CN108088064B - Active noise reduction device and control method provided at ventilation duct and capable of acoustic interaction - Google Patents

Abstract

The invention relates to an active noise reduction device arranged at a ventilation pipe opening and capable of acoustic interaction, comprising a housing, a loudspeaker, a reference microphone, a plurality of error microphones, a controller and a diffuser; the loudspeaker Located in the housing; the reference microphone is located on the top of the housing and the acoustic input end of the reference microphone faces the outside of the housing; each of the error microphones is located on the bottom surface of the housing And it is arranged near the edge of the bottom surface; the controller is arranged in the housing, and the speaker, the reference microphone and each error microphone are connected to the controller by lead wires; the center of the diffuser is opened A third through hole, the bottom surface of the housing is disposed in the third through hole. The invention also provides an active noise reduction control method capable of acoustic interaction. The invention has simple structure and small volume, and can realize acoustic interaction through an active noise reduction device while actively reducing noise.

Description

Active noise reduction device and control method provided at ventilation pipe opening and capable of acoustic interaction

Technical Field

The present invention relates to the technical field of noise control, and in particular to an active noise reduction device which is arranged at a ventilation pipe opening and capable of acoustic interaction, and a control method thereof.

Background Art

With the progress of modern society and the improvement of people's living standards, the problem of noise pollution has become increasingly prominent and has seriously affected people's daily lives. With the widespread application of noise control technology, the noise from the air inlet and outlet of central air conditioners and fresh air systems has become one of the main noise sources in indoor environments. The low-frequency component of this noise is more prominent and it is difficult to effectively suppress it through passive noise reduction methods in a limited space.

Active noise control is an active noise control technology based on the principle of coherent superposition of sound waves. Its basic principle is to introduce a secondary sound source into the sound field and use it to emit "anti-noise" with the same amplitude as the original noise but opposite phase, so that the original noise can be coherently offset, thereby achieving the purpose of noise suppression. Active noise control technology is small in size and light in weight, and has significant advantages in controlling low-frequency noise. The invention patent "An Active and Passive Composite Muffler for Ventilation Pipes" with announcement number CN104165255B is mainly optimized for high-speed ventilation ducts (such as ships). It has a complex structure and a large volume, and is difficult to use in central air-conditioning, fresh air systems and other ducts in daily indoor environments. At the same time, the speakers and microphones in this technical solution are placed inside the pipe, which can only achieve noise reduction and do not have acoustic interaction functions.

With the popularization of devices such as smart speakers and the widespread application of voice interaction, the demand for acoustic interaction functions will become increasingly urgent. In order to suppress low-frequency noise, the active noise control system must have acoustic input and output devices (such as microphones and speakers) that can pick up acoustic signals in the target area (such as human voice) and emit sounds (such as music) in the target area, which provides the necessary conditions for realizing acoustic interaction functions at the same time. However, active noise reduction and acoustic interaction interfere with each other: the microphone of the active noise reduction system is generally placed in a position where low-frequency noise is more prominent, which causes greater interference to the extraction of acoustic signals such as voice in the target area; in addition, when the active noise reduction system's speakers are used to play music and other acoustic signals directly, it will cause strong interference to the adaptive noise reduction algorithm, seriously affecting the normal operation of the system. There is still a lack of effective methods to organically combine active noise reduction and acoustic interaction.

Summary of the invention

In order to solve the above technical problems, the present invention provides an active noise reduction device and a control method which are arranged at a ventilation pipe opening and capable of acoustic interaction, and which can realize acoustic interaction while realizing active noise reduction.

Technical solution 1:

An active noise reduction device arranged at a ventilation duct opening and capable of acoustic interaction comprises a shell, a loudspeaker, a reference microphone, a plurality of error microphones, a controller and a diffuser; a first through hole is provided on the bottom surface of the shell, and a second through hole is provided on the top of the shell; the loudspeaker is arranged in the shell, and its paper cone faces outward and is arranged on the first through hole, and the inner space enclosed by the shell and the loudspeaker forms a back cavity of the loudspeaker; the reference microphone is arranged on the second through hole and the acoustic input end of the reference microphone faces the outside of the shell; each of the error microphones is arranged at the The shell body is provided with a reference microphone and an acoustic input end of each error microphone facing the outside of the shell body; the controller is provided in the shell body, and the speaker, the reference microphone and each error microphone are connected to the controller by leads; a third through hole is provided at the center of the diffuser, and the bottom surface of the shell body is provided in the third through hole. When the diffuser is installed on the ventilation pipe mouth, the side of the shell body provided with the reference microphone is used as the inner side and is provided inside the ventilation pipe mouth, and the bottom surface of the shell body is used as the outer side and is provided at the ventilation pipe mouth, so that the speaker can directly emit sound into the room.

More preferably, the controller includes a wireless communication module and an A/D converter, a digital signal processor, a D/A converter and a power amplifier connected in sequence, the external sound control device performs two-way communication with the digital signal processor through the wireless communication module, the A/D converter collects output signals of the reference microphone and each of the error microphones, the D/A converter outputs the signal to be played by the speaker to the power amplifier, and the power amplifier is connected to the speaker.

More preferably, the active noise reduction device further comprises a sound absorbing material layer. Except for the bottom surface of the shell, the outer surface of the shell is provided with the sound absorbing material layer, including the outer side of the reference microphone. The sound absorbing material layer is used to increase the passive noise reduction effect and reduce the wind noise of the reference microphone.

More preferably, the shell is a conical shell, the second through hole is opened at the top of the conical shell, the reference microphone is sealingly mounted on the second through hole, and the sealing connection method includes gluing or adding a cushion.

Technical solution 2:

An active noise reduction control method capable of acoustic interaction, wherein the reference microphone picks up an acoustic signal in a ventilation duct, which includes an original noise signal in the duct, an anti-phase noise signal emitted by a speaker, and an audio signal input from an external sound control device played by the speaker; the acoustic signal picked up by the error microphone includes: a residual noise signal formed by superposition of the original noise signal propagated from the inside of the duct and the anti-phase noise signal emitted by the speaker, an audio signal input from an external sound control device played by the speaker, and a control voice emitted by a person; the digital signal processor is provided with software modules for processing acoustic signals that run simultaneously, including a noise reduction module, an echo cancellation module, and a noise reduction enhancement module; in the process of processing noise reduction and acoustic interaction, the noise reduction module, the echo cancellation module, and the noise reduction enhancement module run simultaneously, regardless of the order;

The control method is specifically as follows:

The low-frequency radiation noise generated in the ventilation duct is suppressed by the noise reduction module: the noise reduction module includes a noise reduction filter, an acoustic feedback channel model, a secondary channel model and a filter update algorithm module; the noise reduction filter filters the original noise inside the duct and outputs an anti-phase noise with the same amplitude but opposite phase as the original noise, and the anti-phase noise is superimposed with the audio signal input by the external sound control device as the output signal of the noise reduction module to drive the speaker and also as the input of the acoustic feedback channel model; the output signal of the noise reduction module is inverted after being filtered by the acoustic feedback channel model, and then superimposed with the acoustic signal picked up by the reference microphone, and the anti-phase noise signal picked up by the reference microphone and the audio signal input by the external sound control device played by the speaker are eliminated after superposition, leaving only the original noise inside the duct, and the original noise is used as the input of the noise reduction filter, and at the same time, the original noise is also used as the input of the secondary channel model, and the original noise is filtered by the secondary channel model and used as the input of the filter update algorithm module;

The control voice emitted by a person is obtained through the echo cancellation module: the echo cancellation module includes a plurality of submodules, the submodules are arranged in a one-to-one correspondence with the error microphones, and each submodule includes an echo cancellation filter and an adaptive algorithm module; the input of the echo cancellation filter is an audio signal input by an external voice control device, which is filtered and inverted, and then superimposed with the sound signal picked up by the error microphone, and the superimposed signal is used as the output signal of the submodule. The output signal of the submodule removes the audio signal input by the external voice control device played by the speaker, but includes the control voice of the person, and the output signal of the submodule is sent to the external voice control device through the wireless communication module. The device realizes acoustic interaction in combination with the audio signal input by the external voice control device through the wireless communication module; at the same time, the output signal of the submodule is also used as the input of the adaptive algorithm module and the input of the filter update algorithm module, the input of the adaptive algorithm module also includes the audio signal input by the external voice control device, the adaptive algorithm module outputs the parameters for updating the echo cancellation filter, and the adaptive algorithm module adaptively adjusts the parameters of the echo cancellation filter according to the audio signal input by the external voice control device and the output signal of the submodule, the purpose of which is to better eliminate the audio signal input by the external voice control device and played by the speaker picked up by the error microphone;

The input of the filter update algorithm module includes the output signal of the secondary channel model and the superposition of the output signals of each echo cancellation submodule. The output of the filter update algorithm module is used to adaptively adjust the coefficients of the noise reduction filter, the purpose of which is to better suppress the residual noise signal picked up by the error microphone, thereby achieving a better active noise reduction effect.

The input of the noise reduction enhancement module is the parameters of each echo cancellation filter. The noise reduction enhancement module superimposes the parameters of each echo cancellation filter and smoothes them in time domain, and uses the result as the new parameter of the secondary channel model to achieve real-time update of the secondary channel model.

More preferably, the noise reduction filter, the acoustic feedback channel model, the secondary channel model and the echo cancellation filter all adopt FIR filters, wherein the secondary channel model and the echo cancellation filter have the same order.

The present invention has the following beneficial effects:

(1) The active noise reduction device provided by the present invention does not require modification of the original piping system, is small in size, easy to use, and can effectively suppress the low-frequency noise radiated from the ventilation pipe opening into the room;

(2) The active noise reduction device provided by the present invention picks up the noise in the ventilation duct through a reference sensor and plays the anti-phase noise through a speaker, so that the noise in the duct and the anti-phase noise cancel each other out, thereby effectively suppressing the low-frequency noise radiated from the ventilation duct opening into the room;

(3) The active noise reduction device provided by the present invention picks up human voice through an error microphone and plays an external input audio signal through a speaker to achieve acoustic interaction;

(4) The active noise reduction control method provided in the present invention has the functions of sound reproduction and array sound perception, which organically combines the active noise control and acoustic interaction functions and promotes each other, so that the two have better performance than when they work separately.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic structural diagram of an active noise reduction device according to the present invention;

FIG2 is a front view of the active noise reduction device of the present invention;

FIG3 is a schematic diagram of a controller in the active noise reduction device of the present invention;

FIG4 is a system block diagram of the active noise reduction control method of the present invention;

FIG5 is a flowchart of the active noise reduction control method of the present invention.

The reference numerals in the figure are as follows:

1. Shell; 11. First through hole; 12. Second through hole; 2. Speaker; 21. Paper cone; 3. Reference microphone; 4. Error microphone; 5. Controller; 51. Wireless communication module; 52. A/D converter; 53. Digital signal processor; 54. D/A converter; 55. Power amplifier; 6. Diffuser; 61. Third through hole; 7. External sound control device; 8. Sound absorbing material layer; 100. Noise reduction module; 101. Noise reduction filter; 102. Acoustic feedback channel model; 103. Secondary channel model; 104. Filter update algorithm module; 200. Echo cancellation module; 201. Echo cancellation filter; 202. Adaptive algorithm module; 300. Noise reduction enhancement module.

DETAILED DESCRIPTION

The present invention will be described in detail below in conjunction with the accompanying drawings FIG. 1 to FIG. 5 and specific embodiments.

Embodiment 1:

Please refer to Figures 1 and 2. An active noise reduction device that is disposed at the outlet of a ventilation duct and is capable of acoustic interaction comprises a shell 1, a loudspeaker 2, a reference microphone 3, a plurality of error microphones 4 (four are shown in Figures 1 to 3), a controller 5, and a diffuser 6. A first through hole 11 is provided on the bottom surface of the shell 1, and a second through hole 12 is provided on the top of the shell 1. The loudspeaker 2 is disposed in the shell 1, and its paper cone 21 faces outward and is disposed on the first through hole 11. The internal space enclosed by the shell 1 and the loudspeaker 2 forms a back cavity of the loudspeaker 2. The reference microphone 3 is disposed on the second through hole 12, and the acoustic input end of the reference microphone 3 faces the outside of the shell 1. Each of the error microphones 4 is disposed on the bottom surface of the shell 1 and is disposed near the bottom surface. The edge of the housing 1 is located at the acoustic input end of each error microphone 4, and the acoustic input end of each error microphone 4 faces the outside of the housing 1, that is, the reference microphone 3 is used to measure the noise outside the housing 1; the controller 5 is arranged in the housing 1, and the speaker 2, the reference microphone 3 and each error microphone 4 are connected to the controller 5 by wires; a third through hole 61 is opened at the center of the diffuser 6, and the bottom surface of the housing 1 is arranged in the third through hole 61. When the diffuser 6 is installed at the ventilation pipe opening, the side of the housing 1 provided with the reference microphone 3 is used as the inner side, which is arranged inside the ventilation pipe opening to pick up the sound signal in the ventilation duct, and the bottom surface of the housing 1 is used as the outer side, which is arranged at the ventilation pipe opening, so that the speaker 2 directly emits sound into the room, and the bottom surface error microphone 4 is used to pick up the sound signal in the room. In this embodiment, the diffuser is rectangular as a whole, and the bottom surface of the housing 1 is correspondingly arranged at the rectangular ventilation pipe opening; in other embodiments, the bottom surface and the grille of the housing 1 can also be circular, and the diffuser is circular as a whole, and is correspondingly arranged at the circular ventilation pipe opening.

The various components of the active noise reduction device of the present invention are further described below.

In this embodiment, as shown in FIG. 1 , the active noise reduction device further includes a sound absorbing material layer 8. In addition to the bottom surface of the shell 1, the outer surface of the shell 1 is provided with the sound absorbing material layer 8, including the outer side of the reference microphone 3. The sound absorbing material layer 8 is used to increase the passive noise reduction effect and reduce the wind noise of the reference microphone 3.

The housing 1 is a conical housing, the second through hole 12 is opened at the top of the conical housing, and the reference microphone 3 is installed on the second through hole 12. The advantages of using a conical housing are: low wind resistance and no additional noise.

The reference microphone 3 is sealed at the connection with the second through hole 12, and the sealing connection is carried out in a manner of gluing or adding a soft pad.

Please refer to Figure 3. The controller 5 includes a wireless communication module 51 and an A/D converter 52, a digital signal processor 53, a D/A converter 54 and a power amplifier 55 connected in sequence. The external sound control device 7 performs two-way communication with the digital signal processor 53 through the wireless communication module 51. The A/D converter 52 collects the output signals of the reference microphone 3 and each of the error microphones 4. The D/A converter 54 outputs the signal that needs to be played by the speaker 2 to the power amplifier 55. The power amplifier 55 is connected to the speaker 2.

The working principle of the present invention is: the active noise reduction device is installed at the ventilation pipe mouth, the reference microphone 3 is located in the ventilation duct, and the sound signal picked up by the reference microphone 3 includes: the original noise signal inside the duct, the anti-phase noise signal emitted by the speaker 2, and the audio signal input by the external sound control device 7 played by the speaker 2, wherein the anti-phase noise signal is the noise with the same amplitude but opposite phase as the original noise emitted after processing by the digital signal processor 53, and the anti-phase noise cancels out the original noise outside the pipe mouth, thereby greatly attenuating the low-frequency noise radiated from the ventilation pipe mouth to the room, and realizing active noise reduction. The error microphone 4 is located at the ventilation duct mouth, and the sound signal picked up includes: the residual noise signal formed by the superposition of the original noise signal propagated from the inside of the pipe and the anti-phase noise signal emitted by the speaker (2), the audio signal input by the external sound control device 7 played by the speaker 2, and the control voice emitted by a person. Among them, active noise reduction can significantly attenuate the residual noise signal, and the audio signal input by the external voice control device 7 can be processed and eliminated by the digital signal processor 53, so as to obtain pure control voice, and then transmit the control voice to the external voice control device 7 through the wireless communication module 51; at the same time, the sound source input of the external voice control device 7 is sent to the digital signal processor 53 through the wireless communication module 51, and after being processed by the digital signal processor 53, the power amplifier 55 drives the speaker 2 to make sound, thereby realizing the acoustic interaction function with the external voice control device 7.

Embodiment 2:

Please refer to FIG. 4 and FIG. 5 , an active noise reduction control method capable of acoustic interaction is combined with the active noise reduction device described in Example 1, wherein the reference microphone 3 picks up the sound signal in the ventilation duct, which includes the original noise signal inside the duct, the anti-phase noise signal emitted by the speaker 2, and the audio signal input by the external sound control device 7 played by the speaker 2; the sound signal picked up by the error microphone 4 includes: the residual noise signal formed by the superposition of the original noise signal propagated from the inside of the duct and the anti-phase noise signal emitted by the speaker 2, the audio signal input by the external sound control device 7 played by the speaker 2, and the control voice emitted by a person; the digital signal processor 53 is provided with a software module for processing the sound signal, including a noise reduction module 100, an echo cancellation module 200 and a noise reduction enhancement module 300; the three modules run simultaneously, regardless of the order;

The control method is specifically as follows:

The low-frequency radiation noise generated in the ventilation duct is suppressed by the noise reduction module 100. Specifically, the noise reduction module 100 includes a noise reduction filter 101, an acoustic feedback channel model 102, a secondary channel model 103 and a filter update algorithm module 104. The noise reduction filter 101 filters the original noise inside the duct and outputs an anti-phase noise with the same amplitude but opposite phase as the original noise. The anti-phase noise is superimposed on the audio signal input by the external sound control device 7 as the output signal of the noise reduction module 100 to drive the speaker 2. At the same time, the output signal of the noise reduction module 100 is also used as the output signal of the acoustic feedback channel model 102. Input; the output signal of the noise reduction module 100 is filtered by the acoustic feedback channel model 102 and then inverted, and then superimposed with the acoustic signal picked up by the reference microphone 3. After superposition, the anti-phase noise signal emitted by the loudspeaker 2 picked up by the reference microphone 3 and the audio signal input by the external sound control device 7 played by the loudspeaker 2 are eliminated, and only the original noise inside the pipeline is left. The original noise is used as the input of the noise reduction filter 101. At the same time, the original noise is also used as the input of the secondary channel model 103. The original noise is filtered by the secondary channel model 103 and used as the input of the filter update algorithm module 104;

The control voice emitted by a person is obtained through the echo cancellation module 200. Specifically, the echo cancellation module 200 includes a plurality of submodules, and the submodules are arranged in a one-to-one correspondence with the error microphone 4. Each submodule includes an echo cancellation filter 201 and an adaptive algorithm module 202. The input of the echo cancellation filter 201 is the audio signal input by the external sound control device 7, which is filtered and inverted, and then superimposed with the sound signal picked up by the error microphone 4. The superimposed signal is used as the output signal of the submodule. The output signal of the submodule removes the audio signal input by the external sound control device 7 played by the speaker 2, but includes the control voice of the person. The output signal of the submodule is sent to the external sound control device 7 through the wireless communication module 51, and combined with the audio signal input by the external sound control device 7 through the wireless communication module 51, acoustic interaction is realized.

At the same time, the output signal of the submodule is also used as the input of the adaptive algorithm module 202 and the input of the filter update algorithm module 104. The input of the adaptive algorithm module 202 also includes the audio signal input by the external voice control device 7. The output of the adaptive algorithm module 202 is used to update the parameters of the echo cancellation filter 201.

The adaptive algorithm module 202 adaptively adjusts the parameters of the echo cancellation filter 201 according to the audio signal input by the external sound control device 7 and the output signal of the submodule, so as to better eliminate the audio signal input by the external sound control device 7 picked up by the error microphone 4 and played by the speaker 2;

The input of the filter update algorithm module 104 includes the output signal of the secondary channel model 103 and the superposition of the output signals of each echo cancellation submodule. The output of the filter update algorithm module 104 is used to adaptively adjust the coefficients of the noise reduction filter 101, the purpose of which is to better suppress the residual noise signal picked up by the error microphone 4, thereby achieving a better active noise reduction effect.

The algorithms adopted by the filter update algorithm module 104 and the adaptive algorithm module include common adaptive algorithms such as LMS, RLS, and AP.

In this embodiment, the noise reduction filter 101, the acoustic feedback channel model 102, the secondary channel model 103 and the echo cancellation filter 201 all adopt FIR filters, wherein the secondary channel model 103 and the echo cancellation filter 201 have the same order.

The input of the noise reduction enhancement module 300 is the parameters of each echo cancellation filter 201. The noise reduction enhancement module 300 superimposes the parameters of each echo cancellation filter 201 and smoothes them in time domain, and uses the result as the new parameter of the secondary channel model 103 to achieve real-time update of the secondary channel model 103.

In the application scenario of the active noise reduction control method of the present invention, at the initial stage of the operation of the noise reduction module 100, the noise reduction module 100 is not running stably, and the sound signal picked up by the error microphone 4 is composed of three parts: one is the low-frequency noise radiated from the ventilation duct outlet to the room, the second is the audio signal (such as music) input by the external sound control device 7 played by the speaker 2, and the third is the control voice issued by people. These three sound signal components interfere with each other, thereby affecting the functional effects of different acoustic processing modules.

First, when the noise reduction module 100 works stably, the low-frequency noise component radiated from the pipe mouth to the room will be greatly suppressed, and the low-frequency noise radiated from the pipe mouth of the ventilation duct to the room picked up by the error microphone 4 will be attenuated to a large extent. The basic principle of the noise reduction module 100 is: the reference microphone 3 picks up the original noise signal inside the pipe, and the noise reduction filter 101 filters and drives the speaker 2 to emit an anti-phase noise with the same amplitude as the original noise but opposite phase. The noise and the original noise cancel each other, thereby greatly attenuating the low-frequency noise radiated from the ventilation pipe mouth to the room. However, in actual work, the signal picked up by the reference microphone 3 mainly includes four parts: one is the original noise signal inside the pipe, the second is the anti-phase noise emitted by the speaker 2, the third is the audio signal (such as music) input by the external sound control device 7 played by the speaker 2, and the fourth is the control voice emitted by a person. Since the reference microphone 3 is placed inside the pipe, the fourth part of the signal is weak and its influence can be ignored; however, the second and third parts will form a feedback loop, which seriously interferes with the normal operation of the noise reduction module 100. In order to eliminate this feedback loop, the acoustic feedback channel model 102 is used to establish an acoustic transfer function model between the loudspeaker 2 and the reference microphone 3, so as to predict the second and third part signals and eliminate them from the reference signal.

The signal driving the speaker 2 includes two parts: one is the output of the noise reduction filter 101, which drives the speaker 2 to emit anti-phase noise to complete the active noise reduction function; the other is the audio signal input by the external voice control device 7 to drive the speaker 2 to complete sound playback, that is, to emit the sound expected by the user, such as music.

After the noise reduction module 100 works stably, the anti-phase noise emitted by the speaker 2 will cancel out the original noise of the ventilation duct, thereby greatly attenuating the first part of the noise energy picked up by the error microphone 4. However, since the speaker 2 is close to the error microphone 4, the audio signal (i.e., the sound expected by the user) input by the external sound control device 7 played by the speaker 2 has a large energy, which will strongly interfere with other sound signal components in the room. The function of the echo cancellation module 200 is to remove this interference, and the signal-to-noise ratio of other sound signal components in the room in the output signal of the module will be greatly improved.

Secondly, when the noise reduction module 100 has not entered a stable working state, the output signals of each submodule of the echo cancellation still have a strong low-frequency radiation noise component of the pipe mouth. Therefore, the output signals of all echo cancellation submodules are superimposed and input into the noise reduction module 100, and the filter update algorithm module 104 adjusts the parameters of the noise reduction filter 101 according to the low-frequency noise energy in the superimposed signal, so that the noise reduction module 100 enters a stable working state. Another input of the filter update algorithm module 104 is the original noise signal in the pipe filtered by the secondary channel model 103, and the acoustic transfer function model of the sum of the signals picked up by the speaker 2 to all error microphones 4 is established by using the secondary channel model 103. The filter update algorithm module 104 can adopt general adaptive algorithms such as LMS, RLS, AP, etc.

In the field of active noise control, the secondary channel model 103 refers to the transfer function between the loudspeaker 2 and the error microphone 4, and in the present invention, refers to the transfer function from the loudspeaker 2 to the sum of all error microphone 4 signals. As mentioned above, the noise reduction module 100 needs to use the model of this transfer function to complete the adaptive adjustment function of the parameters of the noise reduction filter 101. In practical applications, the frequency response of the secondary channel is often time-varying due to obstacles, moving objects in the room, and the fatigue characteristics of the loudspeaker 2. This will cause a difference between the secondary channel model 103 and the actual response, thereby affecting the noise reduction effect and even the stability of the overall module. The echo cancellation module 200 just contains an effective estimate of the time-varying secondary channel model 103, so the echo cancellation module 200 and the noise reduction module 100 are organically combined using the noise reduction enhancement module.

Now we give the specific implementation process in combination with the specific algorithm:

，第i路误差传声器信号

，外部音源输入信号

，降噪滤波器输出信号

，降噪模块输出信号

，声反馈通道模型输出信号

，次级通道模型输出信号

，对应于第i路误差传声器的回声抵消子模块输出信号

。The signals are shown below: Reference microphone signal

, the i-th error microphone signal

, external audio source input signal

, the noise reduction filter output signal

, the noise reduction module output signal

, the output signal of the acoustic feedback channel model

, the secondary channel model output signal

, corresponding to the output signal of the echo cancellation submodule of the i-th error microphone

.

Noise Reduction Module 100:

，滤波器输入、输出分别为：The noise reduction filter 101 is an N-order FIR filter with coefficients

, the filter input and output are:

，

,

；

;

和

的叠加：

，该信号通过功率放大器55驱动扬声器2发声；The output of the noise reduction module 100 is

and

The superposition of:

, the signal drives the speaker 2 to produce sound through the power amplifier 55;

还作为声反馈通道模型102的输入信号，产生输出

；声反馈通道模型102是一K阶FIR滤波器，其滤波器系数

，其输入输出关系为：at the same time,

It also serves as the input signal of the acoustic feedback channel model 102, generating the output

The acoustic feedback channel model 102 is a K-order FIR filter, whose filter coefficients

, its input-output relationship is:

；

;

，输入输出关系为：The secondary channel model 103 is an L-order FIR filter, whose filter coefficients are

, the input-output relationship is:

；

;

叠加得到

，利用

和

、根据LMS算法对降噪滤波器101系数进行自适应更新：All

Superposition

,use

and

, adaptively update the coefficients of the denoising filter 101 according to the LMS algorithm:

，

,

，

,

为一常数，典型取值为0.1。In the formula

is a constant, with a typical value of 0.1.

置0，

和

则通过离线系统辨识的方法得到，其中声反馈通道模型102是功率放大器55输入信号和参考传声器3输出信号间的传递函数，而次级通道模型103则是功率放大器55输入信号和所有误差传声器4输出叠加信号间的传递函数。Initialization: Before the algorithm is executed,

Set to 0,

and

It is obtained through an offline system identification method, wherein the acoustic feedback channel model 102 is a transfer function between the input signal of the power amplifier 55 and the output signal of the reference microphone 3, and the secondary channel model 103 is a transfer function between the input signal of the power amplifier 55 and the superimposed signal of all error microphones 4 outputs.

Echo cancellation module 200:

，输入

与输出

的关系为：The echo cancellation filter 201 corresponding to the i-th echo cancellation submodule is an L-order FIR filter with a filter coefficient

,enter

With output

The relationship is:

；

;

The output of the echo cancellation submodule is:

；

;

进行更新：Using LMS algorithm

To update:

，

,

为一常数，典型取值为0.1。In the formula

is a constant, with a typical value of 0.1.

置为0。Initialization: Before the algorithm is executed,

Set to 0.

Noise Reduction Enhancement Module 300:

相加，利用时域平滑算法进行处理，得到的结果对

进行更新：Set all echo cancellation filter 201 coefficients

Add and process using time domain smoothing algorithm. The result is

To update:

，

,

为一常数，典型取值为0.01。In the formula

is a constant, with a typical value of 0.01.

The present invention provides an active noise reduction control method capable of acoustic interaction, which has a sound playback function - the user uses an external sound control device 7 to provide a sound source input signal, and the audio that the user wants to hear is played through a loudspeaker 2; and an array sound perception function - an array is formed by multiple error sensors to perceive the control voice emitted by a person and send it to the external sound control device 7; and at the same time, active noise control and echo cancellation control are organically combined. The active noise control adopts a feedforward structure and algorithm, which can suppress the ventilation duct noise at the error microphone 4, so that the array formed by the error microphone 4 is less interfered by low-frequency noise when picking up indoor acoustic signals (i.e., the control voice emitted by a person); at the same time, when the sound source of the external sound control device 7 is input for indoor sound playback, the intermediate parameters of the adaptive algorithm of echo cancellation are used to dynamically and real-time adjust the parameters of the secondary channel model 103, so as to realize the real-time update of the physical channel model between the loudspeaker 2 and the error microphone 4, thereby improving the stability and noise reduction effect of the active noise reduction system. In addition, since the ventilation pipe opening is generally located in the middle of the indoor ceiling, the array formed by the error microphone 4 will be less affected by the sound reflection and scattering caused by indoor furniture and moving objects, thereby improving the accuracy of acoustic signal pickup and the comfort of acoustic interactive products.

The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims (3)

1. Locate the active noise reduction device that ventilates the mouth of pipe and can acoustic interaction, its characterized in that: comprises a shell (1), a loudspeaker (2), a reference microphone (3), a plurality of error microphones (4), a controller (5) and a diffuser (6); a first through hole (11) is formed in the bottom surface of the shell (1), and a second through hole (12) is formed in the top of the shell (1); the loudspeaker (2) is arranged in the shell (1), the cone (21) of the loudspeaker is outwards arranged on the first through hole (11), and the inner space enclosed by the shell (1) and the loudspeaker (2) forms a rear cavity of the loudspeaker (2); the reference microphone (3) is arranged on the second through hole (12) and the acoustic input end of the reference microphone (3) faces the outside of the shell (1); each error microphone (4) is arranged on the bottom surface of the shell (1) and is arranged at a position close to the edge of the bottom surface, and the acoustic input end of each error microphone (4) faces the outside of the shell (1); the controller (5) is arranged in the shell (1), and the loudspeaker (2), the reference microphone (3) and the error microphones (4) are all connected to the controller (5) through leads; a third through hole (61) is formed in the center of the diffuser (6), the bottom surface of the shell (1) is arranged in the third through hole (61), when the diffuser (6) is installed on a ventilation pipe orifice, one side of the shell (1) provided with the reference microphone (3) is used as the inner side and is arranged in the ventilation pipe orifice, the bottom surface of the shell (1) is used as the outer side and is arranged in the ventilation pipe orifice, and the loudspeaker (2) directly sounds indoors; the controller (5) comprises a wireless communication module (51), and an A/D converter (52), a digital signal processor (53), a D/A converter (54) and a power amplifier (55) which are sequentially connected, wherein an external sound control device (7) is in bidirectional communication with the digital signal processor (53) through the wireless communication module (51), the A/D converter (52) collects output signals of the reference microphone (3) and each error microphone (4), the D/A converter (54) outputs signals required to be played by a loudspeaker (2) to the power amplifier (55), and the power amplifier (55) is connected with the loudspeaker (2); the sound absorption device further comprises a sound absorption material layer (8), wherein the sound absorption material layer (8) is arranged on the outer surface of the shell (1) except the bottom surface of the shell (1), the sound absorption material layer (8) comprises the outer side of the reference microphone (3), and the sound absorption material layer (8) is used for increasing the passive noise reduction effect and reducing the wind noise of the reference microphone (3); the shell (1) is a conical shell, the second through hole (12) is formed in the top of the conical shell, and the reference microphone (3) is mounted on the second through hole (12) in a sealing mode.

2. An active noise reduction control method capable of acoustic interaction, characterized in that it combines the active noise reduction device of claim 1, the reference microphone (3) picks up an acoustic signal in the ventilation duct, which comprises an original noise signal inside the duct, an inverted noise signal emitted by the speaker (2) and an audio signal input by an external sound control device (7) played by the speaker (2); the acoustic signal picked up by the error microphone (4) comprises: the device comprises a residual noise signal formed by superposition of an original noise signal transmitted from the inside of a pipeline and an inverted noise signal emitted by a loudspeaker (2), an audio signal input by an external sound control device (7) played by the loudspeaker (2) and control voice emitted by a person; the digital signal processor (53) is internally provided with a software module which is operated simultaneously and is used for processing acoustic signals, and the software module comprises a noise reduction module (100), an echo cancellation module (200) and a noise reduction enhancement module (300), wherein the noise reduction module (100), the echo cancellation module (200) and the noise reduction enhancement module (300) are operated simultaneously in the process of processing noise reduction and acoustic interaction, and the noise reduction are not sequential;

the control method specifically comprises the following steps:

-suppressing, by the noise reduction module (100), low frequency radiation noise generated in the ventilation duct: the noise reduction module (100) comprises a noise reduction filter (101), an acoustic feedback channel model (102), a secondary channel model (103) and a filter updating algorithm module (104); the noise reduction filter (101) filters original noise in a pipeline and outputs inverted noise with the same amplitude as the original noise but opposite phase, and the inverted noise is overlapped with an audio signal input by the external sound control equipment (7) and then is used as an output signal of the noise reduction module (100) to drive the loudspeaker (2) and is also used as an input of the acoustic feedback channel model (102); the output signal of the noise reduction module (100) is filtered by the acoustic feedback channel model (102) and then is inverted, the inverted noise signal which is picked up by the reference microphone (3) and is sent by the loudspeaker (2) and the audio signal which is input by the external sound control equipment (7) and is played by the loudspeaker (2) are eliminated after being overlapped, only original noise in a pipeline is left, the original noise is used as the input of the noise reduction filter (101), meanwhile, the original noise is also used as the input of the secondary channel model (103), and the original noise is filtered by the secondary channel model (103) and is used as the input of the filter updating algorithm module (104);

the echo cancellation module (200) is used for acquiring control voice sent by a person: the echo cancellation module (200) comprises a plurality of sub-modules, the sub-modules are arranged in one-to-one correspondence with the error microphone (4), and each sub-module comprises an echo cancellation filter (201) and an adaptive algorithm module (202); the input of the echo cancellation filter (201) is an audio signal input by an external sound control device (7), the audio signal is filtered and inverted, and then the audio signal is overlapped with an acoustic signal picked up by the error microphone (4), the overlapped signal is used as an output signal of the sub-module, the output signal of the sub-module removes the audio signal input by the external sound control device (7) played by the loudspeaker (2), but comprises control voice of a person, the output signal of the sub-module is sent to the external sound control device (7) through the wireless communication module (51), and the audio signal input by the external sound control device (7) through the wireless communication module (51) is combined to realize acoustic interaction; meanwhile, the output signals of the sub-modules are also used as the input of the self-adaptive algorithm module (202) and the input of the filter updating algorithm module (104), the input of the self-adaptive algorithm module (202) also comprises an audio signal input by an external sound control device (7), the self-adaptive algorithm module (202) outputs a parameter used for updating the echo cancellation filter (201), and the self-adaptive algorithm module (202) carries out self-adaptive adjustment on the parameter of the echo cancellation filter (201) according to the audio signal input by the external sound control device (7) and the output signal of the sub-module, so as to better eliminate the audio signal which is picked up by an error microphone (4) and is input by the external sound control device (7) played by the loudspeaker (2);

the input of the filter updating algorithm module (104) comprises superposition of the output signal of the secondary channel model (103) and the output signal of each echo cancellation sub-module, and the output of the filter updating algorithm module (104) is used for adaptively adjusting the coefficient of the noise reduction filter (101) so as to better inhibit the residual noise signal picked up by the error microphone (4), thereby realizing better active noise reduction effect;

the input of the noise reduction enhancement module (300) is the parameter of each echo cancellation filter (201), and the noise reduction enhancement module (300) superimposes and time-domain smoothes the parameter of each echo cancellation filter (201), and then the result is used as a new parameter of the secondary channel model (103) to realize the real-time update of the secondary channel model (103).

3. An active noise reduction control method capable of acoustic interaction according to claim 2, characterized in that: the noise reduction filter (101), the acoustic feedback channel model (102), the secondary channel model (103) and the echo cancellation filter (201) all adopt FIR filters, wherein the secondary channel model (103) and the echo cancellation filter (201) have the same order.

Images