples, the phase rotation of a signal
of about 550 Hz or more will be 180 ° or more. When the delay is increased to such an extent,
not only it is difficult to obtain the noise cancellation effect, but also there may be a phenomenon
that the external sound is emphasized. As described above, in the configuration of the digital
noise canceling system as illustrated in FIG. 1, the allowable noise canceling effect is limited to
the range of the frequency band lower than about 550 Hz, for example, the audible band. Even in
comparison with the standard case of 20 Hz to 20 kHz, the noise cancellation effect can be
obtained only in a very narrow frequency range. That is, it is difficult to obtain a noise
cancellation effect that is sufficient for practical use. This is the reason that most of the noise
canceling systems of headphone devices currently put into practical use are analog systems.
[0044]
However, as described above, the advantages obtained by digitizing the noise canceling system
are significant. Therefore, in the present embodiment, as will be described later, a configuration
for achieving the practical use of the noise canceling system of the headphone device while
eliminating the problem of the above-mentioned signal delay while using a digital method.
suggest.
[0045]
FIG. 5 is a block diagram showing a configuration example of a noise canceling system according
to a first embodiment of the present invention. In this figure, the same parts as those in FIG. The
overall configuration of the noise canceling system according to the first embodiment shown in
FIG. 5 is similar to that of the noise canceling system shown in FIG. Then, in the present
embodiment, first, in the A / D converter unit 4, the minimum phase shift type digital filter
(minimum phase shift type FIR) by the FIR system, instead of the linear phase type FIR
decimation filter 42 of FIG. , And the decimation filter 42A is provided. Similarly, in the DAC /
10-04-2019
17
amplifier section 6, in place of the linear phase FIR interpolation filter 61 of FIG. 4, an
interpolation filter 61A of minimum phase shift FIR is provided. As the basis of the minimum
phase shift type FIR digital filter, the tap coefficient can be formed by setting the peak value on
the head side (closer to the input) so that the minimum phase can be obtained as the FIR system.
.
[0046]
FIGS. 17 (b) and 17 (c) show characteristics of the linear phase FIR digital filter and the
characteristics of the minimum phase shift FIR. FIG. 17 (b) shows the impulse response waveform
and phase frequency response characteristics of the linear phase FIR, and FIG. 17 (c) shows the
impulse response waveform and phase frequency response characteristics of the minimum phase
shift FIR. Indicates The characteristics shown in FIGS. 17 (b) and 17 (c) are obtained as outputs
when a signal having frequency response characteristics of gain shown in FIG. 17 (a) is input.
First, the frequency response characteristic of the phase of the linear phase FIR shown in FIG.
17B indicates that the phase rotation is proportional to the frequency. That is, it shows that it is
the frequency response characteristic as a linear phase type. Also, looking at the impulse
response waveform, the output in response to the input has the number of taps (order), as can be
seen from the fact that the peak is obtained at a timing delayed for a certain period of time with
respect to the input timing. It shows that it has the delay (group delay) by the fixed time
according to. On the other hand, the frequency response characteristic of the phase of the
minimum phase shift type FIR in FIG. 17B indicates that the phase rotation with respect to
frequency is non-linear. Further, as the impulse response waveform, a peak is obtained at a
timing faster than the input timing, which corresponds to, for example, several taps. As
understood from the characteristics of FIGS. 17 (b) and 17 (c), as the same FIR digital filter, the
minimum phase shift type FIR has an output delay (, signal It can be said that the time of delay,
input / output delay) is very short.
[0047]
As described above with reference to FIG. 4, the noise canceling system may be configured by
actually adopting the devices as the A / D converter unit 4 and the DAC / amplifier unit 6 which
have been known so far. It is difficult to put them to practical use because the delay time of these
devices is too large. Looking at the inside in more detail, the delay due to the decimation filter 42
is dominant in the A / D converter unit 4 as the delay factor of these devices, and the delay due
to the interpolation filter 61 in the DAC / amplifier unit 6 is It has become dominant. Such a large
delay in the decimation filter 42 and the interpolation filter 61 is due to the constant group delay
10-04-2019
18
of the linear phase FIR as understood from FIG. 17 (b). The reason why the decimation filter 42
and the interpolation filter 61 are configured as a linear phase type FIR is that the processing
target is an audio signal, and it is inherent that no phase distortion or the like according to the
frequency occurs. It is derived from the fact that it is preferable. The linear phase type FIR causes
a group delay between input and output, but it is premised that the device up to now is used for
reproduction (recording) of an audio sound source that the user is actively listening to. As a
result, there were no particular problems. For example, in the case of reproducing an audio
source, even if the signal processing of the audio source is input to the signal processing device
and then reproduced as a sound, even if a corresponding delay due to the signal processing
occurs, This is nothing other than what is normally reproduced and output continuously, and
therefore, when the user reproduces and listens to the audio sound source, the delay in signal
processing is regarded as a problem. On the other hand, when trying to divert the previous
device to the noise canceling system instead of reproducing the audio source, the group delay of
the device can not obtain a phase that can cancel out the external sound, or It becomes difficult
and emerges as a problem.
[0048]
Therefore, in the present embodiment, the decimation filter 42A to be provided in the A / D
converter unit 4 and the interpolation filter 61A to be provided in the DAC / amplifier unit 6 are
each set to the minimum phase shift type FIR. Thereby, the signal delay in the decimation filter
42A and the interpolation filter 61A is greatly shortened. In this manner, the signal system for
noise noise processing shown in the figure is obtained by eliminating the dominant delay factor
of the D / A conversion site and the A / D conversion site in the noise canceling system. The
delay of the noise sound processing system is very short. In response to this, the frequency band
of voice in which noise cancellation is supposed to work effectively is also greatly expanded, and
as a result, a noise cancellation effect that is sufficient for practical use is obtained. That is, it is
possible to obtain a noise canceling system of a headphone device which can be put into practical
use although it is a digital system.
[0049]
As understood from the frequency response characteristic of the phase shown in FIG. 17C, in the
case of the minimum phase shift type FIR, phase distortion occurs according to the frequency.
Therefore, in the case of an audio signal, the possibility of the sound quality deterioration due to
this phase distortion is inevitable. This is the reason why, up to now, the linear phase type is used
for the digital filter implemented in the A / D converter or D / A converter compatible with audio
10-04-2019
19
signals. However, even if the signal processing target in this case is an audio signal, it is, for
example, an external sound to be subjected to noise cancellation, and the required reproduction
fidelity is considerably low compared to, for example, an audio source. On top of that, the sound
components that are actually canceled and considered to have a large effect are the low
frequency band called so-called low frequency, and there is also a trade-off with the
characteristics of the device etc. Noise cancellation up to about several kHz is effective It is
considered to be sufficient for practical use if it works. From this point of view, for example, even
if a filter such as an A / D converter or a D / A converter is formed by the minimum phase shift
type FIR, there is hardly any defect in the sound quality from the viewpoint of noise cancellation.
It can be said. That is, in the present embodiment, the decimation filter at the A / D conversion
site and the interpolation filter at the D / A conversion site, assuming the application as a noise
canceling system, is better than the linear phase FIR. From the viewpoint of the minimum phase
shift type FIR being preferable and practical, the minimum phase shift type FIR is used.
[0050]
Subsequently, a noise canceling system according to a second embodiment will be described with
reference to FIG. In FIG. 6, the same parts as those in FIG. 5 corresponding to the first
embodiment are given the same reference numerals, and the description will be omitted or will
be simplified. Further, following the description of the second embodiment, the description will
be made in order up to the twelfth embodiment. Similarly, the same reference numerals are given
to parts that are the same as those in the diagram showing the configuration of the embodiment
after that, and description will be omitted or simplified.
[0051]
In the first embodiment previously shown in FIG. 5, the A / D converter unit 4, the DSP 5 and the
DAC / amplifier unit 6 are individually separate components. On the other hand, in the second
embodiment shown in FIG. 6, the functional parts provided in each of A / D converter unit 4, DSP
5 and DAC / amplifier unit 6 shown in FIG. It is configured to be housed in one chip as a large
scale integration chip 100 or a physical structural unit which is an integrated circuit part. That is,
among the integrated circuit components as the LSI chip 100, as shown in the figure, at least as
the noise cancellation filter 51 corresponding to the Δ 変 調 modulator 41, the decimation filter
42A of the minimum phase shift type FIR, and the DSP 5 of FIG. Circuit portions as a DSP block
5A formed including functional blocks, an interpolation filter 61A of a minimum phase shift type
FIR, a noise shaper 62, a PWM circuit 63, and a power drive circuit 64 are the same as those in
FIG. It is provided by the relationship of the output. In this case, for example, an Δ 変 調
10-04-2019
20
modulator 41 corresponding to the A / D converter unit 4 shown in FIG. 5, a decimation filter
42A, and an interpolation filter corresponding to the D / A converter unit 6 shown in FIG. 61A,
the noise shaper 62, the PWM circuit 63, the power drive circuit 64 and the like can be
configured by hardware and incorporated. The DSP block 5A can be configured to include, for
example, a CPU that executes a program for signal processing.
[0052]
Thus, if the A / D conversion site, the DSP equivalent site, and the D / A conversion site are
integrated into one chip, for example, the A / D converter unit 4, the DSP 5, and the DAC /
amplifier unit 6 are each The process of manufacturing is simplified and the efficiency is higher
than when configuring the noise canceling system using the independent chip.
[0053]
FIG. 7 shows a configuration example of a noise canceling system according to the third
embodiment.
As the configuration shown in this figure, for example, instead of the linear phase FIR decimation
filter 42A and the linear phase FIR interpolation filter 61A shown in FIG. 5 corresponding to the
first embodiment, A decimation filter 42B and an interpolation filter 61B by digital filters (IIR
filters) as an IIR (Infinite Impulse Response) system are respectively provided.
[0054]
FIG. 17 (d) shows the impulse response waveform for the IIR filter and the frequency response
characteristic of the phase. The characteristics shown in FIG. 17 (d) are also for the output when
the signal of the characteristics shown in FIG. 17 (a) is input. Further, it is assumed that the IIR
filter in this case is formed in a second order. Then, as can be seen in comparison with FIG. 17
(d), FIG. 17 (b) showing the characteristics of the linear phase type FIR described above, for
example, and FIG. 17 (c) showing the characteristics of the minimum phase shift type FIR. As the
impulse response waveform, a peak is obtained at a fast timing corresponding to, for example,
about several taps with respect to the input timing. That is, it has a very small input / output
delay as in the minimum phase shift FIR. Therefore, in the third embodiment, the decimation
filter 42B and the interpolation filter 61B are configured by IIR filters to obtain the same effect
as that of the previous embodiment.
10-04-2019
21
[0055]
FIG. 8 shows a configuration example of a noise canceling system according to the fourth
embodiment. In the noise canceling system shown in this figure, one functional portion provided
in each of the A / D converter unit 4, the DSP 5 and the DAC / amplifier unit 6 shown in FIG. 7 is
one chip as one LSI chip 100. It is configured to be stored in That is, as shown in the drawing, the
ΔΣ modulator 41, the decimation filter 42B of IIR, the DSP block 5A, the interpolation filter 61B
of IIR, the noise shaper 62, the PWM circuit 63, and the power drive circuit 64 in the LSI chip
100. Are provided in the same signal input / output relationship as in FIG.
[0056]
FIG. 9 shows a configuration example of a noise canceling system according to the fifth
embodiment. The noise canceling system shown in this figure adaptively changes the settings of
the filter characteristics of the decimation filter 42A, the noise canceling filter 51, and the
interpolation filter 61A based on the configuration of FIG. 5, for example. Configured to be able
to For this purpose, for example, an analysis and control processing unit 52 is provided in the
DSP 5 and, further, each of the A / D converter unit 4 and the DAC and amplifier unit 6 receives
an instruction from the analysis and control processing unit 52. A filter setting control unit 43
and a filter setting control unit 65 are provided. Further, in correspondence with the decimation
filter 42A, a filter parameter table TB1 storing filter parameters A1 to An is prepared as a filter
parameter which is a set of predetermined parameter items including tap coefficients and the like
for the decimation filter 42A. Do. Similarly, a filter parameter table TB2 storing the filter
parameters B1 to Bn is prepared corresponding to the noise canceling filter 51, and a filter
parameter storing the filter parameters C1 to Cn corresponding to the interpolation filter 61A.
Prepare table TB3. The filter parameter table TB2 corresponding to the noise canceling filter 51
in the DSP 5 is held in the RAM 8 used by the DSP 5. The writing of the filter parameter table
TB2 to the RAM 8 may be performed, for example, when the DSP 5 is activated.
[0057]
The analysis / control processing unit 52 in this case branches and inputs the digital audio signal
of the external sound input from the A / D converter unit 4 to the DSP 5 and executes analysis on
a predetermined item. Items to be analyzed here may include, for example, the level of an audio
10-04-2019
22
signal (i.e., the volume of an external sound) or the frequency band distribution characteristic
thereof. Then, based on the result of the analysis process, the analysis / control processing unit
52 first outputs a signal specifying one filter parameter in the filter parameter table TB1 to the
filter setting control unit 43. In response to this, the filter setting control unit 43 sets, in the
decimation filter 42A, the filter parameter designated from the filter parameters A1 to An stored
in the filter parameter table TB1. At the same time, the analysis / control processing unit 52
reads out, from the filter parameters B1 to Bn stored in the filter parameter table TB2 stored in
the RAM 8, the filter parameters specified based on the analysis processing result, The noise
cancellation filter 51 is set. Further, together with this, the analysis / control processing unit 52
outputs a signal specifying one filter parameter in the filter parameter table TB3 to the filter
setting control unit 65 based on the analysis processing result. The filter setting control unit 63
sets a filter parameter designated from among the filter parameters C1 to Cn stored in the filter
parameter table TB3 in the interpolation filter 61A according to the signal input.
[0058]
By executing the above-described operation, the filter characteristics of each of the abovedescribed digital filters are varied in accordance with the state of the sound that is obtained by
the microphone 2 at that time. . That is, the system of the noise canceling system shown in FIG. 9
is adapted to obtain the most effective noise canceling effect by adapting to, for example, the
volume and frequency band distribution of the external sound analyzed by the analysis / control
processing unit 52. To work. This means that as a noise canceling system, a stable and good
noise cancellation effect can always be obtained regardless of changes in ambient sound. The
variable setting of the filter characteristics as described above can be easily realized by using the
above-described filters as digital filters. Further, the analysis / control processing unit 52 in the
actual DSP 5 can be implemented as a function obtained by executing a program by a CPU
provided in the DSP 5, for example.
[0059]
FIG. 10 shows a configuration example of a noise canceling system according to the sixth
embodiment. In the noise canceling system shown in this figure, functional parts provided in
each of the A / D converter unit 4, the DSP 5 and the DAC / amplifier unit 6 shown in FIG. 9 and
the RAM 8 are one LSI chip 100. It is housed in a part. In FIG. 10, a portion as the RAM 8 of FIG.
9 provided in the LSI chip 10 is shown as a RAM block 8A. In this manner, the above-described
portions shown in FIG. 9 are included in the same device as the LSI chip 100, so that the analysis
and control processing unit 52 in the DSP block 5A can be configured to interpolate the
10-04-2019
23
decimation filter 42A. The variable setting of the filter parameter for the relationship filter 61A
can be directly executed without passing through the filter setting control units 43 and 65.
[0060]
FIG. 11 shows an example of the configuration of a noise canceling system according to the
seventh embodiment. The noise canceling system shown in this figure is developed based on the
configuration of FIG. 10, and includes not only the filter parameter table TB2, but also the filter
parameter table TB1 corresponding to each of the decimation filter 42A and the interpolation
filter 61A. The TB 3 is also to be held in the RAM block 8A. As described above, by integrating
the portions corresponding to the A / D converter, DSP, RAM, and D / A converter into one with
respect to the LSI chip 100, all of the plurality of filter parameter tables are LSI chips. A
configuration can be easily realized in which the internal RAM area, that is, the RAM block 8A is
held. In this case, the analysis / control processing unit 52 selects and determines filter
parameters to be set for the decimation filter 42A, the noise cancellation filter 51, and the
interpolation filter 61A based on the analysis processing result. Next, after the selected and
determined filter parameter is read out from each of the filter parameter tables TB1, TB2 and
TB3 held in the RAM block 8A, each of the parameter values possessed by these read filter
parameters is appropriately set as described above. It is set for each filter. For example, in the
case of FIG. 10, the filter parameter tables TB1 and TB3 are held not on the RAM 8 but on the
decimation filter 42A and the interpolation filter 61A side. For this purpose, for example, it is
necessary to provide a non-volatile storage element or the like, and to store the data of the filter
parameter tables TB1 and TB3 here. On the other hand, with the configuration of FIG. 11, these
hardware elements for storing the filter parameter tables TB1 and TB3 become unnecessary, so
the circuit scale as the LSI chip 100 is reduced. Therefore, it is possible to expect miniaturization
of the LSI chip, cost reduction and improvement in yield. Further, as described above, all the filter
parameter tables are held in the RAM 8, so that when the signal processing for noise cancellation
is executed, the arithmetic operation is executed inside the DSP 5 as needed, and the filter
parameter table is executed. It is also possible to generate and hold the filter parameters to be
stored in B. Thus, the degree of freedom and the flexibility in setting the filter characteristics are
increased.
[0061]
Next, a noise canceling system according to the eighth embodiment will be described. The eighth
embodiment has been described as the noise canceling system as the embodiment described
above as will be described later with reference to FIG. 12, and a system of a feedforward system
10-04-2019
24
based on the model shown in FIG. Not only that, but also a system of feedback system based on
the model shown in FIG.
[0062]
As described above, the feedback method and the feed forward method have the characteristic of
a relationship that is mutually traded off. For example, in the feedforward method described
above as the embodiment, the frequency band where noise can be effectively canceled
(attenuated) is wide and the stability of the system is high, but it is considered that a sufficient
amount of noise cancellation can not be obtained. ing. For this reason, depending on conditions
such as the positional relationship with the noise source, for example, it may not match the
transfer function of the system, and for example, the noise may not be canceled or may increase
in a specific frequency band. Is pointed out. In this case, in spite of the fact that noise
cancellation works effectively over a wide frequency band, a phenomenon occurs in which noise
is noticeable only in a specific frequency band. , It becomes difficult to feel the noise cancellation
effect. On the other hand, the feedback method is characterized in that although the frequency
band where noise can be canceled is narrow, a sufficient amount of noise cancellation can be
obtained. From this point of view, if the noise canceling system is constructed by combining the
feed forward method with the feedback method, it is possible to expect a better noise
cancellation effect than in the case based only on the feed forward method. That is, first, noise
cancellation is performed by the feedforward method, and noise noise components in a specific
frequency band that can not be canceled by this are effectively canceled by the feedback method.
This makes it possible to effectively cancel noise noise over a wide frequency band.
[0063]
A configuration example of a noise canceling system according to the eighth embodiment shown
in FIG. 12 will be described. In this figure, first, as a microphone unit corresponding to a feed
forward system, a digital microphone 2A is provided in place of the microphone 2 of the
embodiments described above. The digital microphone 2A is configured to be able to output an
audio signal (audio signal) based on the collected voice as a component or device as one
microphone unit in a predetermined digital signal format, for example As shown, a microphone
21 and a ΔΣ modulator 22 for converting a signal obtained by collecting the microphone into a
1-bit digital signal at a predetermined sampling frequency and outputting the digital signal are
formed. To be The signal output from the digital microphone 2A is input to the minimum phase
shift FIR decimation filter 42A in the LSI chip 100. The output of the decimation filter 42A is
input to the IIR filter 51-1 in the DSP block 5A, and is also branched and input to the analysis /
10-04-2019
25
control processing unit 52. The output of the IIR filter 51-1 is combined with another audio
signal to be described later in a combiner 54. The output of the synthesizer 54 is the output of
the DSP block 5A and is input to the interpolation filter 61A. The configuration of the subsequent
stage from the interpolation filter 61A is the same as that of FIG. 11 except that the filter 7 is
omitted. If the system of the above configuration is, for example, the IIR filter 51-1 as the noise
canceling filter 51 in the above embodiments, the noise canceling system by the feedforward
method shown in the above embodiments can be obtained. It is understood that the system of
[0064]
However, in this embodiment, the function as the noise canceling filter 51 (corresponding to the
FF filter circuit in FIG. 3) is the decimation filter 42A-1, the IIR filter 51-1, and the interpolation
filter 61A. Is realized by a filter circuit system consisting of That is, the processing for the middle
and high frequency bands can be performed by enabling the decimation filter 42A, which is an
FIR filter on the A / D conversion side, and the interpolation filter 61A, which is an FIR filter on
the D / A conversion side. These filter characteristics are set by processing the remaining low
frequency band by the IIR filter 51-1 in the DSP block 5A.
[0065]
Generally, with regard to digital filters used for A / D conversion processing and D / A conversion
processing, FIR tends to be adopted rather than IIR from the viewpoints of arithmetic accuracy
and stability. However, when performing noise cancellation at the audio level, in order to obtain
sufficient noise cancellation effect even in the low band by the FIR filter, the number of taps is
considered to be huge corresponding to the oversampling area. Will be required. Based on such
reasons, as the decimation filter 42A and the interpolation filter 61A, which are FIR filters
included in the A / D conversion processing and the D / A conversion processing, respectively,
according to the present embodiment, the middle and high height excluding low frequency It is
formed to perform signal processing in the range of area. On the other hand, for example,
assuming that it is possible to use a high-precision arithmetic unit inside the DSP (DSP block 5A),
for example, the low-range signal processing is performed in this DSP to perform IIR. Arithmetic
resources can be reduced by configuring the filter. That is, in the system according to the
feedforward method in the eighth embodiment, FIR filters on the A / D conversion processing
side and the D / A conversion processing side (decimation filter 42A-1, interpolation filter 61A)
And the IIR filter 51-1 in the DSP are operated in coordination for noise cancellation, thereby
reducing the input / output delay while reducing the amount of calculation (calculation
resources) It is what you are doing.
10-04-2019
26
[0066]
In addition, in the eighth embodiment, a system of the following feedback system is added as a
system of the above-mentioned feed forward system. First, the digital microphone 2B is provided
as a microphone unit corresponding to a feedback system. Similarly to the digital microphone 2A,
the digital microphone 2B also includes the microphone 21 and the ΔΣ modulator 22 to output
the sound obtained by the sound collection as a digital audio signal of one bit string. The digital
microphone 2A compatible with the feedforward method is attached, for example, so that
external sound (external noise) can be picked up outside the headphone case, whereas the digital
microphone 2B compatible with the feedforward method is The sound is output from the driver
1a and attached to a position where it can pick up a sound that is going to reach the headphone
wearer's ear. Therefore, depending on the digital microphone 2B, the sound produced by
combining the sound output from the driver 1a and the sound that intrudes from the headphone
case to which the driver 1a is attached and tries to reach the ear of the headphone wearer is
collected. , Is to output the audio signal.
[0067]
The audio signal output from the digital microphone 2B is input to the decimation filter 42A-1 in
the LSI chip 100. The decimation filter 42A-1 is a minimum phase shift type FIR, as with the
decimation filter 42A provided corresponding to the feedforward system. The signal output from
the decimation filter 42A-1 is input to the IIR filter 51-2 in the DSP block 5A, and is input to the
synthesizer 54 after being given predetermined characteristics here. . The system from the
interpolation filter 61A to the driver 1a to which the output of the synthesizer 54 (DSP block 5A)
is input is shared with the system corresponding to the feedforward system. Also in this feedback
type system, as a filter (corresponding to the FB filter circuit in FIG. 1) for giving a signal
characteristic for noise cancellation, as in the case of the above-mentioned feed forward type
system, It is realized by the decimation filter 42A-1 which is the minimum phase shift type FIR on
the D conversion side, the interpolation filter 61A which is the minimum phase shift type FIR on
the D / A conversion side, and the IIR filter 51-2 in the DSP block 5A. It is supposed to be
configured to In other words, the interpolation filter 61A is shared with a feedforward system,
and the decimation filter 42A-1 is configured to execute signal processing corresponding to the
middle and high frequencies. The filter 51-2 is configured to execute signal processing
corresponding to the low band.
10-04-2019
27
[0068]
Furthermore, in this case, a configuration is shown in which a digital audio signal of a
predetermined format as a digital audio source, which is a sound to be originally listened to by
the headphone device, is input. The signal of this digital audio source is input to the equalizer 53
formed in the DSP block 5A, subjected to processing such as equalization according to the
purpose such as sound quality adjustment, and then output to the synthesizer 54. It will be The
signal component of the digital audio source output from the synthesizer 54 is also output as
sound from the driver 1a through the system from the interpolation filter 61A to the driver 1a.
[0069]
In this configuration, noise cancellation is performed by the above-described feedforward noise
canceling system and feedback noise canceling system so that only the reproduced sound of the
digital audio source can be clearly heard. As understood from the above description, in this
embodiment, the noise canceling system by the feed forward method and the noise canceling
system by the feedback method are simultaneously operated. As a result, as described above, the
feedforward method was used to first perform stable noise cancellation over a wide frequency
band, and the feedforward method could not sufficiently attenuate the noise. It is possible to
obtain a complex noise cancellation operation of effectively canceling and attenuating
components of noise noise of a specific frequency by a feedback method. As a result, for example,
a higher noise cancellation effect is expected than when noise cancellation is performed by a
system based on only one of the feedforward method and the feedback method.
[0070]
Furthermore, even in this case, noise is canceled by providing the analysis / control processing
unit 52 in the DSP block 5A and holding the filter parameter tables TB11, TB12, TB13, and TB3
in the RAM block 8A. It is possible to give a filter characteristic adapted to the change in the
external sound condition for the digital filter related to. In this case, the filter parameter table
TB11 is formed by storing filter parameters for setting filter characteristics for the decimation
filters 42A and 42A-1, which are both the minimum phase shift type FIR. The filter parameter
table TB12 is formed by storing filter parameters for setting the filter characteristics of the IIR
filter 51-1, and the filter parameter table TB13 is for setting the filter characteristics of the IIR
filter 51-2. Is formed by storing the filter parameters of Further, the filter parameter table TB3 is
formed, for example, by storing filter parameters for setting the filter characteristic of the
10-04-2019
28
interpolation filter 61A, as in the case of FIG.
[0071]
FIG. 13 shows an example of the configuration of a noise canceling system according to the ninth
embodiment. The noise canceling system shown in this figure includes the adaptive signal
processing unit 55 as a portion corresponding to the noise canceling filter in the DSP block 5A
based on the configuration of the eighth embodiment. To be The adaptive signal processing unit
55 receives noise sounds (cancel target sound respectively obtained by the microphone 2A
outside the headphone unit 1h (headphone case) and the microphone 2B inside) received from
the decimation filters 42A and 42A-1. Audio signal as an analog signal) and performing adaptive
signal processing using these audio signals to generate an audio signal for noise cancellation,
which is output to the synthesizer 54. In this way, by applying adaptive signal processing, an
audio signal for noise cancellation that is generated is always adapted to changes in the state (a
feature of a timbre or level) of the external sound at that time. Signal characteristics are provided
which are considered to provide the best noise cancellation effect. With such a configuration,
there is no need to prepare a filter parameter table, for example, for changing and setting the
characteristics of the digital filter for noise cancellation in the DSP. As a result, for example, the
area size required to hold the filter parameter table in the RAM 8, RAM block 8A, etc. is reduced,
and the storage capacity of the RAM 8, RAM block 8A can be saved accordingly, and the RAM 8,
RAM block can be used. It also becomes possible to set a small storage capacity as 8A. Further,
the decimation filters 42A and 42A-1 and the interpolation filter 61A, for example, in the same
manner as the eighth embodiment shown in FIG. By setting filter parameters selected and read
from the filter parameter tables TB11 and TB3 held in the RAM block 8A, more efficient
adaptation of filter characteristics can be achieved.
[0072]
In the ninth embodiment, there is no particular limitation on what kind of method and system is
adopted as the adaptive signal processing unit 55. Incidentally, as a method of adaptive signal
processing, learning by LMS (Least Mean Square) algorithm or learning identification method is
known. Also, in the eighth embodiment shown in FIG. 12, digital microphones 2A and 2B are
used as microphone units, but in the ninth embodiment shown in FIG. 13, ordinary microphones
2A and 2B are used. It is assumed that 1-bit quantization of the signal obtained by the sound
collection is performed by the Δ 、 modulators 41 and 41-1, respectively. The microphones 2A
and 2B correspond to the feedforward method and the feedback method, respectively, like the
digital microphones 2A and 2B in FIG. 12. First, the microphone 2A is external to the outside of
10-04-2019
29
the headphone unit (headphone housing) 1h. The microphone 2B is attached so as to be able to
pick up a sound (external noise), and is attached to a position where it can pick up a sound
output from the driver 1a and reaching the ear of the headphone wearer. For confirmation, in the
eighth embodiment of FIG. 12 as well, the portions as the microphones 21 and 21 in the digital
microphones 2A and 2B are provided to the headphone unit in the same manner as described
above. is there.
[0073]
A specific configuration example of the adaptive signal processing unit 55 shown in FIG. 12 is
shown as a tenth embodiment in FIG. The adaptive signal processing unit 55 shown in this figure
first comprises an FIR filter 56a and an adaptive filter 56b. The filter 56a gives a signal
characteristic according to the filter characteristic represented by γ2 to the signal input from
the decimation filter 42A corresponding to the microphone 2A outside the headphone unit 1h,
and outputs the signal characteristic to the synthesizer 54. The output of the filter 56a is an
audio signal for noise cancellation.
[0074]
Further, the adaptive filter 56b in this case is formed based on the LMS method, and includes a
characteristic variable filter 57a and a synthesizer 57b. The characteristic variable filter 57a is,
for example, an FIR filter and is also referred to as an adaptive linear combiner.
[0075]
As illustrated, the characteristic variable filter 57a uses as an input signal a digital signal output
from the decimation filter 42A corresponding to the microphone 2A outside the headphone unit
1h, and outputs a signal characteristic according to the filter characteristic γ1. The synthesizer
57b is adapted to obtain the difference between the signal output from the filter 57a and the
signal from the decimation filter 42A-1 corresponding to the microphone 2B inside the
headphone unit 1h. The characteristic variable filter 57a is made to change its own filter
characteristic γ1 based on the above-mentioned difference in accordance with a predetermined
algorithm given in advance. In this way, the filter characteristic γ1 that is optimal for noise
cancellation is obtained. Then, in the adaptive signal processing unit 55, the filter 56a is
controlled to have the same filter characteristic as the filter characteristic γ1 set in the
10-04-2019
30
characteristic variable filter 57a as described above. For example, a predetermined parameter
actually obtained as the filter characteristic γ1 is also set to the filter 56a at a predetermined
timing. As a result, the noise canceling audio signal output from the filter 56a has characteristics
such that a noise canceling effect that is always optimal is generated in accordance with the
external sound condition and the like.
[0076]
FIG. 15 shows an example of the configuration of a noise canceling system according to the
eleventh embodiment. The noise canceling system shown in this figure is based on, for example,
the seventh embodiment shown in FIG. In addition to the noise cancellation filter 51 and the
analysis / control processing unit 52, the DSP block 5A provided in the LSI chip 100 is further
configured to include the decimation filter 42A, the interpolation filter 61A, the noise shaper 62,
and the PWM circuit 63. It is formed. That is, for the decimation filter 42A, the interpolation filter
61A, the noise shaper 62, the PWM circuit 63, etc., the circuit is not formed in hardware in the
LSI chip 100, but is realized in DSP more. It is intended to be realized by the function.
[0077]
FIG. 16 shows a configuration example of a noise canceling system according to the twelfth
embodiment. As the noise canceling system shown in this figure, for example, a circuit formed in
the LSI chip 100 also as a part as the amplifier 3 and the filter 7 shown as the one outside the
LSI chip 100 in FIG. Further, sound collecting and sound emitting parts such as the microphone 2
and the driver 1 a are to be integrated with the LSI chip 100. In such a configuration, the
microphone 2 may be formed on the LSI chip 100 using the technology of MEMS (Micro Electro
Mechanical Systems). As for the driver 1a, it is conceivable to adopt a capacitor driver or the like
as illustrated. This capacitor driver can also be formed by applying the technology of MEMS.
[0078]
In the description of the embodiments described above, a noise canceling system based on a feed
forward system or a noise canceling system combining a feedback system with a feed forward
system has been described as an example. However, even in the case of a noise canceling system
configured based only on the feedback method, when a proper configuration is considered, for
example, the A / D conversion site and the D / A conversion site are the same A / D as in FIG. The
10-04-2019
31
configurations of the converter unit 4 and the DAC / amplifier unit 6 are adopted. According to
this, the configuration based on the present invention can be applied also to the noise canceling
system configured based only on the feedback method, and a sufficient effect can be expected.
[0079]
Further, regarding the configuration including the LSI chip 100 in the embodiments described
above, it is necessary to include all parts as the A / D converter unit 4, the DSP 5 (DSP block 5),
and the DAC / amplifier unit 6. Is integrated into the LSI chip 100. However, depending on the
case, for example, the DSP 5 may be configured as an independent chip or package component
for the LSI chip 100. Further, under the present invention, the LSI chip 100 may be treated not
only as one bare chip but also as one chip part, that is, one part called a package. In this case, a
plurality of chips (bare chips) corresponding to functional parts as the A / D converter unit 4, the
DSP 5 (DSP block 5), and the DAC / amplifier unit 6 are mounted in a package as the LSI chip
100. Will be configured. Further, in this case, for example, three chips corresponding to each of
the A / D converter unit 4, the DSP 5 (DSP block 5), and the DAC / amplifier unit 6 may be
mounted. Any of the above may be integrated into one chip. Furthermore, at least one of the A /
D converter unit 4, the DSP 5 (DSP block 5), and the DAC / amplifier unit 6 is packaged after
being configured to be composed of two or more chips. It is also conceivable to For example, in
some cases, it may be more efficient to divide and manufacture the DAC / amplifier unit 6 etc.
into two chips by the system of amplification processing in the latter stage and the digital signal
processing system in the latter stage. it is conceivable that.
[0080]
In each of the above embodiments, the decimation filter at the A / D conversion site and the
interpolation filter at the D / A conversion site are both configured as the minimum phase shift
type FIR, or both are configured as the IIR filter. However, it is also possible to adopt a
configuration in which one of the two filters is the minimum phase shift FIR and the other is the
IIR filter. As understood from the above description, the minimum phase shift type is one type of
FIR filter. Therefore, the minimum phase shift type digital filter has a high degree of freedom in
design and high stability with respect to the amplitude characteristics with respect to frequency.
However, when the filter order (the number of taps) becomes long, a large number of operation
resources for multiplication and addition are required. On the other hand, the IIR filter can
significantly reduce operation resources, but it is difficult to obtain high stability, for example,
errors accumulate when the accuracy of the operation unit is originally low. There is. In addition,
there are limitations in designing the amplitude characteristics. In this way, the FIR filter and the
10-04-2019
32
IIR filter have different characteristics in such a way as to make a trade-off with each other.
Therefore, in consideration of, for example, the above characteristics and conditions for actually
constructing a noise canceling system, any of the minimum phase shift FIR and IIR filters for the
above decimation filter and interpolation filter is used. Determining whether to adopt it is also
considered to be effective in order to obtain a high-performance noise canceling system
according to the present embodiment. Furthermore, for each of the decimation filter and the
interpolation filter (or any one thereof), for example, two systems of one by the minimum phase
shift type FIR and one by the IIR filter are prepared, and the use environment and conditions at
that time. It is also conceivable to adopt a configuration in which either one of the systems can be
adaptively selected according to the analysis content of the external sound or the like.
Alternatively, it may be considered that such selection of the filter can be performed according to
the user operation.
[0081]
Furthermore, according to the present invention, the minimum phase shift FIR or IIR filter is
adopted only for at least one of the decimation filter for the A / D conversion site and the
interpolation filter for the D / A conversion site. It may be configured. Even with such a
configuration, for example, the delay of the signal processing system for noise cancellation is
shortened as compared with the case where both linear and phase decimation filters are adopted
for the above-mentioned decimation filter and interpolation filter. Therefore, the effects
according to the present invention are produced.
[0082]
In addition, as the decimation filter of the A / D conversion part and the digital filter used for the
interpolation filter of the D / A conversion part, the delay time is small enough to satisfy the
required noise cancellation effect, for example, sound quality, stability Other configurations such
as minimum phase shift FIR and IIR filters are also conceivable as long as other conditions such
as the above satisfy a certain level or more.
[0083]
In addition, how to actually mount each component for realizing the noise canceling system in
the embodiments described so far on a device, this point is actually based on the present
invention. The noise canceling system may be arbitrarily determined according to the apparatus
to which the noise canceling system is applied, the configuration of the system, the application,
10-04-2019
33
and the like.
For example, if it is intended to construct a headphone device having a noise cancellation
function by itself, mount almost all the components that are supposed to form the noise
cancellation system in the housing of the headphone device. Is considered. Alternatively, if the
noise canceling system is to be configured by a set of devices such as a headphone device and an
external adapter, at least one of the components other than the microphone and driver should be
mounted on the adapter side. Is considered. Furthermore, the noise canceling system according
to the present invention is not an headphone device but an audio reproducing device configured
to reproduce audio content and output it to a headphone terminal, a mobile phone device, a
network voice communication device, etc. If it is decided to mount, at least one of the
components other than the microphone and the driver may be mounted on the side of these
devices.
[0084]
It is a figure showing an example model about a noise canceling system of a headphone device by
a feedback method. It is a Bode diagram showing the characteristic about the noise canceling
system shown in FIG. It is a figure showing an example model about a noise canceling system of a
headphone device by a feedforward method. It is a block diagram showing an example of basic
composition of a noise canceling system of a headphone device by a digital method. BRIEF
DESCRIPTION OF THE DRAWINGS It is a block diagram which shows the structural example of
the noise canceling system as 1st Embodiment in this invention. It is a block diagram showing an
example of composition of a noise canceling system as a 2nd embodiment. It is a block diagram
which shows the example of a structure of the noise canceling system as 3rd Embodiment. It is a
block diagram showing an example of composition of a noise canceling system as a 4th
embodiment. It is a block diagram which shows the example of a structure of the noise canceling
system as 5th Embodiment. It is a block diagram which shows the example of a structure of the
noise canceling system as 6th Embodiment. It is a block diagram which shows the example of a
structure of the noise canceling system as 7th Embodiment. It is a block diagram which shows
the example of a structure of the noise canceling system as 8th Embodiment. It is a block
diagram showing an example of composition of a noise canceling system as a 9th embodiment. It
is a block diagram which shows the example of a structure of the noise canceling system as 10th
Embodiment. It is a block diagram which shows the structural example of the noise canceling
system as 11th Embodiment. It is a block diagram showing an example of composition of a noise
canceling system as a 12th embodiment. It is a figure which shows each characteristic of linear
phase type FIR, the minimum phase shift type FIR, and IIR as a characteristic of a digital filter.
10-04-2019
34
Explanation of sign
[0085]
DESCRIPTION OF SYMBOLS 1 headphone, 1a * 1 b driver (capacitor driver), 2 * 21 microphone,
2A, 2B digital microphone, 22 delta-sigma modulator, 3 amplifier, 4 A / D converter, 41 * 41-1
delta-sigma modulator, 42A * 42A- 1 decimation filter (minimum phase shift type FIR), 5 DSP, 5
A (DSP block), 6 DAC / amplifier section, 7 filter, 8 RAM, 8 A RAM block, C1 capacitor, 43/65
filter setting control section, 51 noise cancellation Filter, 51-1 and 51-2 IIR filter, 52 analysis and
control processing unit, 53 equalizer, 61A interpolation filter (minimum phase shift type FIR), 62
noise shaper, 63 PWM circuit, 64 power drive unit, 100 LSI
10-04-2019
35
Документ
Категория
Без категории
Просмотров
0
Размер файла
64 Кб
Теги
description, jp2008197438
1/--страниц
Пожаловаться на содержимое документа