KR20160118156A - Noise reducing device using active noise control and method for the same - Google Patents

Abstract

According to one embodiment of the present invention, a noise reducing device using active noise control comprises: a main body installed to be attached to and detached from a portion of a human body within a predetermined distance from an ear of a user; a plurality of microphone sensors installed in the main body and measuring a noise generated from a noise source; a processor installed in the main body and generating a control sound signal having the same size with and an opposite phase to a noise signal transmitted from the noise source by using an active noise control algorithm; and a plurality of speakers installed in the main body and outputting a control sound to reduce the noise by using the control sound signal.

Description

TECHNICAL FIELD [0001] The present invention relates to a noise reduction apparatus and method using active noise control,

Embodiments of the present invention relate to a noise reduction apparatus and method using active noise control.

As the industry develops, audio reproduction apparatuses capable of reproducing audio such as music are becoming portable. Portable audio playback devices can be used for music appreciation or language learning without being restricted to places such as subways, buses or streets. Examples of such portable audio players include MP3 players, mini cassettes and compact disc players.

Generally, a portable audio reproducing apparatus is a device for reproducing data recorded on a recording medium such as a memory or a CD, or a broadcast signal received from a broadcasting station. An audio signal reproduced by the portable audio reproducing apparatus is amplified through an amplifier and outputted to an earphone.

An earphone outputs an audio signal amplified unilaterally through an internal speaker unit. However, the earphone of the portable audio reproducing apparatus lowers the sound quality due to the noise introduced from the external environment. Therefore, the user gradually increases the sound pressure of the device in order to maintain the sound level in consideration of the external noise. Therefore, the portable audio reproducing apparatus using the earphone has a problem of causing hearing loss or the like when listening for a long time.

In order to solve such a problem, a headphone / earphone having a noise canceling function has been proposed. Conventional noise cancellation headphones / earphones are provided with a headphone / earphone unit, so that the sound quality of the headphone / earphone equipment is fixed, so it is not easy to select a device that satisfies both the active noise control function and the personal hearing characteristic . In addition, the noise cancellation headphone / earphone of the related art is able to feel a noise reduction effect only when the equipment is in close contact with the vicinity of the ear of the user, which limits the activity and convenience of the user.

Therefore, unlike the prior art or the product which requires the equipment to be in close proximity to the ear of the user, there is no restriction to place the equipment near the ear by using the active noise control algorithm using the virtual microphone technique or the like. It is necessary to develop a technique that minimizes the inconvenience caused by the use of the apparatus without restricting the activity of the user because the noise reduction function is provided to the body near the ear such as the neck or the head.

A related prior art is Korean Patent Publication No. 2007-0075664 (Title of the Invention: Noise Reduction Device and Method of Earphone Device, Published Date: 2007.07.24).

An embodiment of the present invention provides an apparatus and method for reducing noise using active noise control that can effectively reduce noise near a user's ear while providing a wearing method that assures the user's activity.

The problems to be solved by the present invention are not limited to the above-mentioned problem (s), and another problem (s) not mentioned can be clearly understood by those skilled in the art from the following description.

A noise reduction apparatus using active noise control according to an embodiment of the present invention includes a main body detachably installed on a part of a body within a predetermined distance from a user's ear; A plurality of microphone sensors installed in the main body and measuring noise generated from a noise source; A processor, installed in the main body, for generating a control signal having the same magnitude and phase as a noise signal transmitted from the noise source using an active noise control algorithm; And a plurality of speakers installed in the main body and outputting control sounds for reducing the noise using the control sound signal.

Wherein the processor is configured to operate a Filtered-x Least Mean Square (FxLMS) algorithm using the virtual microphone in consideration of physical distances between the plurality of speakers, the plurality of microphone sensors, and virtual microphones existing at ear positions of the user .

Wherein the processor estimates a first sound transmission path between the plurality of microphone sensors and the plurality of speakers, estimates a second sound transmission path between the plurality of microphone sensors and the virtual microphone, The FxLMS algorithm can be operated using the error signal defined by the difference between the control signal passed through the transmission path and the noise signal transmitted from the noise source and the estimated second sound transmission path.

The processor can operate the FxLMS algorithm such that the power of the error signal is reduced in consideration of the estimated second sound transmission path.

The processor may operate the FxLMS algorithm such that the value of the cost function defined based on the error signal and the estimated second sound transfer path is reduced.

The value of the cost function can be calculated based on the following equation (1).

[Equation 1]

| e _v (n) | ² = | e _p (n) * h ^{^} (n) | ²

Here, | e _v (n) | ² is the value of the cost function, e _p (n) is the error signal, h ^{^} (n) is a second sound transmission path, the estimated, * denotes the convolution operation, respectively.

The processor can estimate the first sound transmission path by using an LMS algorithm that inputs and outputs white noise output from the plurality of speakers and modified white noise input to the plurality of microphone sensors respectively .

Wherein the processor includes a first modified white noise input to the plurality of microphone sensors corresponding to white noise output from the plurality of speakers and a second modified white noise input to the virtual microphone corresponding to white noise output from the plurality of speakers, It is possible to estimate the second sound transmission path by using an LMS algorithm that uses input and output of the modified white noise.

The processor may operate a multi-channel ANC algorithm using a stereo array-based stereophonic technique in consideration of the physical distance between the plurality of speakers, the plurality of microphone sensors, and the ear position of the user.

Wherein the processor is further configured to determine the direction of the user's ear position relative to the control sound output from each of the plurality of speakers based on the physical distance and direction between the plurality of speakers and the user & The sizes and phases of the control signals may be set differently from each other by adjusting the delays and gains of the control signals.

A noise reduction method using active noise control according to an embodiment of the present invention includes a plurality of microphone sensors installed on a main body detachably attached to a part of a body within a predetermined distance from a user's ear and measuring noise generated from a noise source ; Generating a control sound signal having the same magnitude and phase as a noise signal transmitted from the noise source using an active noise control algorithm in a processor installed in the main body; And outputting a control sound for reducing the noise by using the control sound signal from a plurality of speakers installed in the main body.

Wherein the step of generating the control sound signal comprises the steps of: calculating a filtered-x Least (FxLMS) using the virtual microphone in consideration of the physical distance between the plurality of speakers, the plurality of microphone sensors, Mean Square < / RTI >

Wherein the step of generating the control signal comprises: estimating a first sound transmission path between the plurality of microphone sensors and the plurality of speakers; Estimating a second sound transmission path between the plurality of microphone sensors and the virtual microphone; And operating the FxLMS algorithm using an error signal defined as a difference between the control signal passed through the estimated first sound transmission path and a noise signal transmitted from the noise source and the estimated second sound transmission path . ≪ / RTI >

The step of operating the FxLMS algorithm may include operating the FxLMS algorithm to reduce the power of the error signal, taking into account the estimated second sound transfer path.

Operating the FxLMS algorithm may include operating the FxLMS algorithm such that the value of the cost function defined based on the error signal and the estimated second sound transfer path is reduced.

The value of the cost function can be calculated based on the following equation (1).

[Equation 1]

| e _v (n) | ² = | e _p (n) * h ^{^} (n) | ²

Here, | e _v (n) | ² is the value of the cost function, e _p (n) is the error signal, h ^{^} (n) is a second sound transmission path, the estimated, * denotes the convolution operation, respectively.

Wherein the estimating of the first sound transmission path includes a step of estimating a first sound transmission path by using an LMS algorithm that inputs and outputs white noise output from the plurality of loudspeakers and modified white noise input to the plurality of microphone sensors, And estimating the sound transmission path.

Wherein the step of estimating the second sound transmission path includes a first modified white noise that is input to the plurality of microphone sensors corresponding to white noise output from the plurality of speakers and a second modified white noise that corresponds to the white noise output from the plurality of speakers And estimating the second sound transmission path using an LMS algorithm using input and output of the second modified white noise input to the virtual microphone.

The generating of the control sound signal may include operating the multi-channel ANC algorithm using a speaker array based stereophonic technique in consideration of the physical distance between the plurality of speakers, the plurality of microphone sensors, and the ear position of the user Step < / RTI >

Wherein the act of operating the multi-channel ANC algorithm comprises: determining physical distance and direction between the plurality of speakers and the user ' s ear to direct the user ' s ear position with respect to the control sound output from each of the plurality of speakers And adjusting the delay and the gain of each of the plurality of speakers based on the magnitude and phase of the control signal.

The details of other embodiments are included in the detailed description and the accompanying drawings.

According to an embodiment of the present invention, it is possible to effectively reduce the noise near the user's ear while providing a wearable manner for assuring the user's activity.

In accordance with an embodiment of the present invention, independent active noise control hardware is incorporated, so that no additional equipment is required for noise reduction purposes.

1 is a block diagram of a noise reduction apparatus using active noise control according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating an appearance of a noise reduction apparatus using active noise control according to an embodiment of the present invention. Referring to FIG.
3 and 4 are views showing the wearing state of the noise reducing apparatus using active noise control according to an embodiment of the present invention.
5 and 6 are diagrams for explaining the operation of the FxLMS algorithm using a virtual microphone in an embodiment of the present invention.
7 is a view for explaining the operation of a multi-channel ANC algorithm using a stereo array based speaker array in another embodiment of the present invention.
8 to 10 are flowcharts for explaining a noise reduction method using active noise control according to an embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and / or features of the present invention, and how to accomplish them, will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. It should be understood, however, that the invention is not limited to the disclosed embodiments, but is capable of many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, To fully disclose the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

The present invention aims at reducing noise near the user's ear, and ideally it is necessary to position a microphone (microphone sensor) for noise acquisition near the ear. However, according to the present invention, the position of the microphone for noise acquisition is separated from the ear position of the user by a certain distance according to the part to be placed on the body. In addition, since the physical distance between the speaker and the user's ear is relatively far from that of the conventional acoustic device, the anti-phase noise signal (control sound signal) is transmitted not only to the user's ears but also to the user's surroundings, There is a risk of interference.

Accordingly, in the present invention, by using a virtual sensor technique (FxLMS algorithm using a virtual microphone) and (2) a directional sound source output method using a speaker array (a multi-channel ANC algorithm using a stereo array based speaker array) Can overcome the limitations of the prior art.

(1) In the case of the virtual sensor technique, the acoustic path from the noise acquisition microphone (microphone sensor) to the virtual sensor (virtual microphone) existing at the user's ear position is predicted, and the acoustic path and noise acquisition micro- It is possible to effectively reduce the noise near the ear by predicting the noise signal of the virtual microphone input near the ear and operating the algorithm in the direction of minimizing the noise signal.

Next, in the case of (2) a directional sound source output method using a speaker array, the size and phase of signals output from the respective speakers are set differently according to the distance and direction between the respective speakers and the user's ear, And thus the influence of the speaker output signal propagating to the user's surroundings is minimized, so that the noise near the ear can be effectively reduced.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram of a noise reduction apparatus using active noise control according to an embodiment of the present invention, and FIG. 2 is a view illustrating an appearance of a noise reduction apparatus using active noise control according to an embodiment of the present invention .

1 and 2, a noise reduction apparatus 100 using active noise control according to an embodiment of the present invention includes a main body 105, a plurality of microphone sensors 110, a processor 120, (130), and a power supply (140).

The body 105 is detachably attached to a part of the body within a predetermined distance from the user's ear.

For example, the body 105 may be removably worn in a manner similar to wearing a necklace around a neck within a certain distance from the user ' s ear 310, as shown in FIG. Alternatively, the body 105 may be detachably attached to the head, such as a hair band, within a certain distance from the user's ear 310, as shown in FIG.

For this, the main body 105 may be formed in a partially ring-shaped shape. However, the present invention is not limited to this, and the body 105 may be formed in a different shape (e.g., a quadrangle) so long as the body 105 can be detachably attached to a part of the body within a predetermined distance from the user's ear.

A plurality of microphone sensors 110 and a plurality of speakers 130 may be provided outside the main body 105. The main body 105 may include a processor 120 and a power supply unit 140 ) May be provided.

As described above, the main body 105 may function as a casing means for protecting the components provided therein, and may also function as a mounting means for detachably attaching to a part of the user's body.

The plurality of microphone sensors 110 are installed in the main body 105. At this time, the plurality of microphone sensors 110 may be arranged in parallel on opposite sides of the main body 105. The plurality of microphone sensors 110 may be spaced apart from each other by a predetermined distance so as to reduce shaded areas where noise is not detected, and may be arranged to have different orientations.

In the present embodiment, a total of eight microphone sensors 110 may be disposed on each of opposite sides of the main body 105, four in total. However, the present invention is not limited thereto, and the plurality of microphone sensors 110 may be disposed on four opposite sides of the main body 105 on opposite sides thereof, or may be arranged in four or less positions.

The plurality of microphone sensors 110 measure noise generated from a noise source. The plurality of microphone sensors 110 may generate a noise signal, which is an electrical signal corresponding to the measured noise, and transmit the generated noise signal to a processor 120 to be described later.

In this embodiment, when the diaphragm is displaced according to the sound pressure of the noise generated from the noise source, the plurality of microphone sensors 110 may be a direct-current-type electrostatic microphone for measuring noise by changing capacitance have. However, the present invention is not limited to this, and a different type of dynamic microphone, such as a moving coil type microphone, may be used as the plurality of microphone sensors 110.

The processor 120 is installed in the main body 105. At this time, the processor 120 may be installed in an internal space between opposite surfaces of the main body 105. That is, the processor 120 may be installed inside the main body 105 of the area where the plurality of microphone sensors 110 and a plurality of speakers 130 described later are not installed.

The processor 120 may receive the noise signal from the plurality of microphone sensors 110, convert the analog signal into an analog-digital signal, and analyze the frequency and magnitude (sound pressure) of the noise signal. The processor 120 generates a control sound signal having the same magnitude and opposite phase as the noise signal using the active noise control algorithm based on the frequency and size of the detected noise signal.

At this time, a physical distance difference occurs between the ear position of the user and the plurality of microphone sensors 110 and a plurality of speakers 130 described later. This is because the main body 105 on which the plurality of microphone sensors 110 and a plurality of speakers 130 described later are installed is worn on the body of a part spaced a certain distance from the user's ear (see FIGS. 3 and 4) ). This difference in physical distance can cause the noise reduction effect to deteriorate. Therefore, in an embodiment of the present invention, a FxLMS (Filtered-x Least Mean Square) algorithm using a virtual microphone may be used to compensate for the physical distance difference.

That is, the processor 120 determines the physical distance between the plurality of microphone sensors 110, a plurality of speakers 130 described later, and virtual microphones (see "510" It is possible to operate an FxLMS (Filtered-x Least Mean Square) algorithm using the virtual microphone.

More specifically, the processor 120 may estimate a first sound transmission path between the plurality of microphone sensors 110 and a plurality of speakers 130 described below. To this end, the processor 120 uses an LMS algorithm that inputs / outputs white noise output from the plurality of speakers 130 and modified white noise input to the plurality of microphone sensors 110 So as to estimate the first sound transmission path.

Then, the processor 120 may estimate a second sound transmission path between the plurality of microphone sensors 110 and the virtual microphone. The processor 120 may include a first modified white noise input to the plurality of microphone sensors 110 corresponding to white noise output from the plurality of speakers 130, It is possible to estimate the second sound transmission path by using the LMS algorithm which inputs and outputs the second modified white noise (see "520" in FIG. 5) input to the virtual microphone corresponding to the outputted white noise.

Next, the processor 120 calculates an error signal, which is defined as a difference between the control signal passed through the estimated first sound transmission path and the noise signal transmitted from the noise source, outputted from the plurality of speakers 130, And operating the FxLMS algorithm using the estimated second sound transfer path.

At this time, the processor 120 may operate the FxLMS algorithm such that the power of the error signal is reduced in consideration of the estimated second sound transmission path. That is, the processor 120 may operate the FxLMS algorithm such that the value of the cost function defined based on the error signal and the estimated second sound transfer path is reduced.

Here, the value of the cost function can be calculated based on the following equation (1).

[Equation 1]

| e _v (n) | ² = | e _p (n) * h ^{^} (n) | ²

Here, | e _v (n) | ² is the value of the cost function, e _p (n) is the error signal, h ^{^} (n) is a second sound transmission path, the estimated, * denotes the convolution operation, respectively.

In another embodiment of the present invention, a multi-channel ANC algorithm using a stereo array based speaker array may be used to compensate for the physical distance difference.

That is, the processor 120 uses the speaker array-based stereo sound technique in consideration of the physical distance between the plurality of microphone sensors 110, a plurality of speakers 130 described later, and the ear position of the user The multi-channel ANC algorithm can be operated.

More specifically, the processor 120 may direct the user to the ear position of the control sound output from each of the plurality of speakers 130. To do this, the processor 120 determines the delay and gain of each of the plurality of speakers 130 based on the physical distance and direction between the plurality of speakers 130 and the user's ear Can be adjusted. Accordingly, the processor 120 can set the sizes and phases of the control sound signals to be different from each other, whereby the control sounds output from the plurality of speakers 130 are concentrated to the ear position of the user, It is possible to reduce the noise generated from the speaker.

The plurality of speakers 130 are installed in the main body 105. At this time, the plurality of speakers 130 may be disposed at adjacent positions corresponding to the plurality of microphone sensors 110, respectively. That is, the plurality of loudspeakers 130 may be arranged on both sides of the main body 105, respectively.

The plurality of speakers 130 output control sounds for reducing noise by using control sound signals generated by the processor 120. Here, it is preferable that the control sound has the same magnitude and opposite phase as the noise. At this time, the control sounds do not have the same magnitude and opposite phase (reverse phase) as the noise at the position output from the plurality of speakers 130, but have the same magnitude and reverse phase . Therefore, the noise generated from the noise source is canceled by the control sound, so that the noise can be effectively reduced.

In the meantime, when the multi-channel ANC algorithm using the stereo array based stereophonic technique is operated by the processor 120, the plurality of speakers 130 may be arranged in the same direction as the noise source, The opposite phases of the noise can be generated as the control sounds, respectively, so as to be in opposite phases.

The power supply unit 140 serves to supply power to the plurality of microphone sensors 110, the processor 120, and the plurality of speakers 130. The power supply unit 140 may be installed inside the main body 105.

5 and 6 are diagrams for explaining the operation of the FxLMS algorithm using a virtual microphone in an embodiment of the present invention.

Referring to FIGS. 5 and 6, in the case of the active noise reduction system using a virtual microphone, the steps performed in the algorithm are three stages in total.

The first step is Estimating the transmission path S (z), i.e., the first sound transmission path, of how the sound generated in the speaker 130 sounds in the actual microphone 110. [ This step is preferably performed before activating the noise reduction function. In FIG. 6, it is a step of estimating S (z) which is a transmission path (first sound transmission path) between the speaker 130 and the microphone 110. The estimated result is S ^{^} (z) and included in the FxLMS algorithm part using virtual microphone.

Specifically, the propagation path S (z) is estimated using the LMS algorithm as a system identification problem. To this end, white noise is emitted from the speaker 130, and the distorted white noise heard by the microphone 110 is measured. The white noise of the speaker 130 is input to the transmission path S (z), and the modified white noise is output. From this input / output data, the impulse response of the propagation path S (z) can be estimated using the LMS algorithm. This process is called system identification in that it obtains the impulse response information of the actual transmission path system.

In the second step, the sound heard at the actual microphone 110 is different from the sound heard at the virtual microphone (see "510" in Fig. 5 and 610 in Fig. 6) 510, and 610, respectively. This step is also preferably performed before the noise reduction function is activated. In FIG. 6, it is a step of estimating H (z) as a transmission path between the actual microphone 110 and the virtual microphone 610. The estimated result is H ^{^} (z) and included in the FxLMS algorithm part using virtual microphone.

More specifically, the method of estimating the propagation path H (z) between the actual microphone 110 and the virtual microphones 510 and 610 is similar to the method of estimating S (z) in the first step. The microphone 1 is disposed at the position of the actual microphone 110 and the microphone 2 is disposed at the position of the virtual microphones 510 and 610. Then, the speaker 130 outputs white noise. Then, the modified white noise is input to the microphone 1 and the microphone 2, respectively. The input of the transmission path H (z) is the modified white noise introduced into the microphone 1, and the output of the transmission path H (z) is another modified white noise entering the microphone 2. Since there is input / output data in this way, the impulse response information of the propagation path H (z) can be estimated using the LMS algorithm.

In the third step, the FxLMS algorithm is operated considering the propagation path H ^{^} (z) between the actual microphone 110 and the virtual microphones 510 and 610 thus estimated. As a result, the FxLMS algorithm passes the transmission path S ^{^} (z) between the speaker 130 and the actual microphone 110 and the transmission path H ^{^} (z) between the speaker 130 and the virtual microphones 510 and 610 There is a difference from the existing algorithm in that the algorithm operates so that the power of a signal e _v (n) is reduced. The virtual microphone 110 and the virtual microphones 110 and 610 are not located in the actual microphone 110 through the FxLMS algorithm that operates in consideration of the transmission path H ^{^} (z) between the actual microphone 110 and the virtual microphones 510 and 610 as well as S ^{^} (z) 510 and 610, respectively.

Here, the signal e _v (n) is then passed through the estimated transmission path S ^{^} (z), the actual microphone 110, noise d _p (n) and speaker generated in the negative y _p (n) corresponding to the error signal, The combined sound of two sounds, e _p (n), passes through H ^{^} (z). In addition, the FxLMS algorithm S ^{^} (z) and H ^{^} was (z) operates to reduce the sound after passing through the so name is attached called filtered-x (Fx), e v (n) operates to reduce the power of the Therefore, the least mean square (LMS) is given.

More specifically, the third step will be described below using the mathematical expression.

The FxLMS algorithm is a modified version of the LMS algorithm. LMS algorithm is input to the noise x (n), y (n ) = a a e (n) the difference between the modified noise make it a output coming through the sound transmission path, d _p (n), e (n) d _p ( (n) -y (n) = d _p (n) -x (n) ^T w (n). When switching to the optimization problem, it is a goal to reduce the cost function of the following equation (2) for every sample n (natural number).

&Quot; (2) "

Both the above and the following optimization problems differ only in the cost function J (n) and can be solved using the steepest descent method.

&Quot; (3) "

FxLMS algorithm noise x (n) as inputs to the x ^{^} (n) signal which has passed through the transmission path S ^{^} (z) which are estimated, samneunda the y (n) as output. And the modified noise coming through the sound transmission path, d _p (n) and y (n) of the error signal the difference between the transfer path S (z) a signal is y _p (n) passed through the e _p (n), d _{p (n) -y p (n)} = d p (n) - a (s (n) * x ( n)) T w (n) = d p (n) -x ^ (n) T w (n) It works to reduce squared. If this is switched back to the optimization problem, the goal is to reduce the cost function of equation (4) for each sample n.

&Quot; (4) "

Finally, to reduce the signal passed through H ^{^} (z), the cost function of the FxLMS algorithm | e _p (n) | ² | e _v (n) | ² = | e _p (n) * h ^{^} (n) | ^2, and operates the algorithm so that this value is reduced. Likewise, if we turn this into an optimization problem, the goal is to reduce the cost function of Equation 5 below.

&Quot; (5) "

7 is a view for explaining the operation of a multi-channel ANC algorithm using a stereo array based speaker array in another embodiment of the present invention.

Stereophony is a method of realizing a virtual sense of space by utilizing the wavelength characteristics of sound appearing using many speakers. Specifically, delay and gain are adjusted to each speaker to give directionality.

As shown in FIG. 7, when a plurality of speakers 130 are used, the speakers 130 are operated at positions different from the positions of the noises. However, these synthesized sounds can be made to produce sounds of opposite phases at the positions of the noise sources. This technique can be implemented using a multi-channel ANC (Active Noise Control). The multi-channel ANC algorithm is an algorithm that generates control sounds through a plurality of speakers 130 to minimize the sound pressure errors of the left and right microphones regardless of the positions of the noises. At this time, the multi-channel ANC algorithm generates a control sound source at the same position as the noise source by generating control sounds in different sizes and phases through the plurality of speakers 130, do. That is, when a multi-channel ANC algorithm is implemented, directional control sounds can be generated in a plurality of speakers 130. [

Specifically, when a multi-channel ANC is used, control sounds having different sizes and phases are generated in the respective speakers 130. However, this regulation is the result of a multi-channel FxLMS algorithm that maximizes noise reduction in the ear by summing each speaker control sound. To maximize noise reduction, 1) the direction is the same direction as the noise source, and 2) the phase is opposite to the noise. Accordingly, since each speaker 130 operates so as to change its size and phase according to the incoming direction of noise, directionality can be given to the sound source (controlled sound) of each speaker 130 according to the direction in which noise is introduced .

8 to 10 are flowcharts for explaining a noise reduction method using active noise control according to an embodiment of the present invention.

Referring to FIGS. 1 and 8, in step 810, the plurality of microphone sensors 110 measures noise generated from a noise source.

Next, in step 820, the processor 120 generates a control signal having the same magnitude and phase as the noise signal transmitted from the noise source using the active noise control algorithm.

1 and 9, in step 910, the processor 120 estimates a first sound transmission path between the plurality of microphone sensors 110 and the plurality of speakers 130. In step 910, Thereafter, in step 920, the processor 120 estimates a second sound transmission path between the plurality of microphone sensors 110 and the virtual microphones 510, 610. Thereafter, in step 930, the processor 120 calculates an error signal, which is defined as the difference between the control sound signal passing through the estimated first sound transfer path and the noise signal transmitted from the noise source, The FxLMS algorithm is operated using the propagation path. At this time, the processor 120 may operate the FxLMS algorithm such that the power of the error signal is reduced in consideration of the estimated second sound transmission path.

1 and 10, since the processor 120 operates as a multi-channel ANC in step 1010, the delay and gain of each of the plurality of speakers 130 gain of the control sound signal are adjusted so that the control sound signals have different sizes and phases.

Next, referring again to FIGS. 1 and 8, in step 830, the plurality of speakers 130 outputs a control sound for reducing the noise using the control sound signal.

Embodiments of the present invention include computer readable media including program instructions for performing various computer implemented operations. The computer-readable medium may include program instructions, local data files, local data structures, etc., alone or in combination. The media may be those specially designed and constructed for the present invention or may be those known to those skilled in the computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floppy disks, and ROMs, And hardware devices specifically configured to store and execute the same program instructions. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the scope of the appended claims and equivalents thereof.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, Modification is possible. Accordingly, the spirit of the present invention should be understood only in accordance with the following claims, and all equivalents or equivalent variations thereof are included in the scope of the present invention.

105: Body
110: microphone sensor
120: Processor
130: Speaker
140: Power supply
510, 610: virtual microphone

Claims (15)

A body detachably attached to a part of a body within a predetermined distance from a user's ear;
A plurality of microphone sensors installed in the main body and measuring noise generated from a noise source;
A processor, installed in the main body, for generating a control signal having the same magnitude and phase as a noise signal transmitted from the noise source using an active noise control algorithm; And
A plurality of speakers provided in the main body and outputting control sounds for reducing the noise by using the control sound signals;
Wherein the active noise control apparatus comprises:
The method according to claim 1,
The processor
(FxLMS) algorithm using the virtual microphone in consideration of the physical distance between the plurality of speakers, the plurality of microphone sensors, and virtual microphones existing at the ear position of the user Noise reduction device using active noise control.
3. The method of claim 2,
The processor
Estimating a first sound transmission path between the plurality of microphone sensors and the plurality of speakers, estimating a second sound transmission path between the plurality of microphone sensors and the virtual microphone, Wherein the FxLMS algorithm is operated using an error signal defined by a difference between the control signal passed and the noise signal transmitted from the noise source and the estimated second sound transfer path. Device.
The method of claim 3,
The processor
And the FxLMS algorithm is operated so that the power of the error signal is reduced in consideration of the estimated second sound transfer path.
5. The method of claim 4,
The processor
Wherein the FxLMS algorithm is operated so that the value of the cost function defined based on the error signal and the estimated second sound transfer path is reduced.
6. The method of claim 5,
The value of the cost function
Wherein the noise reduction apparatus is calculated based on the following equation (1).
[Equation 1]
| e _v (n) | ² = | e _p (n) * h ^{^} (n) | ²
Here, | e _v (n) | ² is the value of the cost function, e _p (n) is the error signal, h ^{^} (n) is a second sound transmission path, the estimated, * denotes the convolution operation, respectively.
The method of claim 3,
The processor
Wherein the first acoustic propagation path is estimated by using an LMS algorithm that inputs and outputs white noise output from the plurality of loudspeakers and modified white noise input to the plurality of microphone sensors as input and output, Noise Reduction Device Using Noise Control.
The method of claim 3,
The processor
A first modified white noise input to the plurality of microphone sensors corresponding to white noise output from the plurality of speakers and a second modified white noise input to the virtual microphone corresponding to white noise output from the plurality of speakers, Wherein the second sound transmission path is estimated using an LMS algorithm using input and output as inputs and outputs.
The method according to claim 1,
The processor
Wherein a multi-channel ANC algorithm using a speaker array based stereophonic technique is operated in consideration of the physical distance between the plurality of speakers, the plurality of microphone sensors, and the ear position of the user Noise reduction device.
10. The method of claim 9,
The processor
A plurality of loudspeakers each having a plurality of loudspeakers, each loudspeaker having a plurality of loudspeakers, each loudspeaker having a plurality of loudspeakers, wherein a delay and a gain of the control signal are adjusted so that magnitude and phase of each of the control sound signals are set differently from each other.
Measuring a noise generated from a noise source in a plurality of microphone sensors installed in a main body detachably attachable to a part of a body within a predetermined distance from a user's ear;
Generating a control sound signal having the same magnitude and phase as a noise signal transmitted from the noise source using an active noise control algorithm in a processor installed in the main body; And
Outputting a control sound for reducing the noise by using the control sound signal from a plurality of speakers installed in the main body
And a noise reduction method using active noise control.
12. The method of claim 11,
The step of generating the control signal
Operating a FxLMS (Filtered-x Least Mean Square) algorithm using the virtual microphone in consideration of a physical distance between the plurality of speakers, the plurality of microphone sensors, and a virtual microphone existing at an ear position of the user
And a noise reduction method using active noise control.
13. The method of claim 12,
The step of generating the control signal
Estimating a first sound transmission path between the plurality of microphone sensors and the plurality of speakers;
Estimating a second sound transmission path between the plurality of microphone sensors and the virtual microphone; And
Operating the FxLMS algorithm using an error signal defined by a difference between the control signal passed through the estimated first sound transmission path and a noise signal transmitted from the noise source and the estimated second sound transmission path,
And a noise reduction method using active noise control.
12. The method of claim 11,
The step of generating the control signal
Operating a multi-channel ANC algorithm using a stereo array based speaker array scheme in consideration of the physical distance between the plurality of speakers, the plurality of microphone sensors, and the ear position of the user
And a noise reduction method using active noise control.
15. The method of claim 14,
The step of operating the multi-channel ANC algorithm
A plurality of loudspeakers each having a plurality of loudspeakers, each loudspeaker having a plurality of loudspeakers, each loudspeaker having a plurality of loudspeakers, setting the magnitude and phase of each control sound signal to be different from each other by adjusting delay and gain,
And a noise reduction method using active noise control.

Images