KR102392460B1 - Apparatus and method of reducing noise - Google Patents

Abstract

A noise canceling apparatus and method are disclosed. The noise removal device according to an embodiment of the present invention includes at least one microphone module for sound pickup for generating a noise pickup signal by collecting sound from a medium, and vibration corresponding to the noise removal signal generated based on the noise pickup signal. and at least one speaker driver that transmits to a medium and a controller that generates the noise removal signal based on the noise pickup signal.

Description

Apparatus and method for noise cancellation

The present invention relates to an apparatus and method for removing noise, and to a technology for generating a noise removing signal based on a noise collecting signal collected through a microphone module, and outputting the noise canceling signal.

Unless otherwise indicated herein, the material described in this section is not prior art to the claims of this application, and inclusion in this section is not an admission that it is prior art.

The most basic method for removing noise is to generate an inverse signal of the same level as the noise to be removed to extinguish the noise.

However, although the above method is possible when it is close to the ear, such as earphones or headphones, there is a problem in that it is difficult to remove the noise generated in the space.

Noise radiated into the air is easily deformed under the influence of diffraction, interference, reflection, etc., and it is virtually impossible to generate a signal to cancel it.

However, if noise transmitted through a medium, such as interlayer noise, is removed in the medium stage before being radiated into the air, noise radiated into the air can be prevented.

Methods for removing noise in the medium stage include installing carpets or cushioning mats that can dampen vibrations in the medium, or functional construction for sound absorption. There is a problem.

In addition, if the medium is difficult to reinforce after construction, such as in a building, there is a problem in that additional construction is more difficult.

Korean Patent Registration No. 10-1365607, registered on February 14, 2014 (Title: Smart TV that removes noise in separated space, noise canceling device and smart TV system)

It is an object of the present invention to eliminate noise transmitted through the medium.

Another object of the present invention is to prevent the vibration of the speaker driver installed at the same point from being transmitted to the microphone module.

It is also an object of the present invention to block sound leakage of a speaker driver.

Another object of the present invention is to provide a noise canceling device that can be combined with general appliances such as electric lights and air conditioners.

Another object of the present invention is to accurately remove noise by analyzing the location of the noise source.

In addition, it is not limited to the above-described objects, and it is obvious that other objects may be derived from the following description.

In order to achieve the above object, an apparatus for removing noise according to an embodiment of the present invention includes one or more microphone modules for collecting noise for generating a noise-collecting signal by collecting sound from a medium, and a noise-removing signal generated based on the noise-collecting signal. and at least one speaker driver for transmitting a vibration corresponding to the to the medium, and a controller for generating the noise removal signal based on the noise-collecting signal.

In this case, a plurality of the microphone modules for sound pickup are attached to the medium, and a direction corresponding to the noise is detected using the noise pickup signals collected by the plurality of microphone modules for sound pickup, and the noise is removed based on the direction. A signal may be generated.

In this case, the noise collection signals are used to calculate the position of the sound source corresponding to the noise, and the noise removal signal may be generated based on the position of the sound source.

In this case, the speaker driver is provided in plurality, distances between each of the plurality of speaker drivers and the sound source are calculated, and a delay corresponding to at least one of the distances corresponds to at least one of the plurality of speaker drivers It can be applied to the noise canceling signal.

In this case, some of the plurality of speaker drivers generate the noise removal signal for removing noise corresponding to the sound source, and other portions of the plurality of speaker drivers attenuate vibration for removing the noise. It can generate damped vibration for

In this case, the plurality of sound-collecting microphone modules and the plurality of speaker drivers may be installed in one structure attached to the medium.

At this time, a honeycomb resonator for accommodating the one or more sound-collecting microphone modules and the one or more speaker drivers, and removing sound leakage occurring at the rear of the speaker driver and low-level noise transmitted from the medium. can do.

At this time, the honeycomb resonator, the inside is divided into a honeycomb structure, the partition wall dividing one or more honeycomb structures into one space may be formed.

In this case, the honeycomb resonator may have different heights of the bottom surfaces of the honeycomb structures formed therein in order to increase the noise absorption and diffuse reflection of the noise.

In this case, the barrier rib may have a through hole having a size corresponding to the frequency to be removed in the space formed by the barrier rib.

In this case, the speaker driver may further include a resonance unit coupled to the rear surface of the speaker driver and formed in a multi-chamber manner in order to cancel the sound leakage generated from the rear surface of the speaker driver.

In this case, the controller calculates a first fundamental frequency value based on the position of the sound source and the noise pickup signal, and generates a first noise removal signal corresponding to the first fundamental frequency value. to calculate a second fundamental frequency value based on the noise pickup signal transmitted to the speaker driver and from which a wavelength corresponding to the first fundamental frequency value is removed, and corresponding to the second fundamental frequency value A second noise canceling signal is generated and transmitted to the speaker driver, and the speaker driver transmits vibrations corresponding to the first noise canceling signal and the second noise canceling signal received by the controller in chronological order to the medium. can

In this case, the controller may predict a Chladni pattern according to the structure information of the medium input by the user, and generate the noise removal signal based on the pattern and the noise collection signal.

In this case, the controller calculates a fundamental frequency value and a harmonics frequency value based on the position of the sound source and the noise pickup signal, and corresponds to the fundamental frequency value and the harmonics frequency value to simultaneously generate the waveforms, and the noise removal signal may be generated based on the simultaneously generated waveforms.

In addition, in order to achieve the above object, in the noise removal method according to an embodiment of the present invention, in the method for removing noise through a noise removal device, a sound is collected from a medium through a microphone module for sound pickup to generate a noise pickup signal. generating, generating a noise-removing signal based on the noise-collecting signal, and transmitting a vibration corresponding to the noise-removing signal to the medium through a speaker driver.

At this time, a plurality of the microphone modules for sound pickup are attached to the medium, and the step of generating the noise removal signal detects a direction corresponding to the noise by using the noise pickup signals collected by the plurality of microphone modules for sound pickup. and the noise removal signal may be generated based on the direction.

In this case, the generating of the noise removal signal includes calculating the position of the sound source corresponding to the noise based on the noise collection signals and generating the noise removing signal based on the position of the sound source. may include

At this time, the speaker driver is provided in plurality, and calculating the distances between each of the plurality of speaker drivers and the sound source and delaying the delay corresponding to at least one of the distances to at least one of the plurality of speaker drivers The method may further include applying to a noise removal signal corresponding to .

In this case, the step of transmitting the vibration to the medium may include transmitting vibration corresponding to the noise removal signal for removing noise corresponding to the sound source to the medium through some of the plurality of speaker drivers, and Another part of the plurality of speaker drivers may include transmitting a damped vibration for attenuating the vibration to the medium.

In this case, the method may further include removing sound leakage and noise generated at the rear side of the speaker driver through a honeycomb resonator accommodating the sound pickup microphone module and the speaker driver.

At this time, the honeycomb resonator, the inside is divided into a honeycomb structure, the partition wall dividing one or more honeycomb structures into one space may be formed.

In this case, the honeycomb resonator may have different heights of the bottom surfaces of the honeycomb structures formed therein in order to increase diffuse reflection of noise absorbed therein.

In this case, the barrier rib may have a through hole having a size corresponding to the frequency to be removed in the space formed by the barrier rib.

In this case, the generating of the noise removing signal includes calculating a first fundamental frequency value based on the noise pickup signal, and generating a first noise removing signal corresponding to the first fundamental frequency value. generating, calculating a second fundamental frequency value based on a noise pickup signal from which a wavelength corresponding to the first fundamental frequency value is removed, and a second noise removing signal corresponding to the second fundamental frequency value and transmitting the vibration to the medium, the vibration corresponding to the first noise removal signal and the second noise removal signal may be sequentially transmitted to the medium through the speaker driver. .

In this case, the generating of the noise removal signal includes predicting a Chladni pattern according to the structure information of the medium input by a user, and removing the noise based on the pattern and the noise pickup signal generating a signal.

In this case, the generating of the noise removal signal includes calculating a fundamental frequency value and a harmonics frequency value based on the noise pickup signal, and the fundamental frequency value and the harmonics frequency value. It may include simultaneously generating a corresponding waveform and generating the noise removal signal based on the waveform.

According to the present invention, it is possible to remove noise transmitted through the medium.

In addition, according to the present invention, it is possible to accurately remove noise by analyzing the location of the noise source.

In addition, according to the present invention, it is possible to prevent the vibration of the speaker driver installed at the same point from being transmitted to the microphone module.

In addition, according to the present invention, it is possible to block the sound leakage of the speaker driver that removes the noise.

In addition, according to the present invention, it is possible to provide a noise canceling device capable of fusion with general appliances such as electric lights and air conditioners.

Effects of the present embodiments are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those skilled in the art from the description of the claims.

1 is a perspective view of a noise canceling device according to an embodiment of the present invention.
2 is an exploded view of a noise canceling device according to an embodiment of the present invention.
3 is a perspective view of a honeycomb resonator according to an embodiment of the present invention.
4 is a flowchart for removing noise according to an embodiment of the present invention.
5 is a block diagram of a noise canceling device according to an embodiment of the present invention.
6 is a conceptual diagram illustrating the speed of a sound wave according to a medium.
7 is an exploded view of a speaker driver according to an embodiment of the present invention.
8 is a cross-sectional view of a speaker driver and a resonator according to an embodiment of the present invention.
9 is an exploded view of a microphone module according to an embodiment of the present invention.
10 is a conceptual diagram of removing noise through a Fourier transform.
11 is a conceptual diagram illustrating an allowable phase difference for removing noise.
12 is an exemplary view showing the extracted Cladni pattern.
13 is an exemplary diagram illustrating a harmonic frequency and a fundamental frequency in a spectrum of continuous noise.
14 is a cross-sectional view of a microphone-built-in speaker device according to an embodiment of the present invention.
15 is a conceptual diagram for separating and removing a direct sound and an indirect sound according to an embodiment of the present invention.
16 is a flowchart for separating and removing a target frequency according to an embodiment of the present invention.
17 is an exemplary view for removing transaural noise according to an embodiment of the present invention.
18 is a flowchart for removing noise from multiple microphones according to an embodiment of the present invention.
19 is an exemplary diagram illustrating an operation state through a display device according to an embodiment of the present invention.
20 is a conceptual diagram for calculating the position of a sound source through multiple microphones according to an embodiment of the present invention.
21 is a flowchart for calculating the position of a sound source according to an embodiment of the present invention.
22 is a conceptual diagram for classifying noise processing according to a position of a microphone according to an embodiment of the present invention.
23 is a flowchart of a method for removing noise according to an embodiment of the present invention.

The present invention will be described in detail with reference to the accompanying drawings as follows. Here, repeated descriptions, well-known functions that may unnecessarily obscure the gist of the present invention, and detailed descriptions of configurations will be omitted. The embodiments of the present invention are provided in order to more completely explain the present invention to those of ordinary skill in the art. Accordingly, the shapes and sizes of elements in the drawings may be exaggerated for clearer description.

Looking at the noise, the low sound is relatively inaudible to the human ear according to the isoacoustic curve, but contains a large amount of energy and may cause discomfort when exposed for a long period of time.

Bass sound has a relatively large wavelength, so it can easily penetrate walls or structures, so it has a strong transmission power, and diffraction and interference easily occur in areas with varying densities (for example, corners of walls, doors, windows, etc.). There is a characteristic.

Therefore, the low-pitched sound is an important part for preventing noise from entering.

High-pitched sound has a relatively small wavelength, so it cannot easily penetrate walls, and radio waves can be blocked by treatment with sound absorption and barrier walls.

The treble includes many harmonic components for the bass, and high frequencies exceeding the audible frequency band may cause discomfort even though people cannot directly hear them.

The impact sound that directly generates the inter-floor noise has a transient characteristic, and its component exists over the entire frequency band.

In addition, the impact sound does not have a general overtone structure, but high energy is included in the low frequency band.

In addition, although the impact sound has a higher sound pressure than the actual listening, the low sound is heard less than the actual sound according to the characteristic of an equal loudness curve (Fletcher & Munson).

In addition, the impact sound may be amplified in the process of propagating through the medium, and in particular, may cause oscillation at a point where the medium changes, thereby causing greater noise between floors.

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

1 is a perspective view of a noise canceling device according to an embodiment of the present invention.

Referring to FIG. 1 , the noise canceling device according to an embodiment of the present invention may be formed in a single structure in which one or more sound-collecting microphone modules and one or more speaker drivers are attached to a medium.

In one embodiment, it may be attached to a ceiling, wall, or floor, and may be built into electric and electronic products such as lighting fixtures, air conditioners, and air purifiers.

In addition, one embodiment may be built into furniture such as a desk or bed, and may be installed in any place where vibration occurs, such as a vehicle.

2 is an exploded view of a noise canceling device according to an embodiment of the present invention.

2, the noise canceling device according to an embodiment of the present invention includes a top cover 201, a microphone built-in speaker device 203 (or a microphone module and a speaker driver may be separately included), a reference microphone module ( 205), honeycomb resonator 207, immersive playback speaker driver 209, side cover 211, LED panel 213, support frame 215, display and sensor 219, DSP 217, Digital Signal Processor) may be included.

At this time, the top cover 201, the side cover 211 may be formed to accommodate the internal components.

In this case, the speaker device 203 with a built-in microphone reduces the sound pickup and reproduction to one point, thereby eliminating latency and reducing processing power.

In addition, the built-in microphone speaker device 203 may include a microphone module and a speaker driver, but may be disposed to direct the same point.

In addition, the built-in microphone speaker device 203 may include a microphone module and a speaker driver, but may include an immersive speaker 209 having a different reproduction direction.

At this time, the reference microphone module 205 may serve as a reference for detecting the direction of the noise.

In this case, the reference microphone module 205 may be used as a reference signal for processing the noise canceling operation.

In this case, the honeycomb resonator 207 may serve to cancel the sound generated from the rear part of the microphone-built speaker device 203 or the speaker driver and the noise introduced through the ceiling by the principle of resonance. This will be described later with reference to FIG. 3 .

In this case, the speaker driver 209 for immersive reproduction may remove noise for indirect sounds. The indirect sound will be described later.

3 is a perspective view of a honeycomb resonator according to an embodiment of the present invention.

Referring to FIG. 3, the honeycomb resonator 207 according to an embodiment of the present invention accommodates one or more sound-collecting microphone modules and one or more speaker drivers. Low-level noise transmitted from the medium can be removed.

A noise canceling device using a plurality of speaker drivers transmits vibration to a medium through the speaker driver, and the noise reproduced from the rear of the speaker driver may be introduced into the interior space, causing another noise regardless of noise removal can do

At this time, the honeycomb resonator 207 is divided into a honeycomb structure inside, and a partition wall 301 dividing one or more honeycomb structures into one space may be formed.

More specifically, the honeycomb resonator 207 has a honeycomb structure inside and has different areas, so that noise generated from the speaker driver can be removed according to the principle of the Helmholtz resonator.

In this case, the partition wall may divide an area for the purpose of making a volume of the resonator. For example, eight honeycombs are made into one partition wall 301 to remove low tones, four honeycombs are made into one partition wall 301 to remove midrange sounds, and one partition wall 301 is used to remove high tones. It can be made of a bulkhead 301 .

This is because, according to the Helmholtz theory, a large area can remove low tones, and a small area can remove high tones.

In this case, the barrier rib 301 may form holes 305 of different sizes corresponding to the target frequency in order to optimize the target frequency.

In addition, the honeycomb resonator 207 may use a sound absorbing material therein in order to efficiently absorb noise.

In addition, the honeycomb resonator 207 may have different heights of the bottom surfaces 303 of the honeycomb structure formed therein in order to increase the noise absorption and diffuse reflection of the noise.

4 is a flowchart for removing noise according to an embodiment of the present invention.

Referring to FIG. 4 , an embodiment of the present invention may collect a noise signal through a contact microphone including a plurality of microphones and a preamplifier.

At this time, the noise signal collected by the input processor measures the level so that it operates at a specific level at the gate, and the unnecessary frequency can be deleted through the filter, and automatic gain according to the output level through the AGC (Auto Gain Cotroller) can be controlled.

At this time, the multi-channel spectrum analyzer can perform spectrum analysis of the noise signal for each channel, analyze the position and direction of the sound source through transient time difference analysis, and detect the frequency at which the impact sound occurs through spectrum analysis. .

In addition, the multi-channel spectrum analyzer may analyze a frequency and a volume exceeding a reference level, and secure a peak point of a target frequency.

In this case, the phase comparator may analyze the spectrum of the input frequency and predict in advance the shape of the wavelength that is changed when the multi-channel speaker is reproduced at the time of output.

At this time, the learning processor can analyze ADSR (Attack time, Decay time, Sustain level, Release time) for frequently occurring noise, and can learn by acquiring data in which the peak flow is identified through spectrum analysis. .

In this case, the phase processor may perform an operation to create an accurate inverse signal of the noise signal, and may set it to generate an inverse phase only at a frequency exceeding a reference. In addition, the wavelength corresponding to the peak for each frequency may be tracked, and the antiphase signal may be optimized according to the data learned by the learning processor.

In this case, the speaker controller may control wavelengths optimized for a plurality of speakers, and may extract wavelength data expected to be changed by a plurality of combinations.

5 is a block diagram of a noise canceling device according to an embodiment of the present invention.

Referring to FIG. 5 , a REF.MIC 501 is a reference microphone when a plurality of microphones are used, and may be a reference for comparing input signals of N microphones.

At this time, the REF.MIC 501 is attached to the center of the device, recognizes the direction of the noise source, can pick up a signal that is a reference for audio processing, and may be a contact microphone or a piezo microphone.

The MIC(N) 503 may be a separate microphone module or a microphone module built into a microphone-built speaker device, and may be a contact microphone or a piezo microphone.

At this time, the MIC(N) 503 may detect the direction and distance of a noise source together with the REF.MIC 501 and pick up a signal that is a reference for audio processing.

The SPECTRUM ANALYZER 505 may extract basic audio data such as a fundamental frequency, a harmonic frequency, a level, a delay, an ADSR, and a noise floor from a signal input through each microphone.

The PHASE COMPARATOR 507 is a phase comparator, and may analyze the phases of the N microphones based on the REF.MIC 501 and transmit the analyzed waveform to the position detection processor.

The DELAY COMPARATOR 509 is a signal delay comparator, which analyzes the delay values for the N microphone inputs based on the REF.MIC 501 and transmits the analyzed value to the position detection processor.

AMPLITUDE/GATE COMPARATOR (511) is an amplification gate comparator, which transmits audio level values for N microphone inputs to the position detection processor based on REF.MIC (501) and compares the dark noise levels of all microphones for processing and bi can be compared to determine a pass.

The FUNDAMENTAL/HARMONICS ANALYZER 515 is a fundamental and harmonics analyzer, detects a fundamental frequency from a frequency component analyzed as a spectrum, analyzes a harmonic frequency for the frequency, and transmits it to the Fourier transform device.

The FEEDBACK DETECTOR 519 is a feedback detector, and when a feedback signal is sensed, the sensed frequency may be transmitted to the feedback removing device with a corresponding frequency value.

The POSITION DETECTOR 513 is a position detector, and it can analyze the position of the noise source by analyzing the phase, delay, and level analyzed through a plurality of microphones, and can transmit the analyzed direction information and the corresponding value to the DSP.

The FOURIER TRANSFORM 517 is a Fourier transformer, which analyzes a fundamental frequency and analyzes a harmonic frequency, so that an inverse signal can be oscillated or used as data that can be verified through DSP.

The FEEDBACK DESTROYER 521 is a feedback canceller, and may block the feedback generation frequency sensed by the feedback detector.

COMPARATOR MODULE is a comparator module, which analyzes signals such as phase, delay, and level to track the location of the noise source, and analyzes the audio waveform to create a reverse-phase audio signal.

The DSP 523 may analyze and process a signal input through the comparator module, and may perform a complex calculation and output the noise canceling signal to a plurality of speakers.

At this time, the DSP 523 may include a control and display function using a wireless IO, and may also learn a noise canceling function through a learning processor.

The PHASE CONTROLLER 525 is a phase controller and may control the phases of N output signals output through the DSP 523 .

The AGC 527 is an automatic level controller, and may control gains of N outputs output through the DSP 523 .

The MATRIX 529 is a matrix controller, and may control a signal matrix for transmitting signals to the speaker for N output signals after phase and gain adjustments have been completed.

The SPEAKER(N) 531 is a noise canceling speaker that can remove noise for direct sound by using the final output signal through the matrix controller, and can be an exciter-type speaker that delivers vibrations directly to the medium. there is.

The POSITION DETECTOR 533 is a position detector, and may detect a point where the user is located and transmit a signal to the user position controller.

The USER POSTION COTROLLER 535 is a user position controller, and can automatically detect a position through a position sensor or generate a trasural signal in a corresponding region by using a transaural processor for an area designated by the user.

The TRANSAURAL PROCESSOR 537 is a trans-oral processor that can remove high frequencies that cannot be removed by direct sound with trans-oral. In this case, the signal output through the matrix controller may be converted into trans-oral and transmitted to the trans-oral reproduction speaker.

The TRASAURAL SPEAKER 539 is a transoral reproducing speaker that can remove room noise by using a signal input through a transoral processor, and may be a loudspeaker device.

LEARNING PROCESSOR (541) and MEMORY are learning processors and storage devices, which store frequently occurring noise as learning data, and when the same noise as the stored content occurs, the reverse phase waveform that can be completely removed can be stored and reproduced.

The WIRELESS I/O 543 is a wireless input/output device, which may be a remote control or a mobile device such as a PC or a smartphone, and transmits a user's control command or information sensed by the DSP 523 and the sensing device to the user. It can be included with any device that can be used.

The USER COTROLLER/MONITOR 547 may be a user controller and monitor, and the DISPLAY I/O 545 may be a display input/output device.

6 is a conceptual diagram illustrating the speed of a sound wave according to a medium.

In general, noise moves along a medium, and the medium vibrates air to generate noise.

At this time, the noise already radiated into the air is deformed by diffraction, reflection, interference, annihilation, etc., and thus it is difficult to remove it even if an opposite-phase signal is generated.

Therefore, it is desirable to remove the noise propagated according to the medium in the medium propagation step before being radiated into the air, and the noise propagated according to the medium is collected, and the reversed-phase signal is transmitted to the medium to remove the noise. .

However, since the speed of sound is different from the speed of propagation in air and in a specific medium, accurate noise removal cannot be achieved when an opposite-phase signal is generated based on the speed of sound.

Referring to FIG. 6, the speed of sound in air is 340 m/s, whereas the speed of sound in concrete in medium (solid) is 3040 m/s, which has a very large speed difference.

In addition, since the frequency propagated in the medium does not change, only the speed of the transmitted sound changes, as a result, the length of the wavelength is changed.

In order to cancel the noise by reproducing the reverse wavelength corresponding to the sound wave, the sound reproduced from the speaker must be in contact with the medium, not the air, so that the diaphragm can be in contact with the medium so that the sound wave can be transmitted to the medium. Exciter speakers that can be used can be used.

The speed of the sound wave corresponding to the target medium may be defined as in Equation 1 below.

In this case, p is the density of the medium (kg/m^3), B is the elastic modulus of the bulk module (N/m^2), P is the pressure, and V is the velocity.

Through Equation 1, the speed of the sound wave in the medium can be known, and the wavelength value for the frequency can be obtained through Equation 2 below.

At this time, since the speed of sound in the medium and the frequency can be known, the exact length of the wavelength can be extracted, and the phase can be reversed by applying the reverse phase to the wavelength.

Through the above-described method, the sound introduced into the noise source can be absorbed through the contact microphone, and the reversed-phase vibration is transmitted to the medium through the exciter speaker to cancel the noise.

7 is an exploded view of a speaker driver according to an embodiment of the present invention.

In order to transmit vibration to a dense medium, a speaker device having sufficient vibration energy is required.

Referring to FIG. 7 , the speaker driver according to an embodiment of the present invention may be formed to be directly attached to a medium to transmit vibration.

In this case, the vibration unit of the speaker driver that oscillates vibration uses an element having the same density as the medium, but can also amplify the vibration by using an element capable of generating vibration even at low output.

In this case, the reproduction characteristic of the speaker driver (diaphragm) may be adjusted so that high-pitched sound components are not reproduced (frequency of 1 kHz or higher), and a signal amplification unit may include a low pass filter (LPF) in order to maintain the reproduction characteristic.

Referring to FIG. 7 in more detail, the speaker driver according to an embodiment of the present invention includes a vibrating unit, a magnet, a voice coil, a voice coil fixing unit, and a fixing bracket, and is attached to a medium to generate vibration.

At this time, the voice coil generates a magnetic field according to the reversed-phase signal applied through the microphone module.

In this case, the signal may be a sound signal output to a speaker driver, and the magnet may be moved by the magnetic field.

At this time, the voice coil fixing unit accommodates each of the above components, but may fix the position of the voice coil from the outside of the voice coil.

At this time, the voice coil can prevent the position relative to the medium from being changed by the voice coil fixing part.

At this time, the voice coil is located inside the voice coil fixing part, but the position relative to the magnet may be variable.

The reason for changing the relative position is to take advantage of the fact that the sound characteristics vary according to the position of the voice coil with respect to the magnet.

In general, the voice coil and the magnet should be located at 1/2 level as the center of the voice coil, and when the voice coil and the magnet are farther apart, the output and low-pitched sound will occur, and only high-pitched tones will be heard in the end. The closer the magnets are, the higher the output and the higher the bass.

Accordingly, the speaker driver according to an embodiment of the present invention allows the position of the magnet or the voice coil to be moved in detail from the outside, so that the efficiency and sound quality can be adjusted as desired by the user.

At this time, the speaker driver according to an embodiment of the present invention is located inside the voice coil fixing part, and further includes a fixing groove for fixing the voice coil on an inner peripheral surface and a voice coil support part having a first screw thread formed on the outer peripheral surface, The voice coil may be fixed to the fixing groove, and a second screw thread corresponding to the first screw thread may be formed on an inner circumferential surface of the voice coil fixing part so that the position of the voice coil support part is changed by rotation of the voice coil fixing part.

At this time, the magnet is located inside the voice coil, but can be moved by the magnetic field.

In this case, the movement may be vertical vibration, and the vibration of the magnet may be transmitted to the vibrating unit.

At this time, the vibrating unit may transmit the vibration received by one surface in contact with the medium to the medium.

At this time, the vibrating unit may be formed in a parabolic shape and may include a microphone receiving unit recessed inside at one side contacting the medium, and the microphone module may be located in the microphone receiving unit spaced apart from the vibrating unit. .

At this time, the vibrating unit has a through hole formed in the center, the microphone module support pole is positioned to pass through the through hole, and is fixed by a rubber ring, and one end of the microphone module support pole is the microphone module or the microphone It may be fixedly coupled to the feedback blocking housing of the module.

At this time, the suspension ring may be included in order to prevent damage from being accumulated due to the five senses of vibration between the vibrating part and the magnet, and may be formed of a soft material.

At this time, the magnet may further include a support spring positioned on one surface of the magnet to return to the original position after vibration.

In this case, the support spring may be a wave spring having multiple layers.

In this case, the support spring may have different thicknesses of the wave spring having a plurality of layers, thereby increasing the reaction speed at low power and improving the problem of distortion even at high power.

For example, the wave spring according to an embodiment of the present invention may have a multi-layer structure including a layer a, a layer b, and a layer c, and the thickness of each layer may be configured to form a < b < c .

In this case, the wave spring may move only the layer a in the reproduction of a low-output sound, and the layers a, b, and c may move together in the reproduction of the high-output loud sound.

Therefore, the wave spring according to an embodiment of the present invention has a different spring restoring force depending on the output, so that even if a sound with very strong transient characteristics is instantaneously input, distortion does not occur and it can have a fast restoring force, and the damping factor is maximized can do.

In addition, the wave spring does not increase the size of the product because it can reduce the thickness by at least 1/2 compared to the existing spring, and because the restoring force is very strong, deformation does not occur even after long-term use.

In addition, the speaker driver may further include an upper cover and a lower cover to accommodate each component, and the voice coil fixing part may be used as a side cover.

At this time, in order to improve the performance of the speaker driver, an aluminum foil may be further included on the inner surface of the voice coil.

At this time, in order to prevent the position of the voice coil from being changed with respect to the medium, one end of the fixing bracket may be fixed to the voice coil fixing part and the other end may be fixed to the medium.

In addition, one end of the fixing bracket may be coupled to the upper cover and fixed to the medium.

At this time, a screw thread may be formed on one end of the inner circumferential surface of the fixing bracket, and a screw thread corresponding to the upper screw thread may be formed on the outer circumferential surface of the voice coil fixing part or the top cover, and may be coupled by screwing to each other.

In addition, the fixing bracket is formed in a cylindrical shape, one end of the outer periphery in contact with the medium may further include a contact portion extending outwardly, the contact portion may include one or more through-holes to be coupled with the medium.

8 is a cross-sectional view of a speaker driver and a resonator according to an embodiment of the present invention.

Referring to FIG. 8 , the noise canceling apparatus according to an embodiment of the present invention may further include a resonance unit 803 in order to effectively absorb and remove a frequency emitted from one surface of the speaker driver 801 .

At this time, the resonance unit 803 may be used as a housing of the speaker driver 801, a plurality of holes may be formed to simultaneously remove various frequencies, and the volumes for the holes may be formed differently. there is.

Here, f is the frequency to be canceled, c is the speed of sound, S is the area of the hole, L is the distance from the hole to the resonator, and V is the volume of the resonator.

At this time, the resonance part 803 may be formed in a multi-chamber method.

9 is an exploded view of a microphone module according to an embodiment of the present invention.

Referring to FIG. 9 , the microphone module according to an embodiment of the present invention may use a contact microphone capable of collecting only vibrations rather than a general microphone in order to collect a vibration signal of a high-density medium.

In this case, the contact microphone does not pick up sound in the air, but can pick up only frequencies vibrated by the medium.

At this time, the microphone module collects sound from the medium using the first band as a target band, and uses a high-pitched contact microphone for generating a first sound pickup signal, and a second band that is a lower frequency band than the first band as the target band. and a low-pitched contact microphone for collecting sound from the medium and generating a second sound collecting signal, and a microphone controller for generating the sound collecting signal by adding up the first and second sound collecting signals.

In this case, the first band and the second band may include a crossover band, and the sound pickup signal may correspond to the crossover band.

At this time, the treble contact microphone and the low tone contact microphone can connect the negative terminal (-) to the same ground, and each use the positive terminal (+) as an individual output to obtain a balanced audio signal (the balanced audio signal is strong against noise characteristics).

The balanced audio signal also has an effect of amplifying the entire signal.

Signals received by the high-pitched contact microphone and the low-pitched contact microphone are summed, and at this time, an overlapping crossover region becomes a substantially target band.

In this case, the crossover frequency may be adjusted by the user.

For example, the crossover frequency range can be set in DSP, and the user can adjust the crossover frequency by designating the high pass contact mic as a high pass filter (HPF) and the low frequency contact mic as an LPF (low pass filter). .

In this case, the high-pitched contact microphone has a relatively narrower area than the low-pitched contact microphone, and the high-pitched contact microphone and the low-pitched contact microphone may be stacked so that their central axes coincide with each other.

In addition, in the microphone module according to an embodiment of the present invention, in order to improve the sound collection rate of the sound transmitted from the medium, one end contacts the medium, and the vibration of the medium is transmitted to the high-pitched contact microphone through the other end. It may further include a boost plate for treble in a funnel shape.

At this time, since the boost plate for high sound is formed in a funnel shape, it is possible to amplify minute vibrations and efficiently transmit the amplified vibrations to the contact microphone for high sound.

At this time, the material of the boost plate for high sound can use a material that can amplify vibration (for example, the density is configured to have the same sound propagation speed as ABS-concrete), and through this, it is difficult to absorb the vibration. Vibration can be efficiently absorbed even in dense media.

In addition, in the microphone module according to an embodiment of the present invention, in order to improve the sound collection rate of the sound transmitted from the medium, one end contacts the medium, and the vibration of the medium is transmitted to the low-pitched contact microphone through the other end. It may further include a donut-shaped boost plate for bass.

In this case, the bass boost plate may be positioned so that the outer periphery coincides with the outer periphery of the low sound contact microphone, and the treble boost plate may be positioned in the inner through hole of the bass boost plate.

In addition, the microphone module according to an embodiment of the present invention accommodates a high-pitched contact microphone and a low-pitched contact microphone, and may further include a feedback blocking housing that is formed in a parabolic shape to improve a sound pickup rate.

At this time, the feedback blocking housing is formed in a parabolic (parabolic) shape, can amplify the sound generated in the medium, and can collect only the sound generated in a target direction.

In addition, since the feedback blocking housing is formed of a radioactive material or a magnetic shield, the feedback phenomenon can be eliminated because it is not subjected to magnetic influence by the magnet of the speaker driver, as will be described later.

In addition, the microphone module including the feedback blocking housing is a contact microphone type that is not affected by acoustic characteristics, so that noise from people or surroundings is not easily picked up.

In this case, the feedback blocking housing may include a rubber plate covering the opening in contact with the medium.

In this case, the rubber plate may be formed by using a material capable of amplifying the target frequency band of the medium, and the target frequency may be obtained by adjusting the size and thickness.

At this time, the rubber plate can improve the reactivity by placing an edge on the edge in order to efficiently amplify and collect the frequency.

In addition, the outer edge of the rubber plate may be treated in a ring shape in order to pick up the sound of the correct spot.

Through the ring shape, the microphone module 210 according to an embodiment of the present invention can block a sound coming from the outside by compression when mounted on a medium, and can be accurately attached to the medium, so that proximity (Proximity) By increasing the effect, the bass sound absorption characteristics can be increased.

At this time, the rubber plate includes one or more through-holes at regular intervals along a circular arc, and the bass boost plate is located inside the rubber plate, and may include one or more protrusions corresponding to the through-holes of the rubber plate, , the protrusion may be positioned to pass through the through hole of the rubber plate.

In this case, the high-pitched contact microphone and the low-pitched contact microphone may be at least one of a piezo microphone and a laser microphone.

10 is a conceptual diagram of removing noise through a Fourier transform.

Noise is a complex sound, with frequencies evenly distributed over the entire audible frequency band. A plurality of pure sounds may be combined to form a complex sound, but even pure sounds may be transformed into complex sounds by reflection, refraction, diffraction, delay, etc.

Pure tones can be easily removed with simple inversion processing, but complex tones are difficult to remove only with reverse phase processing.

In addition, it is easy to remove a sound of a low frequency band with a relatively long wavelength, and it is difficult to remove a sound of a high frequency band with a short wavelength.

In addition, the sound generated by the impact is temporary, has a short duration, and the sound is distributed over the entire frequency band.

At this time, the impact sound generated in the same medium has the same frequency form regardless of the strength of the impact.

At this time, it has a frequency structure similar to that of a percussion instrument, and this frequency oscillation can be described as a Chladni pattern.

For an impact sound that does not have a general harmonic structure, if the fundamental frequency is removed, other harmonics frequencies can be removed together, thereby reducing the overall noise.

In addition, noise reduction for the harmonic frequency can be improved because the Kladni pattern can be applied.

Referring to FIG. 10 , the impact sound has a wide frequency band from a low range to a high range.

At this time, since the low-pitched sound has a long wavelength and the high-pitched sound has a short wavelength, the residual wavelength corresponding to the high sound is drawn on the long wavelength in the analysis image of a general waveform.

In this case, the first fundamental frequency value may be obtained by obtaining the length of the first generated wavelength.

At this time, when a signal having an inverse wavelength corresponding to the first fundamental frequency value is generated and combined with the original signal, a value for the second impact sound in which the bass tone has disappeared can be obtained ( 1001 ).

At this time, the second fundamental frequency value for the second impact sound can be obtained in the same way, and when the wavelength corresponding to the second fundamental frequency value is applied in reverse phase and added together (1003), the A value can be obtained (1005).

The impact sound can be flattened (cancelled) by repeating the above-described method, and this method is the same as the principle of the Fourier transform.

However, although this method is effective in removing low-band frequencies, it is difficult to remove high-band frequencies with short wavelengths. A method for this will be described later.

11 is a conceptual diagram illustrating an allowable phase difference for removing noise.

In the noise cancellation method using the above-described Fourier transform method, the generated wavelength must be accurately generated with the inverse phase of the original signal for cancellation.

In addition, in order to obtain an attenuation effect of -10dB or more in the offsetting waveform having the reverse phase, the reverse phase must be made within 5% of the waveform.

Referring to Equation 4, in order to completely remove the noise, the input wave and the cancellation wave just need to have an exact opposite phase, and the phase difference between the two wavelengths needs to be λ/2.

At this time, when the offset wave to which the inverse phase is applied forms the opposite phase to the input wave, it is located within 5% of the wavelength, and may have a noise attenuation characteristic of -10dB.

That is, when it is within +/-5%, the offsetting effect becomes large.

12 is an exemplary view showing the extracted Cladni pattern.

Referring to FIG. 12 , a region from which inter-floor noise is radiated corresponds to the ceiling of a room, and assuming that the ceiling of the room is a vibrating membrane, the Claddney pattern may be applied.

At this time, the negative harmonic composition can be predicted by obtaining a pattern for interference and annihilation with the ceiling surface as a membrane.

By using the predicted value, it is possible to obtain a more accurate frequency value for a frequency not composed of harmonics, and to obtain a more accurate numerical value by constantizing the density value for the interlayer material.

Taking inter-floor noise as an example, since the ceiling has a rectangular structure, it can have multiple vibration modes, which can be explained by the Kladni pattern that can grasp the structure of harmonic generation according to the rectangular membrane.

The theory of the Kladni pattern is schematically represented by a general formula as shown in Equation 5 below.

In this case, the additionally generated frequency may be defined as 6 in the following math.

Accordingly, the noise removal method through the Claddney pattern can recognize the predicted value and the change in the pattern in advance, and can cancel the input noise by generating an inverse waveform corresponding to the pattern.

13 is an exemplary diagram illustrating a harmonic frequency 1303 and a fundamental frequency 1301 in a spectrum of continuous noise.

Continuous noise having a predetermined pattern can be removed by sensing a wavelength and simply generating an inverse signal, so it is easier to remove noise compared to an impact sound.

In addition, a continuous sound having a predetermined pattern has a lot of noise having harmonics due to oscillation, and in this case, noise due to harmonics generated in a high frequency band can be removed together by simply removing the fundamental frequency 1301 .

At this time, in the noise removal method according to an embodiment of the present invention, if the fundamental frequency and the length of the wavelength are known in the noise removal having an overtone structure, the noise can be removed by generating a corresponding waveform.

In this case, an audio signal similar to actual noise may be generated by simultaneously generating n harmonic frequencies 1301 along with the fundamental.

In this case, the noise removal method according to an embodiment of the present invention may remove the noise by performing inverse phase processing on the generated signal, which is the same as the method of using the Fourier transform inversely.

In this case, the continuous sound may be effectively removed through the audio sample (learning function).

The learning function can maximize the effect of noise reduction by collecting and sampling noise generated at a place where the noise canceling device is applied, and converting it into a database.

At this time, the learning function may collect the noise input through the microphone module as a sample and store it in the memory. This can be set to be saved while the noise canceling function is running.

In this case, the collected sample may be analyzed for level, delay, wavelength, and spectrum, and the analyzed data may be stored together with the sample.

At this time, the learning function can analyze the wavelength, level, etc. of the noise to determine the cause of the noise, and load a sample having the most similar value and apply an inverse phase to remove the noise.

The learning function is a method for supplementing the timing of samples, which is the most important in noise removal, and can recognize and process a value corresponding to the shape of a waveform as a vector value, thereby saving time for audio processing, and actually You can have a process that doesn't cause any delay.

The collected sample may be stored in a memory as well as in a management server connected to the Internet.

In this case, the sample stored in the server may be used as data from other users or other spaces using the same device.

At this time, the supplier may edit and re-upload the sample to enable the most effective noise cancellation in the noise canceling device.

14 is a cross-sectional view of a microphone-built-in speaker device according to an embodiment of the present invention.

The basic condition of the noise canceling method is to generate an inverse wave at the same position as the noise generating position to cancel it.

At this time, when the noise source is blocked by a wall or an obstacle, the point at which the noise is generated may be regarded as the wall or obstacle, and the noise may be extinguished by generating a wave in an opposite phase to the wall or the obstacle.

In this case, a speaker driver may be used as the device for reproducing sound, and a microphone module for collecting noise is also required.

In this case, the microphone module collects the noise, performs reverse-phase processing on the received signal, and transmits it to the speaker driver, and the speaker driver outputs the reverse-phase-processed signal to cancel the noise.

However, the sound reproduced through the speaker driver may be picked up by the microphone module, and the picked-up sound may be amplified through the amplifier and output again by the speaker driver.

This is called howling or feedback, and the feedback can easily occur when the microphone module is located on-axis with respect to the voice coil of the speaker driver.

Therefore, in general, the speaker driver and the microphone module cannot be installed in the same location.

In addition to this, the feedback continuously increases the amplification amount of the amplifier, thereby causing damage to the amplifier circuit, the power circuit, and the speaker driver.

The feedback is easily generated at a specific frequency, and as the bandwidth of a quality factor (Q) value is widened, it may affect the entire frequency band.

Accordingly, there is a need for a method for preventing the feedback by using a graphic equalizer (EQ) to block the frequency at which the feedback occurs.

However, the above-described method has a problem in that it is difficult to use in a noise-removing environment because it cannot accurately process an inverse phase of an input frequency by modifying the characteristics of the frequency.

In addition, if the microphone module and speaker driver are installed in separate locations to prevent feedback, it is not possible to accurately pick up the noise to be removed through the microphone module, but also requires correction through DSP (Digital Signal Processor). There are difficulties in doing

The feedback occurs because both the speaker driver and the microphone module have a certain directivity, and it is advantageous to position the speaker driver and the microphone module in an Off-Axis state in order to prevent the feedback.

However, as described above, the Off-Axis state cannot pick up an accurate sound, so a method is needed to position it in the On-Axis state, but to prevent feedback.

Accordingly, one embodiment of the present invention can prevent feedback by using a cover capable of blocking a magnetic field as a cover of the microphone module in order to prevent feedback even on the on-axis, and positioning the microphone module inside the diaphragm of the speaker driver. .

Referring to FIG. 14 , the speaker device with a built-in microphone according to an embodiment of the present invention includes a microphone module 1401 that generates a sound pickup signal by collecting sound from a medium, and applies a vibration corresponding to an inverse signal of the sound pickup signal to the medium. a controller that receives the sound pickup signal from the speaker driver 1403 and the microphone module 1401 that delivers, generates an inverse signal of the sound pickup signal, and transmits the inverse signal to the speaker driver 1403 there is.

In this case, the microphone module 1401 may include a high-pitched contact microphone, a low-pitched contact microphone, a high-pitched boost plate, a low-pitched boost plate, a rubber plate, and a feedback blocking housing.

In this case, the speaker driver 1403 may include a magnet, a voice coil, a vibrator and a fixing bracket.

In this case, in one embodiment of the present invention, the speaker driver 1403 may be mounted at the same position as the position of the microphone module 1401 in order to efficiently remove spatial noise.

The reason for mounting the speaker driver 1403 in the same position as the microphone module 1401 is to reduce the processing power for correcting a phase difference that may occur when the pickup position and the playback position are the same by making the pickup position and the playback position the same. , to minimize the error.

To this end, in an embodiment of the present invention, the microphone module 1401 is disposed inside the speaker driver 1403 for generating vibration, and a feedback prevention structure is applied to pick up a noise signal and reproduce an inverse signal at the same time. One possible device can be presented.

In more detail, the vibrating unit of the speaker driver 1403 may have a parabolic shape, and a microphone module 1401 capable of picking up high and low tones separately may be mounted therein.

In this case, the microphone module 1401 may be structurally separated so as not to be affected by the vibration of the speaker driver 1403 .

More specifically, the vibrating unit of the speaker driver 1403 may include a microphone accommodating unit recessed inwardly on one side contacting the medium, and the microphone module 1401 is spaced apart from the vibrating unit in the microphone accommodating unit. can be located.

In addition, to prevent the vibration of the speaker driver 1403 from being transmitted to the microphone module 1401 , one end is fixedly coupled to the microphone module 1401 , and the middle end is coupled to the speaker driver 1403 for supporting the microphone module Further comprising a pole, a rubber ring may be interposed between the pole for supporting the microphone module and the speaker driver 1403 .

Through the above-described structure, in one embodiment of the present invention, the microphone module 1401 and the speaker driver 1403 are accurately seated on the medium, so that pickup and playback can be arranged and operated completely independently.

In addition, the microphone module support pole may allow the microphone module 1401 to be accurately adsorbed to the medium, and may be firmly attached to the target medium by using an adhesive on the rubber plate.

At this time, the feedback blocking housing accommodates a high-pitched contact microphone and a low-pitched contact microphone, but may be formed of an anti-magnetic material (anti-magnetic) in order to prevent the influence of external magnetism, and is formed in a parabolic shape to improve the sound pickup rate can be

In addition, an embodiment of the present invention may include an integrated terminal for connecting the speaker driver 1403 and the microphone module 1401 .

In this case, the microphone module 1401 may use piezo diaphragms of different sizes, such as a high-pitched contact microphone and a low-pitched contact microphone, to pick up a center frequency to be collected differently.

Through this, it is possible to broaden the range of the target sound pickup frequency band.

In addition, in one embodiment of the present invention, the microphone module 1401 for sound pickup is mounted at the same location where the speaker driver 1403 is mounted, but the microphone module 1401 uses a contact microphone (eg, a piezo microphone) to It is possible to prevent feedback by collecting only the noise on the close contact surface without collecting noise in the air.

In addition, as described above, the microphone module 1401 includes a magnetic feedback blocking housing to prevent the magnetic field of the speaker driver 1403 from being affected, thereby preventing feedback due to the magnetic field.

At this time, the vibrating unit is connected to the magnet, and may transmit vibration to the medium through the vibrating unit by driving of a moving magnetic method.

At this time, the vibration of the vibrator may not have any effect on the microphone module 1401 by the rubber ring.

In this case, the fixing bracket may be connected to the voice coil fixing part of the speaker driver 1403 or the external housing of the speaker driver 1403 to be fixed to the medium.

At this time, the microphone module 1401 and the vibrator may be mounted horizontally on the medium by means of a fixing bracket.

15 is a conceptual diagram for separating and removing a direct sound and an indirect sound according to an embodiment of the present invention.

Referring to FIG. 15 , in the noise removal method according to an embodiment of the present invention, a plurality of microphones are installed at a predetermined distance in an array form, and when sound is collected through the microphone, the direction of the sound source and the distance between the sound source can be calculated. there is.

At this time, when the noise generation position and the speaker reproduction position are the same, it can be simply canceled through the reverse phase. can

Therefore, the position detection of the noise source makes it possible to accurately create an inverse wavelength by adding distance and direction components.

In addition, the noise removal method according to an embodiment of the present invention can make different wavelengths separated for each frequency by using a plurality of speakers, and can generate different delay values and different wavelengths for each speaker, thereby generating sophisticated noise. Removal is possible.

At this time, if the ceiling 1500 is taken as an example, the impact sound 1503 generated from the sound source 1501 may be directly transmitted to the noise canceling device, or may be transmitted after being reflected 1505 .

In this case, the direct sound 1503 may be removed by the speaker driver 1509 closest to the sound source, and the indirect sound 1505 or reflected sound may be removed by the remaining speaker drivers 1511 .

In more detail, in the noise removal method, a reference microphone module, which is a standard for sound collection, is placed in the center of the device, and n microphone modules may be disposed at a predetermined distance.

At this time, the reference microphone module and the n microphone modules can detect the location and distance of noise by calculating the separated values based on the phase, delay, and level after analyzing the waveform in detail through the spectrum analyzer.

At this time, the speaker driver reproduced in order to remove the noise generated at the detected position can be divided and reproduced into n speakers by forming a matrix to block the sound at the point of impact using the phase control and automatic gain controller.

At this time, since the noise removal method according to an embodiment of the present invention can calculate the position and direction values of the noise source, it is possible to know the value of the noise introduced by reflection or deformation due to oscillation, diffraction, interference, etc., so that additional processing is easier lose

In addition, since the indirect sound has a delay compared to the direct sound, the difference between the direct sound and the indirect sound can be distinguished, and the noise removal method may be applied differently accordingly, thereby increasing the noise removal efficiency.

16 is a flowchart for separating and removing a target frequency according to an embodiment of the present invention.

Referring to FIG. 16 , in the noise removal method according to an embodiment of the present invention, a sound pickup signal is first received through a microphone.

In this case, the GATE may determine the operating point by applying the average value of the ambient noise.

At this time, the LOUDNESS LEVEL DETECTOR/COMPARISON can adjust the automatic gain control by comparing the sound pickup signal input through the microphone preamplifier with the output of the power amplifier.

In this case, the filter may select only an effective frequency band targeted for processing, and a low pass filter (LPF) and a high pass filter (HPF) in which a specific frequency value is specified may be applied.

In this case, the CROSSOVER may separate frequencies according to the purpose of processing, and a Band Pass Filter (BPF) may be applied.

At this time, BAND 1, BAND 2, BAND N PHASE PROCESSOR sets the center frequency according to the target rejection frequency, corrects the phase for each center frequency using a delay, and divides it into N according to need and accuracy. may be

In this case, BAND 1, BAND 2, and BAND N TRIMMER may reset the level compensation according to each center frequency phase correction, and may be subdivided into N pieces according to necessity and accuracy.

At this time, the summer can re-sum the reverse-phase-processed signals, and noise can be removed through the POWER AMPLIFIER and SPEAKER DRIVER.

At this time, the TRANSIENT DETECTOR/FEEDBACK DESTROYER processes whether the input signal has transient characteristics or feedback, recognizes a sudden change in signal as having a transient characteristic, and considers a gradually increasing signal as feedback and gating the signal Ingredients can be deleted.

Also, the TRANSIENT DETECTOR/FEEDBACK DESTROYER can discriminate the signal type using ADSR characteristics. This allows you to check the frequency at which the feedback occurs in advance, allowing you to completely control the feedback.

17 is an exemplary view for removing transaural noise according to an embodiment of the present invention.

In the noise removal method according to an embodiment of the present invention, it may be difficult to completely remove noise because an impact sound may be introduced from outside the medium to which the noise removal device is attached.

Although the high sound can be partially removed by using a honeycomb resonator as described above, noise may be introduced from a direction to which the noise canceling device is not attached.

Accordingly, transoral reproduction can be performed through the speaker driver mounted in the opposite direction to the medium to which the noise canceling device is attached.

In this case, noise generated in a space other than a medium may be predicted in advance through a separate pickup microphone, and the noise canceling device may generate a transoral signal corresponding to the reversed phase to remove noise for a specific space.

At this time, the noise canceling device may analyze the signal input through the reference microphone and the N piezo microphones to simulate the progress of the waveform based on the virtual listening point.

In this case, the position of the listener may be determined using a human body sensor, and the detected data may be transmitted to the position controller.

In this case, the position of the listener may be an area in which noise is to be removed using transoral.

Data that has been processed through DSP can be transmitted to the front speaker through the transoral processor.

At this time, a transoral signal for noise removal based on the listening point is generated, and the noise at the listening point may be canceled.

However, in this case, the degree of noise removal at the listening point can be controlled by the user through a wired or wireless device.

In this case, a listening point when a plurality of listeners exist may be designated as an appropriate point through a wired/wireless device.

Referring to FIG. 17 , in the noise removal method, when the distance from the noise generating point 1701 to the listening point is designated as L, it is possible to predict sound waves transmitted to L in a general situation. In addition to this, by adding factors such as the amount of noise introduced through another medium or wall, delay, and wavelength, it is possible to predict how a sound wave will be formed at the listening point.

As a result, the noise removal method can remove the noise by using the predicted value as a parameter of the transoral processor.

In this case, the transoral speaker 1703 may each be configured with one or more speaker arrays, and may generate a transoral sound image based on a listening point.

In addition, a plurality of transoral speaker sets may be mounted on the device in order to expand an area to which a transoral sound image is applied.

18 is a flowchart for removing noise from multiple microphones according to an embodiment of the present invention.

Referring to FIG. 18 , one or more multiple microphones may be configured according to a microphone method for acquiring a stereophonic sound.

For example, BIBAURAL can be configured with 2, AMBISONIC with 4 or more, and Multiple XY can be configured with 8 or more.

At this time, the sound pickup signal passing through the MIC PRE and AD may pass through a high pass filter (HPF) that removes frequencies below 1 kHz and may be transmitted to the mixer.

At this time, the MIXER can process the sound pickup signal input through the bus through PANNING, LEVEL, etc.

At this time, the DSP may perform audio DSP processing through COMPRESSOR, GATE, EQ, and the like.

In this case, the BINAURAL/AMBISONIC encoder may monitor the final processed audio and apply encoding as a stereophonic sound acquisition method.

In this case, the PHASE ANALYZER can accurately match the phase by analyzing the phase between the encoded data and the transoral decoding.

At this time, the PHASE CONTROLLER can apply the reverse phase through the analyzed analyzer.

In this case, the signal processed as binaural or ambisonic may be reproduced transaurally through TRANSAURAL DECODER, POWER AMP, and SPEAKER.

19 is an exemplary diagram illustrating an operation state through a display device according to an embodiment of the present invention.

When the noise canceling device generates inter-floor noise, the noise is not transmitted to the user, but it can be checked whether the device is operating.

In addition, the operating state of the device is recorded for each time and date, so that the frequency of noise between floors can be checked, and this data can be used to be customized to the user's environment.

Referring to FIG. 19 , the noise canceling apparatus includes a display 1901 that is divided into multiple, and through the divided display 1901, whether direct sound is processed, whether or not an indirect sound is processed, etc. may be displayed.

At this time, the direct sound processing monitor is the inner circle 1905, and may express the amount to which noise removal is applied based on the center, and may have a processing direction.

At this time, the monitor that processes the indirect sound is an external circle 1903, can express the amount of noise removal from the outside to the center, and can recognize the direction of the process.

The display method is not limited to the above-described example, and a color and whether the display is lit may be changed according to a user's setting.

At this time, the operation state of the noise canceling device is recorded according to time, the user can monitor the operation state using a display or a smart phone, and the record can be checked in units of time, day, month, year, etc. there is.

In this case, the record may be used as data to increase the quality of operation of the device. For example, if the maximum and minimum values of occurrence are compared and the average value is known, it can be used as threshold and gating parameters that can maintain the operation of the device more precisely.

In addition, the record may be used as evidence for a noise dispute between floors.

In addition, the noise canceling device may be used as data of a power management system capable of stopping an operation during a time when a user is not in residence or continuously operating during a time when noise between floors is concentrated.

20 is a conceptual diagram for calculating the position of the sound source 2000 through the multi-microphone 2001 according to an embodiment of the present invention, and FIG. 21 is a method for calculating the position of the sound source 2000 according to an embodiment of the present invention It is a flow chart.

20 and 21 , according to an embodiment of the present invention, a direction in which a sound is generated may be calculated by analyzing a level between the microphones using two or more microphones 2001 .

Also, according to an embodiment of the present invention, a distance at which a sound is generated may be calculated by analyzing a time difference between the microphones using two or more microphones 2001 .

At this time, if one embodiment of the present invention is limited to the sound generated in a specific medium (ceiling, etc.), a more accurate operation can be made.

22 is a conceptual diagram for classifying noise processing according to a position of a microphone according to an embodiment of the present invention.

Referring to FIG. 22 , according to the direction and distance of the sound source, the reverse image of the corresponding waveform may be output through the speaker driver 2203 closest to the sound source 2201 .

In this case, interference may occur in the B zone due to a calculation error of the distance or wavelength from the generated sound source.

Accordingly, at this time, the remaining speaker drivers except for the speaker driver closest to the sound source may output an audio signal to which a phase change is applied to suppress interference, thereby preventing noise from being introduced only to the A zone.

Accordingly, in the noise canceling method according to an embodiment of the present invention, a more flattened waveform can be created despite sound diffusion using a plurality of speakers.

23 is a flowchart of a method for removing noise according to an embodiment of the present invention.

Referring to FIG. 23 , in the method for removing noise according to an embodiment of the present invention, first, in the method of removing noise through a noise removing device, a sound is collected from a medium through a microphone module for sound pickup to generate a noise-collecting signal ( S2310).

In addition, the noise removal method according to an embodiment of the present invention generates a noise removal signal based on the noise pickup signal (S2320).

In addition, in the noise removal method according to an embodiment of the present invention, a vibration corresponding to the noise removal signal is transmitted to the medium through a speaker driver (S2330).

At this time, a plurality of the microphone modules for sound pickup are attached to the medium, and the step of generating the noise removal signal (S2320) corresponds to the noise using the noise pickup signals collected by the plurality of microphone modules for sound pickup. The direction may be sensed, and the noise removal signal may be generated based on the direction.

In this case, step S2320 may include calculating the position of the sound source corresponding to the noise based on the noise collection signals and generating the noise removal signal based on the position of the sound source. .

The speaker driver is provided in plurality, and calculating distances between each of the plurality of speaker drivers and the sound source and delay corresponding to at least one of the distances corresponding to at least one of the plurality of speaker drivers The method may further include applying to the noise canceling signal.

In this case, the step S2330 includes the steps of transmitting a vibration corresponding to the noise removal signal for removing noise corresponding to the sound source to the medium through some of the plurality of speaker drivers and the plurality of speaker drivers Another part of these may include transmitting a damped vibration to the medium to attenuate the vibration.

In this case, the method may further include removing sound leakage and noise generated at the rear side of the speaker driver through a honeycomb resonator accommodating the sound pickup microphone module and the speaker driver.

At this time, the honeycomb resonator, the inside is divided into a honeycomb structure, the partition wall dividing one or more honeycomb structures into one space may be formed.

In this case, the honeycomb resonator may have different heights of the bottom surfaces of the honeycomb structures formed therein in order to increase diffuse reflection of noise absorbed therein.

In this case, the barrier rib may have a through hole having a size corresponding to the frequency to be removed in the space formed by the barrier rib.

At this time, the step S2320 includes the steps of calculating a first fundamental frequency value based on the noise pickup signal, generating a first noise removal signal corresponding to the first fundamental frequency value, Calculating a second fundamental frequency value based on a noise pickup signal from which a wavelength corresponding to the first fundamental frequency value has been removed, and generating a second noise removal signal corresponding to the second fundamental frequency value. may include

In this case, in step S2330, vibrations corresponding to the first noise removing signal and the second noise removing signal may be sequentially transmitted to the medium through the speaker driver.

At this time, the step S2320 is a step of predicting a Chladni pattern according to the structure information of the medium input by the user and generating the noise removal signal based on the pattern and the noise collection signal may include steps.

In this case, the step S2320 is a step of calculating a fundamental frequency value and a harmonics frequency value based on the noise pickup signal, and generating a waveform corresponding to the fundamental frequency value and the harmonic frequency value. It may include simultaneously generating and generating the noise removal signal based on the waveform.

The noise canceling device according to an embodiment of the present invention can be used by being attached to a wall or ceiling supporting a building, and thus can relatively easily detect the shaking of the building.

Accordingly, the noise canceling apparatus according to an embodiment of the present invention may provide a function of notifying the user when an earthquake occurs by adding a sensor capable of detecting an earthquake.

At this time, when abnormal vibration such as an earthquake is detected, the noise removal device according to an embodiment of the present invention may sound an alarm and turn on an emergency sensor, and may reset the operation according to the user's preferences when the situation ends.

In addition, the noise removal device according to an embodiment of the present invention may include one or more temperature, humidity, oxygen density, fine dust concentration, fire detection, and gas detection sensor, through which, when an abnormal situation occurs, the user can be visually or auditory, etc. An alarm is issued so that the user can recognize it, and in an emergency such as a fire, the emergency situation can be delivered directly to the fire department by interlocking with the fire fighting device.

As described above, in the noise removal apparatus and method according to the present invention, the configuration and method of the above-described embodiments are not limitedly applicable, but the embodiments are all or all of the embodiments so that various modifications can be made. Some of them may be selectively combined and configured.

203: speaker unit with built-in microphone
205: reference microphone module
207: Honeycomb Resonator
209 : Speaker driver for immersive playback

Claims (26)

one or more sound-collecting microphone modules for generating a noise-collecting signal by collecting sound from a medium;
one or more speaker drivers for transmitting a vibration corresponding to a noise removal signal generated based on the noise collection signal to the medium;
a controller for generating the noise removal signal based on the noise collection signal; and
A honeycomb resonator accommodating the one or more sound-collecting microphone modules and the one or more speaker drivers, and removing sound leakage occurring at the rear of the speaker driver and low-level noise transmitted from the medium,
The honeycomb resonator is
The interior is divided into a honeycomb structure, and a partition wall dividing one or more honeycomb structures into one space is formed. The method according to claim 1,
The sound-collecting microphone module,
A plurality of attachments to the medium,
A noise canceling apparatus, characterized in that a direction corresponding to the noise is detected using the noise-collecting signals collected by the plurality of sound-collecting microphone modules, and the noise canceling signal is generated based on the direction. 3. The method according to claim 2,
The noise pickup signals are
Used to calculate the position of the sound source corresponding to the noise, the noise removal device, characterized in that the noise removal signal is generated based on the position of the sound source. 4. The method according to claim 3,
The speaker driver is
provided in plurality, distances between each of the plurality of speaker drivers and the sound source are calculated, and a delay corresponding to at least one of the distances is applied to a noise canceling signal corresponding to at least one of the plurality of speaker drivers Features a noise canceling device. 5. The method according to claim 4,
Some of the plurality of speaker drivers generate the noise removal signal for removing noise corresponding to the sound source, and other portions of the plurality of speaker drivers are damped vibrations for attenuating vibration for removing the noise. Noise canceling device, characterized in that for generating. 6. The method of claim 5,
The plurality of sound-collecting microphone modules and the plurality of speaker drivers are installed in one structure attached to the medium. delete delete The method according to claim 1,
The honeycomb resonator is
In order to increase the sound leakage absorbed inside and the diffuse reflection of the noise, the noise canceling device, characterized in that the height of the bottom surface of each honeycomb structure formed therein is formed to be different. The method according to claim 1,
The partition wall is
A noise removing device, characterized in that a through hole having a size corresponding to the frequency to be removed is formed in the space formed by the partition wall. The method according to claim 1,
The speaker driver is
a resonance unit coupled to the rear surface of the speaker driver and formed in a multi-chamber manner in order to cancel the sound leakage generated from the rear surface of the speaker driver; Noise canceling device, characterized in that it further comprises. 4. The method according to claim 3,
The controller is
Based on the location of the sound source and the noise-collecting signal, a first fundamental frequency value is calculated, and a first noise removal signal corresponding to the first fundamental frequency value is generated and transmitted to the speaker driver do,
a second fundamental frequency value is calculated based on the noise pickup signal from which a wavelength corresponding to the first fundamental frequency value is removed, and a second noise removal signal corresponding to the second fundamental frequency value is generated, passed to the speaker driver,
The speaker driver is
The noise removal device, characterized in that the vibrations corresponding to the first noise removal signal and the second noise removal signal received by the controller are transmitted to the medium in time series order. The method according to claim 1,
The controller is
Predicting a Chladni pattern according to the structure information of the medium input by a user, and generating the noise removal signal based on the pattern and the noise collection signal. 4. The method according to claim 3,
The controller is
Based on the location of the sound source and the noise pickup signal, a fundamental frequency value and a harmonics frequency value are calculated, and a waveform corresponding to the fundamental frequency value and the harmonic frequency value is simultaneously generated, Noise canceling device, characterized in that generating the noise canceling signal based on the simultaneously generated waveform. A method of removing noise through a noise canceling device, the method comprising:
generating a noise-collecting signal by collecting sound from a medium through a sound-collecting microphone module;
generating a noise removal signal based on the noise pickup signal;
transmitting a vibration corresponding to the noise removal signal to the medium through a speaker driver; and
Through a honeycomb resonator accommodating the sound-collecting microphone module and the speaker driver, removing the sound leakage and noise occurring at the rear side of the speaker driver,
The honeycomb resonator is
The interior is divided into a honeycomb structure, and a partition wall dividing one or more honeycomb structures into one space is formed. 16. The method of claim 15,
The sound-collecting microphone module,
A plurality of attachments to the medium,
The step of generating the noise removal signal comprises:
A noise removal method, comprising: detecting a direction corresponding to the noise by using the noise-collecting signals collected by the plurality of sound-collecting microphone modules, and generating the noise-removing signal based on the direction. 17. The method of claim 16,
The generating of the noise removal signal comprises:
calculating a location of a sound source corresponding to the noise based on the noise pickup signals; and
generating the noise removal signal based on the location of the sound source; Noise removal method comprising a. 18. The method of claim 17,
The speaker driver is
provided in plurality,
calculating distances between each of the plurality of speaker drivers and the sound source; and
applying a delay corresponding to at least one of the distances to a noise canceling signal corresponding to at least one of the plurality of speaker drivers; Noise removal method, characterized in that it further comprises. 19. The method of claim 18,
Transmitting the vibration to the medium comprises:
transmitting a vibration corresponding to the noise removal signal for removing noise corresponding to the sound source to the medium through some of the plurality of speaker drivers; and
transmitting, by another part of the plurality of speaker drivers, attenuated vibration for attenuating the vibration to the medium; Noise removal method comprising a. delete delete 16. The method of claim 15,
The honeycomb resonator is
In order to increase the diffuse reflection of the noise absorbed therein, the noise reduction method, characterized in that the height of the bottom surface of each honeycomb structure formed therein is formed to be different. 16. The method of claim 15,
The partition wall is
A noise removal method, characterized in that a through hole having a size corresponding to the frequency to be removed is formed in the space formed by the partition wall. 16. The method of claim 15,
The generating of the noise removal signal comprises:
calculating a first fundamental frequency value based on the noise pickup signal;
generating a first noise canceling signal corresponding to the first fundamental frequency value;
calculating a second fundamental frequency value based on a noise pickup signal from which a wavelength corresponding to the first fundamental frequency value has been removed; and
generating a second noise removal signal corresponding to the second fundamental frequency value; including,
Transmitting the vibration to the medium comprises:
and sequentially transmitting vibrations corresponding to the first noise removing signal and the second noise removing signal to the medium through the speaker driver. 16. The method of claim 15,
The generating of the noise removal signal comprises:
predicting a Chladni pattern according to the structure information of the medium input by a user; and
generating the noise removal signal based on the pattern and the noise collection signal; Noise removal method comprising a. 16. The method of claim 15,
The step of generating the noise removal signal comprises:
calculating a fundamental frequency value and a harmonics frequency value based on the noise pickup signal;
simultaneously generating a waveform corresponding to the fundamental frequency value and the harmonic frequency value; and
generating the noise canceling signal based on the waveform; Noise removal method comprising a.

Images