close

Вход

Забыли?

вход по аккаунту

?

JPH0495499

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 JPH0495499
[0001]
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an
apparatus for removing noise using an adaptive filter, and in particular, to the noise other than
the intended voice from the output of an acoustic-electric converter used in a room made of
metal. It is used to exclude. (Prior Art) In recent years, car phones have begun to spread. Since
the mobile phone is used during driving operations requiring mental concentration, it is
necessary to prevent obstacles to the driving operations. Conventional car telephones using
handsets are prohibited from using while traveling because they are an obstacle to driving as
well as general telephones. Therefore, there is a hold button for when a phone call is made while
driving, and it is in a state that "I can not release the hand against the other party while holding".
Please wait. The message "" will be streamed automatically. However, since the use value is low
for a car phone that can not be used while traveling, a hands-free phone capable of conversation
without a single-handed driving has been developed. This hands-free telephone is a telephone
that uses an echo canceller or the like to prevent howling and makes interactive conversations,
and since it is possible to talk without being aware of the telephone, it is suitable for use in
mobile telephones. There are various products being developed. However, the hands-free phone
convenient for the driver also has the following problems on the other side. Unlike conventional
phones, hands-free phones do not use a handset but use a microphone that is far away. For this
reason, the noise in the room (car while driving) mixed in with the microphone has a large
energy as compared with the telephone set using the handset. In other words, in the hands-free
telephone, the speaking voice is buried in noise and can not be heard or heard by the other party,
which causes a problem of giving the other party a sense of discomfort. From the above, in order
to realize hands-free telephones that can withstand practical use, it is essential to realize a noise
eliminator. FIG. 2 is an operation conceptual diagram of the conventional noise removal
apparatus. In FIG. 2, 10 is a noise removing device, 11 is an adaptive filter, 12 is a subtractor, 13
08-05-2019
1
is an output terminal, 14.15 is a microphone, 16 is a target signal source, and I7 is a noise signal
source. Further, S is a target signal output from the target signal source 16 and is a speaking
voice in the case of hands-free telephone. n is a noise signal output from the noise signal source
17, and in the case of a hands-free telephone, it is a car engine sound or the like. H1 is a function
representing an acoustic coupling condition between the microphone 14 and the target signal
source 16, H2 is a function representing an acoustic coupling condition between the microphone
15 and the noise signal source 17, C is a microphone 14 and noise signal source It is a function
expressing an acoustic coupling situation with 17, which is a time-varying function that changes
with time.
X is the output of the microphone 14, y is the output of the microphone 15, y 'is the output of
the adaptive filter 11, and e is the output of the subtractor 12. The microphone 14 receives the
target signal S and the noise signal n. Expressed by the equation, x = H, 'xs + cXn The microphone
15 has a noise signal n manually input. Expressed by the equation, 7 = Hz × n where the
adaptive filter 1 is adaptively controlled by the difference signal e to have a function of c / Hz y
′ = (C / H! Xy> cXn, and by calculating x−y ′ in the subtractor 12, a signal equal to H and Xs,
that is, an unmixed signal of a noise signal can be obtained. As is apparent from FIG. 2, in the
noise eliminator using an adaptive filter, a signal composed of only a noise signal as a reference
signal (input of the adaptive filter) (second example) In the figure y) is essential. However, since it
is extremely difficult to obtain "a signal composed only of noise signals" in actual use conditions,
there is a problem that noise can not be removed sufficiently. On the other hand, in the r noise
cancellation method disclosed in Japanese Patent Application Laid-Open No. 61-150,497, two
microphones are disposed equidistant from the target signal source, and the signal strength of
the target signal in each microphone and The signal phase is made equal, and the target signal is
canceled by the subtractor. On the other hand, with regard to noise signals, since the noise signal
sources are generally not equidistant to the two microphones, the signal strengths and signal
phases of the noise signals in the respective microphones are not equal. Hence, the noise signal
at each microphone remains uncancelled by the subtractor. In this way, a signal composed only
of noise signals is obtained. However, the above-mentioned "noise cancellation method J has the
following two problems. First, it is difficult to place the two microphones equidistant from the
target signal source in situations where an actual noise canceler is used. For this reason, a
reference signal consisting only of a noise signal can not be obtained, and the target signal and
the noise signal are mixed in the reference signal, so that complete noise removal can not be
performed. For example, when 3 Å or more in a car speaks simultaneously in a hands-free car
phone, it is impossible to install two microphones equidistant from the three target signal
sources. The same is true when the target signal source moves. Second, when the noise signal
source is equidistant to the two microphones, the signal strength and signal phase of the noise
signal at each microphone are equal and the noise is not removed.
08-05-2019
2
The present invention has the problem that the above-described conventional noise removing
device has a problem that noise removal can not be performed because the reference signal can
not be easily obtained in actual application, but it can be an extremely complicated and expensive
device if possible. To provide a simple and versatile noise removal device. (Means for Solving the
Problems) In order to achieve the above object, the present invention provides an acousticelectric converter for receiving a target signal and a noise signal, a vibration-electric converter
for receiving a noise signal, the vibration A correction circuit for correcting an output signal of
the electric converter so as to be identical to an output signal when the vibration-electric
converter has the same conversion characteristic as the acoustic-electric converter; and an
output signal of the correction circuit And a subtractor for calculating and outputting the
difference between the acousto-electrical converter and the output signal of the adaptive filter,
and oscillating the operation function of the adaptive filter based on the output of the subtractor
Adaptive control to approach the acoustic coupling between the electrical converter and its
mounting surface. (Operation) The acoustic-electric converter converts a target signal (voice) in
the room and noise coming out of the room through the wall of the room into an electric signal
and sends it to the subtractor. On the other hand, the vibration-electric converter converts the
noise transmitted through the wall of the room into an electric signal and sends it as a reference
signal to the adaptive filter through the correction circuit. As a result, a reference signal
consisting only of noise components not including the target signal is obtained. The output of the
adaptive filter is input to the subtractor, and the difference with the electrical signal from the
acousto-electrical converter is calculated. Here, if the operation function of the adaptive filter is
controlled based on the output of the subtractor and noise components included in both electric
signals input to the subtractor are not equalized, the purpose of not including the noise
component from the subtractor You can get a signal. A conventional adaptive algorithm can be
used for the adaptive film by providing a correction circuit for correcting the difference between
the characteristics of the vibration-electric converter and the characteristics of the acousticelectric converter. (Embodiment) FIG. 1 is a block diagram showing an embodiment of the
present invention. In FIG. 1, 1 is an acoustic-electric converter, 2 is a vibration-electric converter,
3 is a correction circuit for absorbing the difference between the conversion characteristics of
the acoustic-electric converter 1 and the vibration-electric converter 2, 4 Is an adaptive filter, 5 is
a subtractor, 6 is a noise eliminator output, 7 is a target signal source from which a target signal
S is output, 8 is a structure constituting a room (which corresponds to the body in a car), 9 is a
noise It is a noise signal source outside the room where the signal n is output. Further, the input
of the acoustic-electric converter 10 is acoustically connected to the target signal source 7 and
the structure 8, and the output is connected to one input of the subtractor 5.
The vibration / electricity converter 2 is closely attached to the structure 8 of the room, the input
is acoustically connected to the structure 8 of the room, and the output is connected to the input
of the correction circuit 3. The output of the correction circuit 3 is connected to the signal input
of the adaptive filter 4, the signal output of the adaptive filter 4 is connected to the other input of
08-05-2019
3
the subtractor 5, and the output of the subtracter 5 is the control signal input and noise rejection
of the adaptive filter 4 Connected to device output 6. When the conversion characteristics of the
acoustic-electric converter 1 and the vibration-electric converter 2 are the same, the correction
port w! 3 becomes unnecessary. Next, the operation of the present embodiment will be described
based on FIG. Now, the acoustic coupling function between the target signal source 7 and the
acoustic / electric converter 1 is HI, the acoustic coupling function between the noise signal
source 9 and the structure 8 in the room is Hz, the structure 8 in the room and the sound・ H2
represents the acoustic function of the structure 8 in the C1 room having an acoustic coupling
function with the electric converter 1, H represents the conversion characteristic of the acoustic /
electric converter 1, and B represents the conversion characteristic of the vibration / electric
converter 2. Do. Here, the above combined functions H +, Hz, Hs and c are each time-variant
functions. As shown in FIG. 1, in the case of the acoustoelectric converter 1, the target signal S
and the noise signal n are manually input via the structure 8 of the room, and the signal X shown
in the equation (1) is output. x = MX (H + Xs + c + HtXH, Xn) (1) The noise signal n is input to the
vibration / electricity converter 2 via the structure 8 of the room, and the signal y shown in
equation (2) is output. y = B x (Ht x Hs x n) (2) The signal y is input to the correction circuit 3 of
the operation function M / B, and a signal y 'shown in equation (3) appears at the output. y '= (M
/ B) x (Bx (H2XH, Xn)) = MX (Hz XH3xn) (3) The signal y' is input to the adaptive filter 4.
Assuming that the operation function of the adaptive filter 4 is C ′, the adaptive filter 4 outputs
a signal y ′ ′ shown in the equation (4). y '= c'xMX (Hz XH: lxn) (4) The subtractor 5 calculates
the difference e between the signal X and y #. e = x-y = M x (H + x s + c x H 2 x H 3 x n) ('x M x
(Hz x H: l x n) = M x (H, X s + (c-c') X (HE X H 3 X n) (5) Here, since the adaptive filter 4 operates
so as to adaptively bring the operation function C ′ closer to C, the relationship of c = c is
established. Therefore, equation (5) becomes equation (6).
e = MXH1Xs (6) That is, the output n of the subtractor 5 is the signal e shown in equation (6),
which is equal to the case where there is no noise signal n and only the target signal S is present.
In other words, noise has been removed. The above-mentioned noise removal operation does not
depend on the position of the target signal source 7 as is apparent from FIG. 1, so that the
adaptive operation of the adaptive filter 4 is also realized with a slower operation compared to
the adaptive operation of the conventional adaptive filter. can do. It should be noted that a
plurality of acoustic to electrical converters l for collecting the target signals shown in FIG. 1 are
provided separately, for example, in the vicinity of the driver's seat and the rear seat for the
driver and passenger in the car. It is also good. Also, with regard to the vibration / electricity
converter 2 that collects noise signals, for example, a plurality of vibration / electric converters
may be provided in a distributed manner in order to reliably collect noise. (Effects of the
Invention) As described above in detail, by using the noise eliminator according to the present
invention, the conventional noise eliminator only uses the signal of the noise signal component
which can not be practically obtained. It is possible to obtain perfect noise removal as In addition,
a characteristic correction circuit is provided between the vibration / electric converter and the
08-05-2019
4
adaptive filter to correct the difference between the characteristics of the vibration / electric
converter and the characteristics of the acoustic / electric converter. Since it is possible to use an
adaptation algorithm of, it is not necessary to newly obtain an adaptation algorithm. The
invention is particularly useful for noise removal with sources outside the room, especially inside
a room made of metallic structure, such as a hands-free phone used in a car room.
[0002]
Brief description of the drawings
[0003]
FIG. 1 is a block diagram showing an embodiment of the present invention, and FIG. 2 is an
operation conceptual diagram of a conventional noise removal apparatus.
DESCRIPTION OF SYMBOLS 1: .. acoustic-electric converter 2 ... vibration-electric converter 3 ...
correction circuit 4 ... adaptive filter 5 ... subtractor, 6 ... output 7 ... Target signal source, 8:
structure, 9: noise signal source.
08-05-2019
5
Документ
Категория
Без категории
Просмотров
0
Размер файла
15 Кб
Теги
jph0495499
1/--страниц
Пожаловаться на содержимое документа