U.S. patent application number 16/085716 was filed with the patent office on 2019-03-28 for noise reduction device and noise reduction method.
This patent application is currently assigned to DENSO TEN Limited. The applicant listed for this patent is DENSO TEN Limited. Invention is credited to Masaaki TAIRA.
Application Number | 20190097663 16/085716 |
Document ID | / |
Family ID | 60478291 |
Filed Date | 2019-03-28 |
![](/patent/app/20190097663/US20190097663A1-20190328-D00000.png)
![](/patent/app/20190097663/US20190097663A1-20190328-D00001.png)
![](/patent/app/20190097663/US20190097663A1-20190328-D00002.png)
![](/patent/app/20190097663/US20190097663A1-20190328-D00003.png)
![](/patent/app/20190097663/US20190097663A1-20190328-D00004.png)
![](/patent/app/20190097663/US20190097663A1-20190328-D00005.png)
![](/patent/app/20190097663/US20190097663A1-20190328-D00006.png)
![](/patent/app/20190097663/US20190097663A1-20190328-D00007.png)
![](/patent/app/20190097663/US20190097663A1-20190328-D00008.png)
United States Patent
Application |
20190097663 |
Kind Code |
A1 |
TAIRA; Masaaki |
March 28, 2019 |
NOISE REDUCTION DEVICE AND NOISE REDUCTION METHOD
Abstract
A noise reduction device according to an aspect of an embodiment
includes a detection unit, a calculation unit, and a reduction
unit. The detection unit detects frequency components of a
plurality of noise signals that are included in a received signal,
based on a frequency spectrum of the received signal. The
calculation unit calculates a coefficient, based on the frequency
components that are detected by the detection unit. The reduction
unit reduces n noise signals that are at least a part of the
plurality of noise signals from the received signal, depending on
the coefficient that is calculated by the calculation unit (where n
is a positive number that is greater than or equal to 2).
Inventors: |
TAIRA; Masaaki; (Kobe-shi,
JP) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
DENSO TEN Limited |
Kobe-shi, Hyogo |
|
JP |
|
|
Assignee: |
DENSO TEN Limited
Kobe-shi, Hyogo
JP
|
Family ID: |
60478291 |
Appl. No.: |
16/085716 |
Filed: |
March 8, 2017 |
PCT Filed: |
March 8, 2017 |
PCT NO: |
PCT/JP2017/009339 |
371 Date: |
September 17, 2018 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04B 1/1027 20130101;
H04B 1/10 20130101; H04B 1/109 20130101; H04B 1/1036 20130101 |
International
Class: |
H04B 1/10 20060101
H04B001/10 |
Foreign Application Data
Date |
Code |
Application Number |
May 30, 2016 |
JP |
2016-107458 |
Claims
1. (canceled)
2. (canceled)
3. A noise reduction device, comprising: a detection unit that
detects frequency components of a plurality of noise signals that
are included in a received signal, based on a frequency spectrum of
the received signal; a calculation unit that calculates a
coefficient, based on the frequency components that are detected by
the detection unit; and a reduction unit that reduces n noise
signals that are at least a part of the plurality of noise signals
from the received signal, depending on the coefficient that is
calculated by the calculation unit, where n is a positive number
that is greater than or equal to 2, wherein the detection unit
detects signal levels of the n noise signals; the calculation unit
includes a coefficient calculation unit that calculates the
coefficient in a frequency domain, based on the frequency
components and the signal levels that are detected by the detection
unit, and a transformation unit that transforms the coefficient in
a frequency domain that is calculated by the coefficient
calculation unit into the coefficient in a time domain; and the
reduction unit reduces the n noise signals from the received
signal, in a time domain, depending on the coefficient in a time
domain that the transformation unit transforms the coefficient in a
frequency domain into.
4. (canceled)
5. The noise reduction device according to claim 3, wherein the
detection unit includes: a difference calculation unit that
calculates a difference between upper and lower sidebands of a
frequency spectrum of the received signal; and a noise detection
unit that detects the frequency components and the signal levels,
based on the difference that is calculated by the difference
calculation unit.
6. The noise reduction device according to claim 5, wherein: the
noise detection unit provides a signal level difference between the
upper and lower sidebands in the frequency components of the noise
signals as the signal levels that are reduced by the reduction
unit.
7. (canceled)
8. A noise reduction method, comprising: detecting frequency
components of a plurality of noise signals that are included in a
received signal, based on a frequency spectrum of the received
signal; calculating a coefficient, based on the detected frequency
components; and reducing n noise signals that are at least a part
of the plurality of noise signals from the received signal,
depending on the calculated coefficient, where n is a positive
number that is greater than or equal to 2, wherein the detecting
includes detecting signal levels of the n noise signals; the
calculating includes calculating the coefficient in a frequency
domain, based on the detected frequency components and the detected
signal levels, and transforming the calculated coefficient in a
frequency domain into the coefficient in a time domain; and the
reducing reduces the n noise signals from the received signal, in a
time domain, depending on the coefficient in a time domain that the
coefficient in a frequency domain is transformed into.
Description
FIELD
[0001] The present invention relates to a noise reduction device
and a noise reduction method.
BACKGROUND
[0002] Conventionally, in a case where a system that receives a
radio broadcast wave is mounted on a vehicle, such a system may
receive, for example, a noise signal that is caused by a switching
process that is executed at a time of power conversion of a battery
of the vehicle. Accordingly, a conventional system reduces an
output intensity for a peak frequency with a highest intensity
among frequency components that are included in a received signal,
so that a noise signal is reduced.
CITATION LIST
Patent Literature
[0003] Patent Literature 1: Japanese Patent Application Publication
No. 2013-009066
SUMMARY
Technical Problem
[0004] However, a conventional technique merely reduces a noise
signal at a peak frequency with a highest intensity. Therefore, for
example, in a case where a plurality of noise signals are included
in a received signal, there is a possibility that such noise
signals are not completely reduced and a noise signal remains in
the received signal.
[0005] The present invention is attained while the above is taken
into consideration, and aims to provide a noise reduction device
and a noise reduction method that are capable of reducing a
plurality of noise signals that are included in a received
signal.
Solution to Problem
[0006] In order to solve a problem as described above and achieve
an object, a noise reduction device according to the present
invention includes a detection unit, a calculation unit, and a
reduction unit. The detection unit detects frequency components of
a plurality of noise signals that are included in a received
signal, based on a frequency spectrum of the received signal. The
calculation unit calculates a coefficient, based on the frequency
components that are detected by the detection unit. The reduction
unit reduces n noise signals that are at least a part of the
plurality of noise signals from the received signal, depending on
the coefficient that is calculated by the calculation unit (where n
is a positive number that is greater than or equal to 2).
BRIEF DESCRIPTION OF DRAWINGS
[0007] FIG. 1A is a block diagram illustrating a configuration of a
receiving device according to a first embodiment.
[0008] FIG. 1B is a diagram illustrating a received signal that is
received by a receiving device according to a first embodiment.
[0009] FIG. 1C is a diagram illustrating a received signal where a
nose signal is eliminated by a receiving device according to a
first embodiment.
[0010] FIG. 2 is a block diagram illustrating a configuration of a
receiving device according to a second embodiment.
[0011] FIG. 3 is a diagram illustrating a frequency distribution of
a noise signal according to a second embodiment.
[0012] FIG. 4 is a flowchart illustrating a noise reduction process
according to a second embodiment.
[0013] FIG. 5 is a block diagram illustrating a configuration of a
receiving device according to a third embodiment.
[0014] FIG. 6 is a flowchart illustrating a noise reduction process
according to a third embodiment.
[0015] FIG. 7 is a block diagram illustrating a configuration of a
receiving device according to a variation.
[0016] FIG. 8 is a flowchart illustrating a noise reduction process
according to a variation.
DESCRIPTION OF EMBODIMENTS
[0017] Hereinafter, embodiments of a noise reduction device and a
noise reduction method that are disclosed in the present
application will be explained in detail with reference to the
accompanying drawings. Additionally, this invention is not limited
by embodiments illustrated below.
First Embodiment
[0018] A receiving device 1 according to a first embodiment of the
present invention will be explained by using FIG. 1A to FIG. 1C.
FIG. 1A is a block diagram illustrating a configuration of the
receiving device 1 according to the present embodiment.
Furthermore, FIG. 1B is a diagram illustrating a received signal
that is received by the receiving device 1 according to the present
embodiment. FIG. 1C is a diagram illustrating a received signal
where a noise signal is eliminated by the receiving device 1
according to the present embodiment.
[0019] Such a receiving device 1 is mounted on, for example, a
vehicle such as a car. Furthermore, the receiving device 1
receives, for example, an amplitude-modulated radio broadcast wave
that is transmitted by using a medium wave band of 526.5 kHz to
1606.5 kHz (that will also be described as an AM radio broadcast
wave, below). An AM radio broadcast wave includes information on a
sound or voice (that will also be described as sound or voice data,
below).
[0020] Additionally, a case where such a receiving device 1 is
mounted on a vehicle will be explained in the present embodiment
and this is not limiting. The receiving device 1 may be mounted on,
for example, a movable body such as a train or an airplane, an
appliance such as a radio receiver or a personal computer, or the
like. Furthermore, the receiving device 1 does not have to receive
a radio broadcast wave as long as an amplitude-modulated receiving
signal is received.
[0021] <Receiving Device 1>
[0022] As illustrated in FIG. 1A, the receiving device 1 has a
noise reduction device 10, an antenna 20, and a Radio Frequency
(RF) unit 30. The antenna 20 receives, for example, an AM radio
broadcast wave as a received signal.
[0023] The RF unit 30 applies a signal process such as an
amplification process, an A/D conversion process, or
down-converting to a received signal that is received through the
antenna 20.
[0024] The noise reduction device 10 reduces a plurality of noise
signals from a received signal. The noise reduction device 10 has a
detection unit 110, a calculation unit 120, and a reduction unit
130.
[0025] The detection unit 110 detects frequency components of a
plurality of noise signals that are included in a received signal,
based on a frequency spectrum of the received signal. As
illustrated in FIG. 1B, a received signal includes a carrier wave
Sc with a carrier frequency fc and modulated waves Sm1 and Sm2 in
an upper sideband Wm1 and a lower sideband Wm2. Furthermore, a
received signal includes, for example, a plurality of noise signals
such as periodic noise signals Sn11 to Sn13 that generate from an
inverter, a noise signal Sn2 that generates from a DCDC converter,
or the like. Hereinafter, noise signals Sn11 to Sn13 and Sn2 will
collectively be described as a noise signal(s) Sn. A noise signal
Sn is a so-called peaky noise that has a peak at a particular
frequency.
[0026] The detection unit 110, for example, compares a frequency
spectrum of a received signal with a threshold and thereby detects
frequency components and signal levels of a plurality of noise
signals Sn. Additionally, FIG. 1B illustrates a case where four
noise signals Sn are provided.
[0027] The calculation unit 120 calculates a coefficient based on
frequency components and signal levels that are detected by the
detection unit 110. Such a coefficient is, for example, a filter
coefficient for a filtering process that is executed by the
reduction unit 130. The calculation unit 120 calculates a
coefficient in such a manner that a signal level of a received
signal at a frequency component that is detected by the detection
unit 110 is reduced depending on a signal level that is detected by
the detection unit 110. Additionally, a coefficient may be
calculated every time the detection unit 110 calculates a noise
signal Sn, where it is more preferable to execute it at timing when
there is a possibility that a frequency that provides noise is
changed. For such timing, there is provided, for example, a time of
vehicle start-up, a time when an operation condition or the like of
a converter (voltage conversion) is changed, timing for each
predetermined period of time or the like where a possibility of
time variation (a factor such as a temperature change) is taken
into consideration, or the like.
[0028] The reduction unit 130 reduces n noise signals Sn that are
at least a part of a plurality of noise signals Sn from a received
signal, depending on a coefficient that is calculated by the
calculation unit 120 (where n is a positive number that is greater
than or equal to 2). The reduction unit 130, for example, has a
filter and reduces a noise signal Sn from a received signal while a
coefficient that is calculated by the calculation unit 120 is
provided as a filter coefficient. As illustrated in FIG. 1C, a
received signal where a plurality of noise signals Sn are reduced
therefrom is provided as a signal that includes a carrier signal Sc
and modulated signals Sm1 and Sm2 where a plurality of noise
signals Sn11 to Sn13 and Sn2 are reduced. Additionally, although
the reduction unit 130 eliminates n=4, that is, all detected noise
signals Sn in FIG. 1C, this is not limiting. The reduction unit 130
may reduce at least a part of detected noise signals Sn, and
reduces, for example, two noise signals or the like.
[0029] Thus, the noise reduction device 10 of the receiving device
1 according to an embodiment detects frequency components of a
plurality of noise signals Sn. Furthermore, the noise reduction
device 10 reduces a plurality of noise signals Sn from a received
signal, depending on a coefficient that is based on a detected
frequency component. Thereby, it is possible for the noise
reduction device 10 to reduce a plurality of noise signals Sn from
a received signal.
Second Embodiment
[0030] A receiving device 1A according to a second embodiment will
be explained by using FIG. 2 to FIG. 4. FIG. 2 is a block diagram
illustrating a configuration of the receiving device 1A according
to the second embodiment. Additionally, a component that is
identical to that of the receiving device 1 as illustrated in FIG.
1 will be provided with an identical sign and an explanation
thereof will be omitted.
[0031] <Receiving Device 1A>
[0032] The receiving device 1A as illustrated in FIG. 2 includes a
noise reduction device 10A. The noise reduction device 10A has a
detection unit 110A, a calculation unit 120A, a reduction unit
130A, and a storage unit 140.
[0033] The detection unit 110A has an analysis unit 111, a
difference calculation unit 112, and a noise detection unit 113.
Furthermore, the reduction unit 130A has n or first to n-th filters
131 to 13n. Additionally, FIG. 2 illustrates a case of n=4.
[0034] The analysis unit 111 analyzes a frequency spectrum of a
received signal. The analysis unit 111 executes a Fast Fourier
transform (FFT) process for a received signal so that a spectral
intensity at each frequency component is calculated.
[0035] The difference calculation unit 112 calculates a difference
between an upper sideband Wm1 and a lower sideband Wm2 in a
frequency spectrum of a received signal. Herein, the receiving
device 1A receives an AM radio broadcast wave as a received signal.
As illustrated in FIG. 1B, in an AM radio broadcast, a signal that
is centered at a carrier frequency fc and provided with
predetermined bandwidths Wm1 and Wm2 is transmitted. Herein, a
predetermined bandwidth is 15 kHz (fc.+-.7.5 kHz).
[0036] Furthermore, a signal that is transmitted in an AM radio
broadcast is provided with respectively identical frequency
distributions in a bandwidth Wm1 on a higher frequency side (an
upper sideband) and a bandwidth Wm2 on a lower frequency side (a
lower sideband) with respect to a carrier frequency fc.
[0037] In the present embodiment, attention is paid to such a
point, a difference between spectral intensities in upper and lower
sidebands Wm1 and Wm2 of a received signal is calculated so that a
modulated component of the received signal is reduced to improve
detection accuracy of a noise component.
[0038] Specifically, the difference calculation unit 112 calculates
a difference between spectral intensities in upper and lower
sidebands Wm1 and Wm2 that are positioned at an equal distance from
a carrier frequency fc among spectral intensities in the upper and
lower sidebands Wm1 and Wm2 of a received signal. The difference
calculation unit 112 calculates, as a difference, for example, a
result provided by subtracting a lower sideband Wm2 from an upper
sideband Wm1 that is positioned at an equal distance from a carrier
frequency fc.
[0039] As described above, a signal that is transmitted in an AM
radio broadcast has a line-symmetric frequency distribution that is
centered at a carrier frequency fc and provided with upper and
lower sidebands Wm1 and Wm2. Accordingly, as, for example, a
difference between upper and lower sidebands Wm1 and Wm2 of a
received signal is calculated, a frequency distribution of a
difference is a distribution where spectral intensities of
modulated signals Sm1 and Sm2 that are transmitted in an AM radio
broadcast are approximately zero and a plurality of noise signals
Sn remain therein as illustrated in FIG. 3.
[0040] Additionally, FIG. 3 is a diagram illustrating a frequency
distribution of a noise signal Sn. Furthermore, a spectral
intensity at a carrier frequency fc is eliminated by, for example,
a notch filter or the like and is not used for correction of a
spectral intensity, detection of a noise, or the like.
[0041] The noise detection unit 113 compares an absolute value of a
difference that is calculated by the difference calculation unit
112 with a threshold Th1. Additionally, in a case where a
difference before calculating an absolute value is negative, a
frequency component in a lower sideband is provided, while in a
positive case, a frequency component in an upper sideband is
provided. The noise detection unit 113 provides a frequency
component with a spectral intensity that is greater than or equal
to a threshold Th1 as a frequency component of a noise signal Sn.
Furthermore, a spectral intensity L of a detected frequency
component, that is, a signal level difference L between upper and
lower sidebands for a detected frequency component is provided as a
signal level of a noise signal.
[0042] Thus, the difference calculation unit 112 calculates a
difference between upper and lower sidebands Wm1 and Wm2 of a
received signal and the noise detection unit 113 detects a
plurality of noise signals Sn based on such a difference, so that
it is possible to detect a noise with a higher degree of
accuracy.
[0043] The calculation unit 120A calculates a coefficient for each
frequency component based on a frequency component and a signal
level that are detected by the noise detection unit 113. Herein, a
coefficient is a filter coefficient of each of the first to n-th
filters 131 to 13n of the reduction unit 130A. The calculation unit
120A selects a coefficient, depending on, for example, a
coefficient database (DB) 141 that is stored in the storage unit
140, and thereby calculates such a coefficient.
[0044] The coefficient DB 141 stores a coefficient for each of a
frequency component and a signal level. The calculation unit 120A
selects a coefficient with reference to the coefficient DB 141
based on a frequency component and a signal level that are detected
by the noise detection unit 113. The calculation unit 120A selects
a coefficient with reference to the coefficient DB 141 for each of
a plurality of noise signals Sn that are detected by the noise
detection unit 113.
[0045] Herein, a case will be explained where the number M of noise
signals Sn that are detected by the noise detection unit 113 is
greater than the number n of noise signals that are capable of
being reduced by the noise reduction device 10A (M>n). In such a
case, it is possible for the noise reduction device 10A to reduce n
noise signals Sn, for example, in descending order of a signal
level. That is, the calculation unit 120A selects n coefficients
dependent on a frequency component and a signal level of a noise
signal Sn in descending order of a signal level.
[0046] Thereby, it is possible for the noise reduction device 10A
to reduce a noise signal Sn depending on a degree of a signal
level.
[0047] Alternatively, the noise reduction device 10A may select n
noise signals Sn that are reduced in an order of priority dependent
on an auditory characteristic. As described above, sound or voice
data are included in an AM radio broadcast wave. Accordingly, the
noise reduction device 10A reduces a noise signal Sn depending on
an auditory characteristic for a sound or voice, so that it is
possible to preferentially reduce a noise signal Sn of a frequency
component that is readily listened to by a user.
[0048] Specifically, for example, the calculation unit 120A selects
a coefficient for each of frequency components of n noise signals
Sn in order of priority dependent on a high-cut process that is
executed by a (non-illustrated) demodulation unit in a latter part
of the receiving device 1A.
[0049] Additionally, although the number M of noise signals S that
are detected by the detection unit 110A is greater than the number
of the first to n-th filters 131 to 13n herein, this is not
limiting. The number M of noise signals Sn may be less than the
number n of the first to n-th filters 131 to 13n. In such a case,
the calculation unit 120A selects, for example, coefficients of the
first to M-th filters 131 to 13M. Furthermore, the noise reduction
device 10A prevents remaining (M+1)-th to n-th filters 13(M+1) to
13n from operating. Alternatively, the calculation unit 120A may
select coefficients in such a manner that the (M+1)-th to n-th
filters 13(M+1) to 13n output signals that are identical to a
received signal that is an input signal.
[0050] The first to n-th filters 131 to 13n of the reduction unit
130A are, for example, Finite Impulse Response (FIR) notch filters
and reduce respective peaky noises from a received signal. The
first to n-th filters 131 to 13n are mutually connected in series.
The first to n-th filters 131 to 13n respectively filter n noise
signals Sn in order, depending on each coefficient that is selected
by the calculation unit 120A.
[0051] The storage unit 140 stores information needed for a process
that is executed by each unit of the noise reduction device 10A,
such as a coefficient DB 141 or a threshold Th1. Furthermore, the
storage unit 140 stores a result of a process that is executed by
each unit of the noise reduction device 10A. The storage unit 140
is, for example, a semiconductor memory element such as a Random
Access Memory (RAM) or a flash memory or a storage device such as a
hard disk or an optical disc.
[0052] Additionally, although the noise reduction device 10A
includes the storage unit 140 herein, this is not limiting. For
example, the receiving device 1A may include the storage unit
140.
[0053] <Noise Reduction Process>
[0054] A noise reduction process that is executed by the noise
reduction device 10A will be explained by using FIG. 4. FIG. 4 is a
flowchart illustrating a noise reduction process according to the
present embodiment. A noise reduction process executes a noise
reduction process, for example, in a case where the antenna 20
receives a received signal. A noise reduction process is repeatedly
executed with a predetermined period of time while the receiving
device 1A receives a received signal.
[0055] First, the noise reduction device 10A analyzes a frequency
of a received signal (step S101). The noise reduction device 10A
calculates a difference between upper and lower sidebands Wm1 and
Wm2 of a received signal based on a result of analysis (step S102).
The noise reduction device 10A compares an absolute value of a
calculated difference with a threshold Th1 (step S103). In a case
where an absolute value of a difference is less than a threshold
Th1 (step S103; No), the noise reduction device 10A ends a
process.
[0056] On the other hand, in a case where an absolute value of a
difference is greater than or equal to a threshold Th1 (step S103;
Yes), the noise reduction device 10A detects a plurality of
frequency components and signal levels of a difference with an
absolute value that is greater than a threshold Th1 (step S104).
The noise reduction device 10A selects a coefficient for each of a
plurality of detected frequency components (step S105). The noise
reduction device 10A reduces at least n noise signals Sn among a
plurality of noise signals Sn from a received signal for each of
frequency components of such noise signals Sn depending on a
selected coefficient (step S106) and ends a process.
[0057] Additionally, although the noise reduction device 10A ends a
process at step S103 in a case where an absolute value of a
difference is less than a threshold Th1, this is not limiting. In a
case where an absolute value of a difference is less than a
threshold Th1, the noise reduction device 10A determines that a
noise signal Sn is not included in a received signal. Accordingly,
the noise reduction device 10A may output a received signal that is
produced by the RF unit 30 to a (non-illustrated) latter processing
unit directly.
[0058] As stated above, the noise reduction device 10A according to
the second embodiment reduces a noise signal Sn by using n or the
first to n-th filters 131 to 13n, so that it is possible to reduce
a plurality of noise signals Sn from a received signal.
[0059] Additionally, in the above-mentioned embodiment, a
difference L between upper and lower sidebands Wm1 and Wm2 of a
received signal is provided as a signal level of a noise signal Sn.
Thereby, it is possible for the noise reduction device 10A to
reduce a noise signal Sn accurately as compared with, for example,
a case where a received signal at a frequency component of a noise
signal Sn is reduced by a predetermined level. Furthermore, as
compared with, for example, a signal level of a received signal at
a frequency component of a noise signal Sn is "0", it is possible
to maintain continuity of a frequency spectrum of a received signal
and it is possible to improve a receiving accuracy of a received
signal. Additionally, depending on a desired reduction accuracy of
a noise signal Sn or a reception accuracy of a received signal, a
signal level of a received signal that is reduced by the reduction
unit 130A may be reduced by a predetermined level or reduced in
such a manner that a signal level of a received signal is "0".
Third Embodiment
[0060] A receiving device 1B according to a third embodiment will
be explained by using FIG. 5 and FIG. 6. FIG. 5 is a block diagram
illustrating a configuration of the receiving device 1B according
to the third embodiment. Additionally, a component that is
identical to that of the receiving device 1A as illustrated in FIG.
2 will be provided with an identical sign and an explanation
thereof will be omitted.
[0061] <Receiving Device 1B>
[0062] The receiving device 1B as illustrated in FIG. 5 includes a
noise reduction device 10B. The noise reduction device 10B has a
calculation unit 120B, a reduction unit 130B, and a storage unit
140B. The calculation unit 120B has a coefficient calculation unit
121 and a transformation unit 122. The reduction unit 130B has a
filter 131B.
[0063] The coefficient calculation unit 121 calculates a
coefficient in a frequency domain based on frequency components and
signal levels of a plurality of noise signals Sn that are detected
by the noise detection unit 113. The coefficient calculation unit
121 calculates a coefficient that reduces signal levels of at least
n frequency components of a plurality of noise signals Sn that are
detected by the noise detection unit 113, by a signal level that is
detected by the noise detection unit 113 (where n is a positive
number that is greater than or equal to 2). Additionally, one
coefficient that is calculated herein is provided. Furthermore,
signal levels of all noise signals Sn that are detected by the
noise detection unit 113 are reduced herein.
[0064] For example, the coefficient calculation unit 121 calculates
a coefficient provided by adding respective coefficients in a
frequency domain of first to n-th filters 131 to 13n that are
calculated by the calculation unit 120A as illustrated in FIG. 2,
as a coefficient .alpha. of the reduction unit 130B.
[0065] The transformation unit 122 transforms a coefficient in a
frequency domain that is calculated by the coefficient calculation
unit 121 into a coefficient in a time domain. The transformation
unit 122 executes, for example, an Inverse Fast Fourier Transform
(IFFT) process.
[0066] For example, the transformation unit 122 applies an inverse
Fourier transformation to a coefficient .alpha. in a frequency
domain to calculate a coefficient .beta. in a time domain.
[0067] The reduction unit 130B has one filter 131B as described
above. The filter 131B is, for example, an FIR notch filter. The
filter 131B reduces at least n noise signals Sn among a plurality
of noise signals Sn from a received signal, depending on a
coefficient .beta. in a time domain that the transformation unit
122 transforms it into.
[0068] The storage unit 140B is identical to the storage unit 140
as illustrated in FIG. 2 except that it does not have the
coefficient DB 141.
[0069] <Noise Reduction Process>
[0070] A noise reduction process that is executed by the noise
reduction device 10B will be explained by using FIG. 6. FIG. 6 is a
flowchart illustrating a noise reduction process according to the
present embodiment. Additionally, a process that is identical to
the noise reduction process as illustrated in FIG. 4 will be
provided with an identical sign and a redundant explanation will be
omitted.
[0071] The noise reduction device 10B calculates a coefficient
.alpha. in a frequency domain that reduces a signal level L of a
detected noise signal Sn in a frequency component of the noise
signal Sn that is detected at step S104 (step S201). The noise
reduction device 10B transforms a coefficient .alpha. in a
frequency domain into a coefficient .beta. in a time domain (step
S202). The noise reduction device 10B reduces n noise signals Sn
from a received signal depending on a coefficient .beta. in a time
domain (step S203) and ends a process.
[0072] As stated above, the noise reduction device 10B according to
the present embodiment transforms a coefficient .alpha. calculated
in a frequency domain into a coefficient .beta. in a time domain,
so that it is possible to reduce n noise signals Sn, for example,
by the one filter 131B. Thereby, it is possible to reduce a
throughput and a circuit size of the reduction unit 130B as
compared with the noise reduction device 10A according to the
second embodiment.
[0073] Furthermore, whereas the noise reduction device 10A selects
a coefficient, for example, with reference to the coefficient DB
141, the noise reduction device 10B according to the present
embodiment adds a coefficient for each of frequency components of a
noise signal Sn to calculate a coefficient .alpha.. Thereby, it is
possible to reliably reduce, for example, even a noise signal Sn
with a frequency component that is not stored in the coefficient DB
141, from a received signal.
[0074] Furthermore, the number of noise signals Sn that are capable
of being reduced in the noise reduction device 10A is equal to the
number of the first to n-th filters 131 to 13n that are possessed
by the reduction unit 130A. On the other hand, it is possible for
the noise reduction device 10B according to the present embodiment
to calculate a coefficient .alpha. independently of the number of
noise signals Sn. Accordingly, even if only one filter 131B is
provided, it is possible to reduce all of a plurality of noise
signals Sn that are detected by the detection unit 110A.
[0075] Additionally, although the noise reduction device 10B
according to the present embodiment is provided with the one filter
131B of the reduction unit 130B, this is not limiting. It is
sufficient to transform coefficients that reduce a plurality of
noise signals Sn into coefficients in a time domain collectively,
and for example, two or more filters of the reduction unit 130B may
be provided.
[0076] (Variation)
[0077] Next, a variation of the third embodiment will be explained
by using FIG. 7 and FIG. 8. FIG. 7 is a block diagram illustrating
a configuration of a receiving device 1C according to the present
variation. The receiving device 1C according to the present
variation is different from the receiving device 1B that reduces a
noise signal Sn in a time domain in that a noise signal Sn is
reduced in a frequency domain.
[0078] The receiving device 1C has a noise reduction device 10C.
The noise reduction device 10C has a calculation unit 120C, a
reduction unit 130C, a domain transformation unit 150, and an
inverse transformation unit 160.
[0079] The calculation unit 120C is identical to the calculation
unit 120B in that it does not have the transformation unit 122. The
calculation unit 120C calculates a coefficient .alpha. in a
frequency domain.
[0080] The domain transformation unit 150 transforms a received
signal from a signal in a time domain into a signal in a frequency
domain. The domain transformation unit 150 executes an FFT process
for a received signal so that a received signal in a frequency
domain is generated.
[0081] The reduction unit 130C has an attenuator that reduces n
noise signals Sn from a received signal in a frequency domain
depending on a coefficient .alpha. in a frequency domain. The
reduction unit 130C reduces n noise signals Sn depending on a
coefficient .alpha. from a received signal in a frequency
domain.
[0082] The inverse transformation unit 160 transforms a received
signal in a frequency domain with n reduced noise signals Sn into a
received signal in a time domain. The inverse transformation unit
160 applies an IFFT process to a received signal with n reduced
noise signals Sn so that a received signal in a time domain with n
reduced noise signals Sn is generated.
[0083] <Noise Reduction Process>
[0084] A noise reduction process that is executed by the noise
reduction device 10C will be explained by using FIG. 8. FIG. 8 is a
flowchart illustrating a noise reduction process according to the
present variation. Additionally, a process that is identical to the
noise reduction process as illustrated in FIG. 6 will be provided
with an identical sign and a redundant explanation will be
omitted.
[0085] The noise reduction device 10C calculates a coefficient
.alpha. in a frequency domain at step S201, and then, transforms a
received signal into a signal in a frequency domain (step S301).
The noise reduction device 10C reduces a noise signal Sn from a
received signal in a frequency domain depending on a coefficient
.alpha. (step S302). The noise reduction device 10C transforms a
received signal with a reduced noise signal Sn into a received
signal in a time domain (step S303) and ends a process.
[0086] Additionally, although a coefficient .alpha. is calculated
at step S201, and subsequently, a received signal is transformed
into a signal in a frequency domain at step S301 herein, this is
not limiting. It is sufficient to transform it into a received
signal in a frequency domain before a noise signal Sn is reduced
from a received signal at step S302, and it may be executed in step
S101 to step S104. Alternatively, a process may be executed in
parallel to these processes.
[0087] As stated above, it is possible for the noise reduction
device 10C according to the present variation to obtain an effect
that is similar to that of the noise reduction device 10B according
to the third embodiment and reduce a plurality of noise signals Sn
from a received signal even in a frequency domain.
[0088] Additionally, although a process that returns a received
signal with a reduced noise signal Sn to that in a time domain is
executed in the above-mentioned variation, it is also possible to
omit such a process. For example, in a case where a signal process
is applied to a received signal in a frequency domain in a latter
process, the noise reduction device 10C may output a received
signal in a frequency domain to a (non-illustrated) latter
processing unit. In such a case, it is possible to omit the inverse
transformation unit 160.
[0089] Furthermore, although each of the analysis unit 111 and the
domain transformation unit 150 executes an FFT process in the
above-mentioned variation, this is not limiting. It is also
possible to integrate processes of the analysis unit 111 and the
domain transformation unit 150, where, for example, the difference
calculation unit 112 calculates a difference between upper and
lower sidebands Wm1 and Wm2 of a received signal by using a
received signal in a frequency domain that the domain
transformation unit 150 transforms it into, or the like.
[0090] As stated above, the noise reduction device 10 or 10A to 10C
according to each of the embodiments and the variation includes the
detection unit 110 or 110A, the calculation unit 120, 120A to 120C,
and the reduction unit 130 or 130A to 130C. The detection unit 110
or 110A detects frequency components of a plurality of noise
signals Sn that are included in a received signal, based on a
frequency spectrum of the received signal. The calculation unit 120
or 120A to 120C calculates a coefficient, based on frequency
components that are detected by the detection unit 110 or 110A. The
reduction unit 130 or 130A to 130C reduces n noise signal(s) Sn
among a plurality of noise signals Sn from a received signal,
depending on a coefficient that is calculated by the calculation
unit 120 or 120A to 120C (where n is a positive number greater than
or equal to 1).
[0091] Thereby, it is possible for the noise reduction device 10 or
10A to 10C according to each of the embodiments and the variation
to reduce a plurality of noise signals Sn from a received
signal.
[0092] Furthermore, the reduction unit 130A of the noise reduction
device 10A according to the second embodiment includes the n or
first to n-th filters 131 to 13n that filter n noise signals Sn,
respectively. Furthermore, the calculation unit 120A calculates a
coefficient for each of the n or first to n-th filters 131 to 13n
for each of n frequency components that are detected by the
detection unit 110A.
[0093] Thus, the noise reduction device 10A reduces a noise signal
Sn by using the n or first to n-th filters 131 to 13n, so that it
is possible to reduce a plurality of noise signals Sn from a
received signal.
[0094] Furthermore, the detection unit 110A of the noise reduction
device 10B according to the third embodiment detects signal levels
of n noise signals Sn. Furthermore, the calculation unit 120B
includes the coefficient calculation unit 121 and the
transformation unit 122. The coefficient calculation unit 121
calculates a coefficient .alpha. in a frequency domain, based on
frequency components and signal levels that are detected by the
detection unit 110A. The transformation unit 122 transforms a
coefficient .alpha. in a frequency domain that is calculated by the
coefficient calculation unit 121 into a coefficient .beta. in a
time domain. Furthermore, the reduction unit 130B reduces n noise
signals Sn from a received signal in a time domain depending on a
coefficient .beta. in a time domain that the transformation unit
122 transforms it into.
[0095] Thus, the noise reduction device 10B transforms a
coefficient .alpha. that is calculated in a frequency domain into a
coefficient .beta. in a time domain, so that it is possible to
reduce n noise signals Sn by, for example, the one filter 131B.
Thereby, it is possible to reduce a throughput and a circuit size
of the noise reduction device 10B according to the third
embodiment.
[0096] Furthermore, the detection unit 110A of the noise detection
device 10C according to the variation detects signal levels of n
noise signals Sn. Furthermore, the calculation unit 120C calculates
a coefficient .alpha. in a frequency domain based on frequency
components and signal levels that are detected by the detection
unit 110A. The reduction unit 130C reduces n noise signals Sn from
a received signal in a frequency domain depending on a coefficient
.beta. in a frequency domain that is calculated by the calculation
unit 120C.
[0097] Thus, it is possible for the noise reduction device 10C
according to the variation to obtain an effect that is similar to
that of the noise reduction device 10B according to the third
embodiment and reduce a plurality of noise signals Sn from a
received signal even in a frequency domain.
[0098] Furthermore, the detection unit 110A of the noise reduction
device 10A to 10C according to each of the embodiments and the
variation includes the difference calculation unit 112 and the
noise detection unit 113. The difference calculation unit 112
calculates a difference between upper and lower sidebands Wm1 and
Wm2 of a frequency spectrum of a received signal. The noise
detection unit 113 detects a frequency component and a signal level
based on a difference that is calculated by the difference
calculation unit 112.
[0099] Thus, the difference calculation unit 112 calculates a
difference between upper and lower sidebands Wm1 and Wm2 of a
received signal and the noise detection unit 113 detects a
plurality of noise signals Sn based on such a difference, so that
it is possible to detect a noise more accurately.
[0100] Furthermore, the noise detection unit 113 of the noise
reduction device 10A to 10C according to each of the embodiments
and the variation provides a signal level difference L between
upper and lower sidebands Wm1 and Wm2 at a frequency component of a
noise signal Sn as a signal level that is reduced by the reduction
unit 130A to 130C.
[0101] Thereby, it is possible for the noise reduction device 10A
to 10C to reduce a noise signal Sn accurately as compared with, for
example, a case where a received signal at a frequency component of
a noise signal Sn is reduced by a predetermined level.
[0102] It is possible for a person skilled in the art to readily
derive an additional effect or variation. Accordingly, a broader
aspect of the present invention is not limited to a specific detail
and a representative embodiment as illustrated and described above.
Therefore, various modifications are possible without deviating
from the spirit or scope of a general inventive concept that is
defined by the appended claim(s) and an equivalent(s) thereof.
* * * * *