close

Вход

Забыли?

вход по аккаунту

?

JP2010266599

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 JP2010266599
[Purpose] An object of the present invention is to provide a "multipoint adaptive equalization
control method and a multipoint adaptive equalization control system" for adaptively equalizing
and controlling observation sound at a plurality of control points in a vehicle cabin. An audio
signal output from an audio source is imparted with a phase characteristic opposite to that from
a speaker to a main control point, and an observation signal at each observation point of an
audio signal output from the speaker and each control point An error signal, which is the
difference from the target signal, is input, and adaptive signal processing is performed to
minimize the sum of the powers of the respective error signals to determine the gain of the audio
signal. And input to the speaker, [Figure]
Multipoint adaptive equalization control method and multipoint adaptive equalization control
system
[0001]
The present invention relates to a multipoint adaptive equalization control method and a
multipoint adaptive equalization control system, and more particularly to a multipoint adaptive
equalization control method and a multipoint adaptive control method for adaptively controlling
observation sound at a plurality of control points in a vehicle compartment. The present
invention relates to an equalization control system.
[0002]
In general, in acoustic space, a reflected wave, a standing wave and the like are generated by a
08-05-2019
1
wall or the like, and sound waves interfere with each other to disturb the acoustic transfer
characteristic in a complicated manner.
In particular, in a narrow space such as a passenger compartment enclosed with a glass-like
sound that is likely to be reflected, the influence of the disturbance of the acoustic transfer
characteristic on the listening of the sound is significant because the influence of the reflected
wave and the standing wave is large. Is big. An adaptive equalization control system is known as
a technique for correcting such disturbance of the acoustic transfer characteristic. According to
the adaptive equalization control system, a predetermined sound field space can be realized at
any control point.
[0003]
FIG. 6 is a block diagram of an adaptive equalization control system applied to an audio device
(Patent Document 1). The audio source (not shown) is composed of a radio tuner, a CD player,
etc., and outputs an audio signal x (n). The target response setting unit 61 is set with a target
response characteristic (impulse response) h, receives an audio signal x (n) output from the audio
source, and outputs a target response signal d (n) corresponding thereto. Output. The
microphone 62 is installed at a listening position (control point) in the vehicle interior sound
space, detects a sound at this control point, and outputs a music signal y (n). The calculation unit
63 calculates an error between the music signal y (n) output from the microphone 62 and the
target response signal d (n) output from the target response setting unit 61, and outputs an error
signal e (n). Adaptive signal processor 60 generates signal x '(n) such that the power of error
signal e (n) is minimized. The speaker 64 emits a sound according to x '(n) output from the
adaptive signal processing device 60 to the in-vehicle acoustic space.
[0004]
The target response characteristic h of the target response setting unit 61 is set to a
characteristic corresponding to the sound field space to be reproduced. For example, when the
delay time corresponding to about half of the number of taps of the adaptive filter is t, a flat
characteristic (characteristic of gain 1) is set in all audio frequency bands with the delay time t.
The delay time t is for the adaptive filter to accurately approximate the inverse characteristic of
the acoustic system, and the target response setting unit 61 having such target response
characteristic is an FIR (Finite Impulse Response) type. This can be realized by setting the
coefficient of the tap corresponding to the delay time t of the digital filter to 1 and setting the
08-05-2019
2
coefficients of the other taps to 0.
[0005]
The adaptive signal processing device 60 receives the audio signal x (n) and the error signal e (n)
output from the above-described operation unit 63, and the power of the error signal e (n) is
minimum. The adaptive signal processing is performed so that the signal x ′ (n) is output. The
adaptive signal processing device 60 includes an LMS (Least Mean Square) unit 60a, an FIR filter
60b as an adaptive filter, and propagation characteristics of an acoustic propagation system from
the speaker 64 to the control point (microphone 62) for the audio signal x (n) ( Transmission
characteristic) A signal processing filter 60c for convoluting C to generate a reference signal
used for adaptive signal processing.
[0006]
The LMS unit 60a receives the error signal e (n) at the control point and the reference signal
output from the signal processing filter 60c, and the filter coefficient of the FIR filter at time n is
W (n). Using the LMS algorithm so that the signal x '(n) at the control point is equal to the target
response signal d (n) using the signal: W (n + 1) = W (n) + 2.mu.x (n). The filter coefficient of the
FIR filter 60b at time (n + 1) is set by C · e (n). The FIR filter 60b performs digital filter processing
on the audio signal x (n + 1) using the set filter coefficient, and outputs a signal x '(n + 1).
[0007]
If the filter coefficients of the FIR filter 60b converge so that the power of the error signal e (n)
becomes minimum by such adaptive processing, music is listened to in the space having the
target response characteristic h set in the target response setting unit 61. It becomes possible to
listen to the same music as in the case of
[0008]
By the way, although the above-mentioned adaptive equalization control system can listen to
music with the same transfer characteristic as the target response characteristic h at the control
point, it does not guarantee the characteristics of points other than the control point.
08-05-2019
3
Therefore, when trying to listen to ideal music at many positions in the acoustic space by the
adaptive equalization control system, many control points are set, and correspondingly many
speakers and microphones are needed. That is, in the multipoint adaptive equalization system, it
is necessary to install many speakers and microphones as controlled sound sources, and the
number of adaptive filters (FIR filters) is also increased, which causes a problem of an increase in
circuit size and computation amount. .
[0009]
Therefore, a multipoint adaptive equalization control system has been proposed that can correct
the transfer characteristic over the entire acoustic space with a small number of speakers and an
adaptive filter. FIG. 7 is a block diagram of such a multipoint adaptive equalization control
system, in which one loudspeaker and one adaptive filter are provided, and a microphone is
provided at each control point to perform multipoint adaptive equalization control. 7 differs from
FIG. 6 in that 1) microphones 621 to 62K are arranged at a large number (= K) of control points,
and 2) target response setting section 61 sets target response characteristics at each control
point. Point 3) A point of arranging operation parts 631 to 63 K for calculating the difference
between the observed sound signals y1 to yK at each control point and the target response
signals d1 to dK at the control point and outputting as error signals e1 to eK The adaptive signal
processing unit 60 calculates the following equation: W (n + 1) = W (n) + μ1 · C1 · e1 (n) · x1 (n)
+ μ2 · C2 · e2 (n) · x2 (n) · · · + μn · Cn · · · The filter coefficients of the adaptive filter 60b are
updated and set in the filter so that the sum of the powers of the error signals is minimized by eK
(n) .xK (n). In the adaptive signal processing device 60, measurement transfer characteristics
from the speaker to each control point are set in the signal processing filters 60c1 to 60cK. As a
result, it is possible to realize the characteristic that the audio signals at the K control points are
close to the desired signal.
[0010]
Japanese Patent Application Laid-Open No. 11-167383
[0011]
However, in the above-mentioned prior art, it operates so as to realize the characteristic that the
error of the audio signal becomes small on average at all of the plurality of control points.
08-05-2019
4
For this reason, when looking at the characteristics of the audio signal at each control point, the
characteristic is deteriorated from the ideal result of only one main control point in FIG. 6, and in
particular, the arrival time of the sound wave for each frequency There is a problem that
unnaturalness is left in the sense of hearing by performing average correction at a plurality of
control points with respect to the frequency phase characteristic representing.
[0012]
Therefore, it is an object of the present invention to remove auditory unnaturalness at one main
control point even when correction of a plurality of control points is performed using one
speaker and one adaptive signal processor. In addition, it is possible to improve the
characteristics at other control points other than the main control point.
[0013]
The present invention is a multipoint adaptive equalization control method and a multipoint
adaptive equalization control system for adaptively equalizing and controlling observation sound
at a plurality of control points in a vehicle cabin.
[0014]
-Multipoint adaptive equalization control method The multipoint adaptive equalization control
method of the present invention provides an audio signal output from an audio source with a
phase characteristic opposite to that from the speaker to the main control point, A sound is
output from the speaker according to the input audio signal given a characteristic, and an error
signal at each control point of the sound output from the speaker is input, and the sum of the
powers of the error signals is minimum The adaptive signal processing is performed to determine
the gain of the audio signal, and the audio signal is imparted with the antiphase characteristic
and the gain and input to the speaker.
[0015]
-Multipoint adaptive equalization control system The multipoint adaptive equalization control
system according to the present invention comprises a filter having a phase characteristic
opposite to that of the main control point and an audio signal input through the filter. An audio
signal, an error signal generator for outputting an error signal which is a difference between a
detection signal at each control point of the sound output from the speaker and a target signal at
each control point; An adaptive signal processing unit that receives an error signal at each
control point and performs adaptive signal processing to determine the gain of the audio signal
by performing adaptive signal processing so that the sum of powers of the error signals is
08-05-2019
5
minimized, and audio before or after the filter And a gain setting unit for multiplying the signal
by the gain and inputting the result to the speaker.
[0016]
According to the present invention, the audio signal output from the audio source is imparted
with a phase characteristic opposite to the phase characteristic from the speaker to the main
control point, and the inverse phase characteristic is imparted to the audio signal according to
the input audio signal. Is output from the speaker to the acoustic space, and an error signal at
each control point of the sound output from the speaker is input, and adaptive signal processing
is performed to minimize the sum of the powers of the error signals, thereby obtaining the gain
of the audio signal. The audio signal is given an opposite phase characteristic and the gain and
input to the speaker, so that one speaker and one adaptive signal processing device are used to
generate audio signals at a plurality of control points. Even if correction is performed, it is
possible to remove auditory unnaturalness at one main control point, and at other control points
other than the main control point. It becomes possible to improve characteristics.
[0017]
It is a block diagram of the multipoint adaptive equalization control system of this invention.
FIG. 6 is the gain characteristics of the present invention and the prior art at the main control
points.
It is the phase characteristics of the present invention and the prior art at the main control
points.
It is a gain characteristic of the present invention and prior art in control points other than a
main control point.
It is the phase characteristics of the present invention and the prior art at control points other
than the main control point.
FIG. 1 is a block diagram of a first prior art adaptive equalization control system applied to an
08-05-2019
6
audio device. It is a block diagram of the conventional multipoint adaptive equalization control
system.
[0018]
(A) Embodiment The present invention corrects the frequency phase characteristic representing
the arrival time of the sound wave for each frequency only at one of the main control points, and
corrects the gain characteristic only at all the plural control points. suggest. As a result, while
correcting the gain characteristics at all control points, the correction of the phase characteristics
only at the main control points improves the control performance at the main control points
while correcting the gain characteristics at other control points. It can be performed.
[0019]
FIG. 1 is a block diagram of the multipoint adaptive equalization control system according to the
present invention, in which adaptive signal processing is performed in the frequency domain, but
adaptive signal processing may be performed in the time domain. Also, the sound of the main
control point and the other first to Kth control points are controlled, suffix M in the figure
corresponds to the main control point, and suffixes S1 to SK correspond to the first to Kth control
points. doing.
[0020]
The FFT unit 1 converts the audio signal x (t) into an audio signal X (f) in the frequency domain,
and inputs it to the target response setting unit 2 and the adaptive signal processing device 8
respectively. The target response setting unit 2 is set with a target response characteristic
(impulse response) H and receives the audio signal X (f), and target response signals dM (f) and
dS1 (f) at corresponding control points. ,..., DSK (f) are output, and the target response signals are
input to the absolute value calculation units 3M, 3S1,.
[0021]
The absolute value calculators 3M, 3S1,..., 3SK are the absolute values DM (f), DS1 (f) of the input
08-05-2019
7
target response signals dM (f), dS1 (f),. ,..., DSK (f) are calculated, and absolute values DM (f), DS1
(f),..., DSK (f) of the target response signal are calculated in calculation units 4M, 4S1,. Enter each
one. The microphones 5M, 5S1,... 5SK are installed at listening positions (main control points,
first to Kth control points) in the vehicle interior acoustic space, and sound signals detected by
detecting the sound at these control points are detected Are input to the FFT units 6M, 6S1, ...,
6SK, respectively.
[0022]
The FFT units 6M, 6S1, ..., 6SK convert the input audio signal into audio signals yM (f), yS1 (f), ...,
ySK (f) in the frequency domain, and an absolute value calculator The voice signals yM (f), yS1
(f),..., YSK (f) are input to 7M, 7S1,. The absolute value calculation units 7M, 7S1,..., 7SK are input
absolute values YM (f), YS1 (f) of the input audio signals yM (f), yS1 (f),. , YSK (f) are calculated,
and the absolute values YM (f), YS1 (f),..., YSK (f) are input to the operation units 4M, 4S1,.
[0023]
The arithmetic units 4M, 4S1, ..., 4SK are the absolute values DM (f), DS1 (f), ..., DSK (f) of the
input target response signal and the absolute value YM (f) of the observation sound signal , YS1
(f),..., YSK (f), an error signal EM (f), ES1 (f),. f)-YM (f) ES1 (f) = DS1 (f)-YS1 (f) · · · · ESK (f) = DSK
(f)-YSK (f) to calculate.
[0024]
The adaptive signal processing unit 8 applies adaptive signal processing to the audio signal so
that the sum of the powers of the error signals EM (f), ES1 (f),..., ESK (f) at each control point is
minimized. The gain to be determined is determined, and the audio signal X ′ (f) multiplied by
the gain is generated and input to the IFFT unit 9.
The IFFT unit 9 converts the input signal X '(f) into an audio signal X' (t) in the time domain and
inputs it to the speaker 10, which responds to the input audio signal X '(t). Emits a loud sound
into the cabin interior acoustic space.
[0025]
08-05-2019
8
The target response characteristic H of the target response setting unit 2 is set to a characteristic
corresponding to the sound field space to be reproduced. For example, when the delay time from
the speaker to the main control point is t, a flat characteristic (characteristic of gain 1) is set in all
audio frequency bands, having a delay time t / 2. The target response setting unit 2 having such
a target response characteristic sets the coefficient of the tap corresponding to the delay time t /
2 of the digital filter of FIR type to 1, and sets the coefficient of the other taps to 0. Can be
realized by The target response characteristics may be common to each control point or may be
different.
[0026]
The adaptive signal processing device 8 receives the audio signal X (f) and also outputs the error
signals EM (f), ES1 (f),... Output from the above-described arithmetic units 4M, 4S1,. , And ESK (f)
is subjected to adaptive signal processing to minimize the sum of the powers, and an audio signal
X ′ (f) is output.
[0027]
The adaptive signal processing device 8 includes an LMS unit 8a that performs calculation
according to the LMS adaptive signal algorithm, a gain setting unit 8b that sets a gain, and a
microphone (control point) 5M, 5S1,. Signal processing filters 8cM, 8cS1,..., Which generate
reference signals by convolving transfer characteristics C ^ M, C ^ S1,..., C ^ SK of respective
sound propagation systems up to 5SK. It has 8cSK, absolute value calculation units 8dM, 8dS1, ...,
8dSK for calculating the absolute value of the input reference signal, and an anti-phase
characteristic setting unit 8e that constitutes an adaptive filter.
The adaptive signal processor 8 controls only the gain so as to minimize the sum of the powers
of the respective error signals, and does not control the phase.
[0028]
The antiphase characteristic setting unit 8e sets the antiphase IPCM (f) of the phase
characteristic from the speaker 10 to the main control point, controls the phase of the audio
signal here, and inputs it to the gain setting unit 8b. That is, the antiphase characteristics are set
by the antiphase setting unit 8e such that the phase delay of the audio signal at the main control
08-05-2019
9
point is zero. An adaptive filter is formed by the antiphase characteristic setting unit 8e and the
gain setting unit 8b. Here, the antiphase IPCM (f) is determined using the following equation:
IPCM (f) = CM (f) * / | CM (f) | (· * is a complex conjugate, | · | is an absolute value). However, CM
(f) is a transfer characteristic from the speaker to the main control point.
[0029]
The LMS unit 8a receives the error signals EM (f), ES1 (f),..., ESK (f) at each control point and the
reference signals output from the absolute value calculation units 8dM, 8S1,. The following
equation is input: G (n + 1, f) = G (n, f) + μM (f) · CM (f) CM · EM (f) + μS1 (f) · CS1 (f) | · ES1 ( f) ·
· · μSK (f) · | CSK (f) | · ESK (f) to determine the gain G (n + 1, f) of the gain setting unit 8b. Where
G (n, f) is the gain before one sampling, μM (f), μS1 (f),..., ΜSK (f) are parameters for adjusting
the amount of filter coefficient updating, EM (f), ES1 ( f), ..., ESK (f) is an error from the target
response at each control point. また、fはFFTのサイズをNとすれば、f1、f2、・・・、
fN/2である。 The gain setting unit 8 b multiplies the audio signal output from the reverse
phase setting unit 8 e by the gain G (n + 1, f) determined by the LMS unit 8 a and inputs the
result to the IFFT unit 9.
[0030]
From the above, assuming that the correction filter coefficient of adaptive signal processing is W
(f), W (f) is given by the following equation W (f) = G (f) * IPCM (f) However, all the initial
characteristics of G (f) are .., FN / 2 are given by 0 at frequencies f1, f2,. That is, the phase of the
correction filter coefficient W (f) is fixed by the antiphase IPCM (f), and only the gain G (f) is
adaptively controlled such that the sum of the powers of error signals at each control point is
minimized.
[0031]
2 to 5 show the measurement results of the gain characteristic and the phase characteristic of
the speech signal in the case of using the multipoint adaptive equalization control system of the
prior art and the speech signal in the case of using the multipoint adaptive equalization control
system of the present invention Fig. 2 shows gain characteristics at main control points, Fig. 3
shows phase characteristics at main control points, and Fig. 4 shows gain characteristics at
control points other than main control points. 5 is a phase characteristic at control points other
than the main control point, and each of (a) shows the result when the multipoint adaptive
equalization control system of the prior art (FIG. 7) is used, and (b) shows the multipoint of the
08-05-2019
10
present invention. It is a result at the time of using an adaptive equalization system.
[0032]
Referring to FIG. 2, in the prior art of (a), the gain fluctuates at all frequencies, but in the present
invention of (b), substantially flat gains are shown at all frequencies.
Referring to FIG. 3, while the phase characteristics in the prior art of (a) are dispersed, in the
present invention of (b), 0 is shown at all frequencies, and the signals of each frequency are up to
the control point. Since the phase characteristic representing the arrival time shows a constant
value (= 0), no signal delay occurs at the desired control point.
[0033]
Referring to FIGS. 4 and 5, the multipoint adaptive equalization system of the present invention
exhibits flatter gain over the prior art over all frequencies (FIG. 4). The present invention does
not perform phase correction at all frequencies, but exhibits phase characteristics equivalent to
those of the prior art (FIG. 5). Although the antiphase characteristic setting unit 8e is provided at
the front stage of the gain setting unit 8b in the above description, the antiphase characteristic
setting unit 8e may be provided at the rear stage of the gain setting unit 8b.
[0034]
As described above, in the prior art, since the average correction is performed at a plurality of
control points, the characteristics of all the control points can be corrected only moderately, and
the auditory sense of unnaturalness remains. Among the control points, correction is performed
on the gain characteristics and phase characteristics of all frequencies of the audio signal at a
desired control point, and only gain characteristics are corrected at the other control points. A
phase characteristic of value (= 0) can be obtained, and auditory unnaturalness can be removed.
Also, at control points other than the main control point, it is possible to obtain the same
characteristic as the conventional one.
[0035]
08-05-2019
11
Further, according to the present invention, the system configuration can be simplified because it
is only necessary to use one speaker and one adaptive signal processing device. Further,
according to the present invention, only the gain is determined by the adaptive signal algorithm,
so adaptive signal processing can be performed at high speed.
[0036]
1 FFT unit 3M, 3S1, ..., 3SK absolute value calculation unit 4M, 4S1, ..., 4SK operation unit 5M,
5S1, ..., 5SK microphone 6M, 6S1, ..., 6SK FFT unit 7M, 7S1 , ..., 7SK absolute value calculation
unit 8 adaptive signal processing device 8a LMS unit 8b gain setting unit 8cM, 8cS1, ..., 8cSK
signal processing filter 8dM, 8dS1, ..., 8dSK absolute value calculation unit 8e antiphase
Characteristic setting unit 9 IFFT unit 10 Speaker
08-05-2019
12
Документ
Категория
Без категории
Просмотров
0
Размер файла
23 Кб
Теги
jp2010266599
1/--страниц
Пожаловаться на содержимое документа