Electrical audio signal processing systems and devices – Acoustical noise or sound cancellation – Within duct
Reexamination Certificate
1997-09-17
2001-02-20
Harvey, Minsun Oh (Department: 2747)
Electrical audio signal processing systems and devices
Acoustical noise or sound cancellation
Within duct
C381S071100
Reexamination Certificate
active
06192133
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to an active noise control apparatus, which can reduce noise emitted from devices placed in a three-dimensional space by automatic control, with a simple structure.
As is well-known, in active noise control apparatus, in order to reduce noise emitted from a noise source, an additional sound source is newly provided. Then, an additional sound emitted from the additional sound source and noise are added to each other. By an interference effect of sound generated by the additional sound source and noise, noise is canceled. In such active noise control apparatus, there is often used an additional sound source controlling method. In the method, a synthesized sound of noise and the additional sound is measured by a microphone, and sound pressure is controlled to be zero or minimum.
In the conventional active noise control apparatus, which is used in a system for one-dimensionally propagating sound such as an air-conditioning duct, in order that noise emitted from a noise source can be prevented from being leaked from an opening portion of the air-conditioning duct to the outside, the following arrangement is provided.
More specifically, a reference microphone for obtaining a noise signal propagated through the air-conditioning duct is provided on the noise source side. An estimation microphone for obtaining a signal of sound leaking to the outside from the opening portion is provided on the opening portion side. Moreover, a speaker, serving as an additional sound source, for emitting additional sound to the air-conditioning duct is provided between both microphones.
The signal of sound obtained by the estimation microphone is a signal of a synthesized sound of the sound from the noise source and the sound from the speaker. Sound emitted from the speaker is controlled by an adaptive control device such that the synthesized sound is made close to zero.
The adaptive control device constitutes a control system based on a filter-MLMS algorithm, which is frequently used in an active noise control. The adaptive control device comprises a transfer compensation filter, which is from the speaker to the microphone, an FIR (Finite Impulse Response) filter, and an adaptive FIR filter. More specifically, an output signal of the reference microphone is input to the FIR filter, and input to the adaptive FIR filter through the transfer function compensation filter. An output signal of the estimation microphone is input to the adaptive FIR filter.
The FIR filter obtains a sum of products of the input signal and the filter coefficient, and outputs it. When a sum of the output of the filter and the outer signal is set to be an error signal, the adaptive FIR filter automatically renews the filter coefficient such that the error signal is made close to zero. In this case, the signal from the estimation microphone becomes the error signal. In the Filter-XLMS algorithm, the filter coefficient of the FIR filter is set to be the same as that of the active FIR filter.
Generally, in the adaptive FIR filter, every time when the sampling input is provided, the following calculations are performed:
e
(
n
)=
d
(
n
)−
y
(
n
) (1)
h
(
k
)
new
=h
(
k
)
old
+&mgr;e
(
n
)×(
n−k+
1)(
k=
1, 2, . . . ,
M
) (2)
In this case,
y
(
n
)
=
∑
k
=
1
M
h
(
k
)
x
(
n
-
k
+
1
)
(
3
)
In equations (1) to (3), e(n) is an error signal at time n, d(n), y(n) are an outer signal and an output signal of the filter, respectively, and h(k) is a filter coefficient. Also, a mark, new, means a state after the filter coefficient is renewed, a mark, old, means a state before the filter coefficient is renewed. Moreover, x (n−k+1) is an input signal, and &mgr; is a constant, which is called a step size parameter, which determines a velocity of convergence. Equation (2) showing the renewal of the filter coefficient can be obtained from the following equation (4) of a steepest descent method:
h
(
k
)
new
=
h
(
k
)
old
-
1
2
μ
∂
ϵ
∂
h
(
k
)
old
(
k
=
1
,
2
,
…
,
M
)
(
4
)
In this case,
&egr;=
E[e
2
(
n
)] (&egr; is a square average value of e(N)) (5)
∂
ϵ
∂
h
(
k
)
old
=
-
2
e
(
n
)
x
(
n
-
k
+
1
)
(
6
)
In the above-mentioned active noise control apparatus, the additional sound for control is emitted from the speaker based on the sound detected by the reference microphone. The additional sound and noise propagated through the air-conditioning duct from the noise source is interfered with each other. Then, sound pressure at the position of the estimation microphone is made close to zero while the degree of interference is being estimated. When a zero point of sound pressure is created, sound is reflected at the point due to a difference in acoustic impedance. As a result, sound is not propagated in the direction of the opening portion, so that noise emitted from the opening portion can be eliminated.
However, such an active noise control apparatus is useful in the system in which noise such as air-conditioning duct is one-dimensionally propagated, but cannot be applied to noise, which is three-dimensionally propagated. The reason can be explained as follows:
In the case of the one-dimensional propagation, only by setting sound pressure at the opening portion, that is, an outlet, to zero, all sound emitted from the noise source can be reflected to prevent sound from being leaked from the outlet. In the case of the three-dimensional propagation, noise is emitted from the noise source all directions. As a result, even if sound pressure is set to zero at one control point, no effect is brought about.
On the other hand, in consideration of the point that noise emitted from devices as the noise sources is three-dimensionally widened, there is proposed the following method for reducing the entire noise from the devices. More specifically, the devices are surrounded by many additional sound sources. Then, each additional sound is emitted from each additional sound source to reduce sound, which is leaked to the outer portion of the surrounded surface. This method is equivalent to the point that the method of one-dimensional propagation of sound in the air-conditioning duct (already explained) is expanded to a three-dimensional space.
In an active noise control apparatus, which can realize the above method, a plurality of additional sound sources is arranged on the surrounded surface with an equal distance. Then, an estimation microphone is provided close to each additional sound source. An output signal of each estimation microphone is input to a controller for drive-controlling each additional sound source. A signal, which is correlated with sound emitted from the devices, such as a vibration signal of a wall surface, is input as a reference signal to the controller through a signal line.
The controller controls each additional sound source such that sound pressure becomes minimum at each estimation microphone. In other words, the controller controls each corresponding additional sound source and the phase such that a sum of squares of sound pressure becomes minimum at each estimation microphone.
However, in the above-structured active noise control apparatus, additional sound emitted from the additional sound sources is mixed with the respective microphones. The control system must be structured to consider influence caused by the mixture of additional sound. As a result, the control system including the controller becomes inevitably complicated. Moreover, the distance between the additional sound sources must be set to be within ½ of the wavelength of sound to be reduced. As a result, a large number of additional sound sources and estimation microphones must be provided when the surrounded surface is set to a position, which is relatively separated from each unit as a noise source.
Moreover, as another method, the followi
Enamito Akihiko
Hayashi Takuro
Harvey Minsun Oh
Kabushiki Kaisha Toshiba
Oblon & Spivak, McClelland, Maier & Neustadt P.C.
LandOfFree
Active noise control apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Active noise control apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Active noise control apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2572521