CN115565515A - Step-by-step virtual sensing noise reduction method - Google Patents
Step-by-step virtual sensing noise reduction method Download PDFInfo
- Publication number
- CN115565515A CN115565515A CN202211186487.8A CN202211186487A CN115565515A CN 115565515 A CN115565515 A CN 115565515A CN 202211186487 A CN202211186487 A CN 202211186487A CN 115565515 A CN115565515 A CN 115565515A
- Authority
- CN
- China
- Prior art keywords
- filter
- physical
- virtual
- microphone
- noise
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
- G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
- G10K11/1781—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase characterised by the analysis of input or output signals, e.g. frequency range, modes, transfer functions
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
- G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
- G10K11/1781—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase characterised by the analysis of input or output signals, e.g. frequency range, modes, transfer functions
- G10K11/17813—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase characterised by the analysis of input or output signals, e.g. frequency range, modes, transfer functions characterised by the analysis of the acoustic paths, e.g. estimating, calibrating or testing of transfer functions or cross-terms
- G10K11/17817—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase characterised by the analysis of input or output signals, e.g. frequency range, modes, transfer functions characterised by the analysis of the acoustic paths, e.g. estimating, calibrating or testing of transfer functions or cross-terms between the output signals and the error signals, i.e. secondary path
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
- G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
- G10K11/1785—Methods, e.g. algorithms; Devices
- G10K11/17853—Methods, e.g. algorithms; Devices of the filter
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
- G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
- G10K11/1787—General system configurations
Abstract
The invention provides a step virtual sensing noise reduction method. The method comprises the following steps: the method comprises the following steps: temporarily placing a virtual microphone at a target noise reduction position to acquire a virtual error signal, and modeling a virtual secondary path; step two: training a noise control filter; step three: placing the virtual microphones in the first step and the second step as physical microphones at physical positions to acquire physical error signals, and modeling a physical secondary path; step four: placing no microphone at the target noise reduction position, placing a physical microphone at the physical position, and training an auxiliary filter; step five: and entering a control stage to realize active noise reduction of the target noise reduction position. The step-by-step virtual sensing noise reduction method provided by the invention has the advantages that the step-by-step calculation is realized, the hardware resource consumption is low, meanwhile, the better noise reduction effect can be achieved when the primary noise intensity range is larger, and the practicability is strong.
Description
Technical Field
The invention relates to a step virtual sensing noise reduction method.
Background
With the rapid development of science and technology, the human society enters the industrial society, people gradually find some defects caused by the technology development while enjoying convenience brought by science and technology, noise pollution is one of the defects, and as noise increasingly affects the daily life of people, the requirement of people for reducing noise interference is increasingly improved. The traditional noise control belongs to Passive Noise Control (PNC), the control mechanism is that noise sound waves interact with acoustic materials or structures to achieve the purpose of noise reduction, the PNC has limitation on the difficulty of noise reduction of low-frequency noise signals, and the appearance and development of Active Noise Control (ANC) make up for the defect. In recent years, ANC has gradually evolved as one of the main research directions in the field of noise control.
ANC active noise reduction technology: refers to a technique for collecting ambient noise and generating a signal in the opposite direction of the noise to cancel the noise. When the control source radiates anti-noise waves of the same amplitude and opposite phase as the noise source, the ANC system may reduce the noise at the desired location where the error microphone is placed.
ANC noise reduction systems are structurally classified into a feedforward type and a feedback type according to the use of a microphone. The feed-forward system has stable performance and is suitable for processing broadband noise; the feedback system can only reduce single-frequency noise and narrow-band noise, and has limitations. ANC noise reduction systems may also be divided into single-channel ANC systems and multi-channel ANC systems, depending on the number of speakers used.
In some application scenarios, such as a wireless bluetooth headset, an error microphone cannot be placed at a target noise reduction position, and a virtual sensing noise reduction technology is developed accordingly. The virtual sensing noise reduction technology is mainly divided into two stages: a training phase and a control phase. In the training stage, a virtual microphone needs to be temporarily placed at a target noise reduction position to obtain a transfer relation between a physical position and a virtual position; in the control stage, the virtual microphone at the target noise reduction position is removed, the noise control filter is updated according to the signal acquired by the physical microphone and the transfer relationship between the physical position and the virtual position obtained in the training stage, and finally the stable state is achieved, so that the active noise reduction at the target noise reduction position is realized. Currently, there are two virtual sensing technologies: remote microphone method (RM) and auxiliary filter method (AF). For the AF method, one more microphone is used in the training stage than in the control stage, and the requirements on the number of interfaces and the calculation amount of a hardware controller are high.
Disclosure of Invention
The invention aims to: in order to solve the defects that one microphone is needed to be used in the training stage of the AF method more than in the control stage and the consumption of hardware resources is high, and simultaneously, in order to ensure that the noise reduction of a target noise reduction position can reach a specified threshold value when the volume range of a reference signal is large, the invention provides a step virtual sensing noise reduction method, which comprises the following steps:
the method comprises the following steps: modeling a virtual secondary path S v Temporarily placing the virtual microphone at a target noise reduction position to collect a virtual error signal, and updating a virtual secondary path filter according to the LMS algorithm under the combined action of a secondary sound source emitted by a secondary loudspeaker and the virtual error signal received by the virtual microphone
Finally reaching a stable state to obtain a virtual secondary path filter
Step two: training a noise control filter W, and according to the FxLMS algorithm, taking a virtual error signal received by the virtual microphone as a reference signal, passing through the output of the noise control filter W and superposing the primary noise, and passing the reference signal through a virtual secondary path filter
The output of the noise control filter W and the virtual error signal received by the virtual microphone act together to update the noise control filter W, and finally the noise control filter W is obtained in a stable state;
step three: modeling a physical secondary pathway S m Taking the virtual microphone in the first step and the second step as a physical microphone to be placed at a physical position to collect a physical error signal, and updating a physical secondary path filter under the combined action of a secondary sound source emitted by a secondary loudspeaker and the physical error signal received by the physical microphone according to an LMS algorithm
Finally reach the stabilityState, get physical secondary path filter
Step four: training an auxiliary filter H, placing no microphone at the target noise reduction position, placing a physical microphone at the physical position, and taking the noise control filter W obtained by training in the step two as a fixed coefficient filter W 0 The physical error signal received by the physical microphone is a reference signal and passes through a fixed coefficient filter W 0 And a physical secondary path filter
The output of the auxiliary filter is superposed with the primary noise, the residual signal generated by the auxiliary filter is the superposition of the physical error signal received by the physical microphone and the output of the reference signal through the auxiliary filter H, the reference signal and the residual signal generated by the auxiliary filter act together to update the auxiliary filter H, and finally the stable state is achieved to obtain the auxiliary filter H;
step five: entering a control stage, placing no microphone at the target noise reduction position, placing a physical microphone at a physical position, taking a physical error signal received by the physical microphone as a reference signal, and passing the reference signal through a noise control filter W and a physical secondary path filter
The output of the step four and the primary noise are superposed, and the auxiliary filter H obtained by training in the step four is used as a fixed coefficient filter H 0 The residual signal generated by the auxiliary filter is the physical error signal received by the physical microphone and the reference signal, and the residual signal is processed by the fixed coefficient filter H 0 The reference signal is passed through a physical secondary path filter
The output of the noise control filter is combined with the residual signal generated by the auxiliary filter to update the noise control filter W, and finally the stable state is achieved, so that the active noise reduction of the target noise reduction position is realized.
Further, in step two, the process of updating the noise control filter W is as follows:
by the formula
Updating the weight coefficient vector, wherein w (n) represents the weight coefficient vector before the noise control filter is updated, w (n + 1) represents the updated weight coefficient vector, x (n) represents the reference signal,
representing a virtual secondary path filter, e v (n) represents a virtual error signal, mu w1 Representing the convergence factor of the noise control filter update.
Further, in step four, the process of updating the auxiliary filter H is as follows:
by the formula h (n + 1) = h (n) - μ h e h (n) x (n) updates the weight coefficient vector, where h (n) represents the weight coefficient vector before the update of the auxiliary filter, h (n + 1) represents the weight coefficient vector after the update, x (n) represents the reference signal, e h (n) represents the residual signal, μ, produced by the auxiliary filter h Representing the convergence factor of the auxiliary filter update.
Further, in step five, the process of updating the noise control filter W is as follows:
by the formula w (n + 1) = w (n) - μ w2 e h (n)x s (n) updating the weight coefficient vector, wherein w (n) represents the weight coefficient vector before the noise control filter is updated, w (n + 1) represents the updated weight coefficient vector, and x s (n) represents the output of the reference signal through the physical secondary path filter, e h (n) represents the residual signal, μ, produced by the auxiliary filter w2 Representing the convergence factor of the noise control filter update.
Further, in order to meet the noise reduction requirement when the primary noise intensity range is large, the convergence step lengths are respectively set as:
wherein, mu w1 Representing the convergence factor, μ, updated by the noise control filter in step two h Represents the convergence factor, μ, of the update of the auxiliary filter in step four w2 Representing the convergence factor updated by the noise control filter in step five, and x represents the reference signal.
In summary, due to the adoption of the technical scheme, the invention has the beneficial effects that: the problem that hardware resource consumption is large because one microphone is used in the training stage of the AF method more than in the control stage is solved, and when the volume range of a reference signal is large, noise reduction of a target noise reduction position can reach a specified threshold value.
Drawings
FIG. 1 is a process block diagram of a step-by-step virtual sensing noise reduction method of the present invention, a) a step-by-step process block diagram; b) Step four, a process block diagram; c) Step five process block diagrams.
Fig. 2 is a diagram of a secondary path when the reference signal volume is medium according to the present invention.
Fig. 3 is a diagram of an error signal in step two and step four when the volume of the reference signal is medium in the embodiment of the present invention.
Fig. 4 is a diagram of an error signal in step five when the reference signal volume is medium according to an embodiment of the present invention.
Fig. 5 is a noise power spectrum of step five when the reference signal volume is low, medium and high respectively according to the embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the following embodiments and accompanying drawings.
The invention provides a step virtual sensing noise reduction method, which aims to solve the defects that one microphone is needed to be used in a training stage of an AF method and hardware resource consumption is high compared with a control stage, and meanwhile, the noise reduction of a target noise reduction position can reach a specified threshold value when a reference signal volume range is large.
FIG. 1 shows a process diagram of the step-by-step virtual sensing noise reduction method of the present invention, in the present embodiment, a step one is to temporarily place a virtual microphone at a target noise reduction position to collect a virtual error signal, and to model a virtual secondary path S v Expressed as:
y v (n)＝s v (n)*y(n)
wherein y (n) denotes a secondary sound source emitted by a secondary loudspeaker, y v (n) representing the virtual error signal received by the virtual microphone, the secondary sound source emitted by the secondary speaker and the virtual error signal received by the virtual microphone together update the virtual secondary path filter
Finally reaching a stable state to obtain a virtual secondary path filter
In this embodiment, step two (refer to fig. 1 (a)) trains the noise control filter W, and according to the FxLMS algorithm, the virtual error signal received by the virtual microphone is the output of the reference signal passing through the noise control filter W and the superposition of the primary noise, which is represented as:
e v (n)＝d v (n)+s v (n)*y(n)
wherein d is v (n) denotes primary noise, e v (n) represents a virtual error signal received by the virtual microphone. The reference signal passes through a virtual secondary path filter
The output of which cooperates with a virtual error signal received by a virtual microphone to update the noise controlThe process of making the filter W and updating the noise control filter W is as follows:
finally, the stable state is achieved, and the noise control filter W is obtained. Wherein w (n) represents the weight coefficient vector before the noise control filter is updated, w (n + 1) represents the weight coefficient vector after the update, x (n) represents the reference signal,
representing a virtual secondary path filter, e v (n) represents a virtual error signal, mu w1 Representing the convergence factor of the noise control filter update.
In the present embodiment, in the third step, the virtual microphones in the first and second steps are placed at physical positions as physical microphones to collect physical error signals, and a physical secondary path S is modeled m Expressed as:
y m (n)＝s m (n)*y(n)
where y (n) denotes a secondary sound source emitted by a secondary loudspeaker, y m (n) represents a physical error signal received by the physical microphone, and a secondary sound source emitted by the secondary speaker co-acts with the physical error signal received by the physical microphone to update the physical secondary path filter
Finally, the stable state is achieved, and the physical secondary path filter is obtained
In this embodiment, in step four (refer to fig. 1 (b)), no microphone is placed at the target noise reduction position, a physical microphone is placed at the physical position, the auxiliary filter H is trained, and the noise control filter W obtained by training in step two is used as the fixed coefficient filter W 0 The physical error signal received by the physical microphone is used as a reference signal and passes through a fixed coefficient filterW 0 And a physical secondary path filter
Is added to the primary noise, expressed as:
wherein d is m (n) represents primary noise, e m (n) represents a physical error signal received by a physical microphone. The residual signal generated by the auxiliary filter is the superposition of the physical error signal received by the physical microphone and the output of the reference signal through the auxiliary filter H, the auxiliary filter H is updated by the combined action of the reference signal and the residual signal generated by the auxiliary filter, and the updating process of the auxiliary filter H is as follows:
h(n+1)＝h(n)-μ h e h (n)x(n)
and finally, reaching a stable state to obtain an auxiliary filter H. Where h (n) represents the weight coefficient vector before the update of the auxiliary filter, h (n + 1) represents the weight coefficient vector after the update, x (n) represents the reference signal, e h (n) represents the residual signal, μ, produced by the auxiliary filter h Representing the convergence factor of the auxiliary filter update.
In this embodiment, in step five (refer to fig. 1 (c)), no microphone is placed at the target noise reduction position, a physical microphone is placed at the physical position, and in the control stage, the physical error signal received by the physical microphone is the reference signal and passes through the noise control filter W and the physical secondary path filter
Is added to the primary noise, expressed as:
wherein d is m (n) represents primary noise, e m (n) denotes physical microphone connectionA received physical error signal. Taking the auxiliary filter H obtained by the training in the step four as a fixed coefficient filter H 0 The residual signal generated by the auxiliary filter is the physical error signal received by the physical microphone and the reference signal which are passed through the fixed coefficient filter H 0 By superposition of the outputs of the reference signal through a physical secondary path filter
And the output of the auxiliary filter, together with the residual signal generated by the auxiliary filter, updates the noise control filter W by:
w(n+1)＝w(n)-μ w2 e h (n)x s (n)
finally, a stable state is achieved, and active noise reduction of the target noise reduction position is achieved. Wherein w (n) represents the weight coefficient vector before the noise control filter is updated, w (n + 1) represents the weight coefficient vector after the update, and x s (n) represents the output of the reference signal through the physical secondary path filter, e h (n) represents the residual signal, μ, produced by the auxiliary filter w2 Representing the convergence factor of the noise control filter update.
In terms of computation amount (M represents one multiplication, A represents one addition, L is the filter length, LM represents L multiplications, and the filter length is the same by default):
the operation amount of the one-time step is as follows: (2L + 1) M +2LA
The operation amount of the first step is as follows: (4L + 1) M + (4L-2) A
The operation amount of the first step is as follows: (2L + 1) M +2LA
The four operation quantities of the first step are as follows: (4L + 1) M + (4L-1) A
The operation amount of the step five is executed once: (6L + 1) M + (6L-2) A
The specific implementation method predicts the noise reduction amount of the target noise reduction position by a method of firstly recording and then simulating the acquired recording in MATLAB R2016b software. The experiment was performed in a local semi-enclosed space. The noise source is noise recording (volume is divided into low, medium and high), reference microphoneThe wind is placed behind the local semi-closed space, and the secondary channel S is virtualized v Physical secondary path S m The secondary sound source emitted by the secondary speaker is white in the modeling. The ear canal mouth of the left ear in the local semi-closed space is set to be a target noise reduction position, the ear canal mouth of the left ear, the secondary loudspeaker and the physical microphone are arranged on the same horizontal line, the secondary loudspeaker is located at the left part of the local semi-closed space, the physical microphone is 1cm away from the secondary loudspeaker, and the ear canal mouth of the left ear is 8cm away from the physical microphone.
All recordings were 25s for the data measured in this experiment. In simulation, the lengths of the noise control filter and the auxiliary filter are set to 384 orders. In order to enable the noise reduction of the target noise reduction position to reach the designated threshold value when the primary noise intensity range is large, and the value of the known convergence factor is related to the inverse ratio of the energy of the reference signal, the convergence step length of the invention is respectively set as:
wherein, mu w1 Representing the convergence factor, μ, updated by the noise control filter in step two h Represents the convergence factor, μ, of the update of the auxiliary filter in step four w2 Representing the convergence factor updated by the noise control filter in step five and x representing the reference signal.
Firstly, the data training and control stage condition when the reference signal volume is middle is specifically analyzed, and finally, the noise reduction effect of the method when the reference signal volume is respectively low, middle and high in the application of the local semi-closed space is discussed. All of the following results were obtained by MATLAB R2016b software simulation:
fig. 2 is the resulting sub-path filter. The sampling rate of the system is set to 24k and the impulse response length of the virtual, physical sub-path is 16ms.
Fig. 3 shows the error signal diagram of step two and step four when the reference signal volume is medium. Observing fig. 3, step two reaches a steady state after t =0.5, and step four reaches a steady state after t =1 s.
From fig. 4, it can be observed that step five reaches a convergence state when t is very small, i.e. the system reaches a steady state in a very short time.
Fig. 5 shows the noise reduction effect when the reference signal volume is low, medium, and high, respectively. From fig. 5 it can be seen that: when the volume of the reference signal is low, the noise reduction effect at the target noise reduction position reaches about 20 dB; when the volume of the reference signal is middle, the noise reduction effect at the target noise reduction position reaches about 22 dB; when the volume of the reference signal is high, the noise reduction effect at the target noise reduction position reaches about 22 dB. Therefore, when the reference signal is in a larger volume range, the active quiet zone can be formed at the target noise reduction position.
In summary, the effectiveness of the method of the present invention is demonstrated by using a step-by-step virtual sensing noise reduction method to reduce the use of one microphone, save hardware resources, and form an active quiet zone for the target noise reduction position when the reference signal is in a large volume range.
It should be noted that the above-mentioned embodiments are not intended to limit the scope of the present invention, and the equivalent changes or substitutions made on the basis of the above-mentioned technical solutions and the modifications based on the above-mentioned technical solutions are all within the scope of the present invention as claimed in the claims.
Claims (5)
1. A step virtual sensing noise reduction method is characterized by comprising the following steps:
the method comprises the following steps: modeling a virtual secondary path S v The virtual microphone is temporarily placed at a target noise reduction position to collect a virtual error signal, and according to the LMS algorithm, a secondary sound source emitted by a secondary loudspeaker and the virtual error signal received by the virtual microphone act togetherNew virtual secondary path filter
Finally, the stable state is achieved, and the virtual secondary path filter is obtained
Step two: training a noise control filter W, and according to the FxLMS algorithm, taking a virtual error signal received by the virtual microphone as a reference signal, passing through the output of the noise control filter W and superposing the primary noise, and passing the reference signal through a virtual secondary path filter
The output of the virtual microphone and the virtual error signal received by the virtual microphone act together to update the noise control filter W, and finally the noise control filter W is obtained after the stable state is reached;
step three: modeling a physical secondary pathway S m Placing the virtual microphones in the first step and the second step as physical microphones at physical positions to collect physical error signals, and updating a physical secondary path filter under the combined action of a secondary sound source emitted by a secondary loudspeaker and the physical error signals received by the physical microphones according to an LMS algorithm
Finally, the stable state is achieved, and the physical secondary path filter is obtained
Step four: training an auxiliary filter H, placing a physical microphone at a physical position without placing any microphone at a target noise reduction position, and taking the noise control filter W obtained by training in the step two as a fixed coefficient filter W 0 The physical error signal received by the physical microphone is a reference signal and passes through a fixed coefficient filter W 0 And a physical secondary path filter
The output of the auxiliary filter is superposed with the primary noise, the residual signal generated by the auxiliary filter is the superposition of the physical error signal received by the physical microphone and the output of the reference signal through the auxiliary filter H, the reference signal and the residual signal generated by the auxiliary filter act together to update the auxiliary filter H, and finally the stable state is achieved to obtain the auxiliary filter H;
step five: in the control stage, no microphone is placed at the target noise reduction position, a physical microphone is placed at the physical position, a physical error signal received by the physical microphone is a reference signal, and the reference signal passes through a noise control filter W and a physical secondary path filter
The output of the step four and the primary noise are superposed, and the auxiliary filter H obtained by training in the step four is used as a fixed coefficient filter H 0 The residual signal generated by the auxiliary filter is the physical error signal received by the physical microphone and the reference signal which are passed through the fixed coefficient filter H 0 The reference signal is passed through a physical secondary path filter
The output of the noise control filter is combined with the residual signal generated by the auxiliary filter to update the noise control filter W, and finally the stable state is achieved, so that the active noise reduction of the target noise reduction position is realized.
2. The step-by-step virtual sensing noise reduction method according to claim 1, wherein in the second step, the updating process of the noise control filter W is:
by the formula
Updating the weight coefficient vector, wherein w (n) represents the weight coefficient vector before the noise control filter is updated, w (n + 1) represents the updated weight coefficient vector, and x (n) represents the reference signal，
Representing a virtual secondary path filter, e v (n) represents a virtual error signal, mu w1 Representing the convergence factor of the noise control filter update.
3. The method of claim 1, wherein in step four, the updating of the auxiliary filter H comprises:
by the formula h (n + 1) = h (n) - μ h e h (n) x (n) updates the weight coefficient vector, where h (n) represents the weight coefficient vector before the update of the auxiliary filter, h (n + 1) represents the weight coefficient vector after the update, x (n) represents the reference signal, e h (n) represents the residual signal, μ, produced by the auxiliary filter h Representing the convergence factor of the auxiliary filter update.
4. The method according to claim 1, wherein in step five, the updating process of the noise control filter W is as follows:
by the formula w (n + 1) = w (n) - μ w2 e h (n)x s (n) updating the weight coefficient vector, wherein w (n) represents the weight coefficient vector before updating the noise control filter, w (n + 1) represents the weight coefficient vector after updating, and x s (n) represents the output of the reference signal through the physical secondary path filter, e h (n) represents the residual signal, μ, produced by the auxiliary filter w2 Representing the convergence factor of the noise control filter update.
5. The method for reducing noise of step-by-step virtual sensing according to claims 2 to 4, wherein the convergence factor is set as:
wherein, mu w1 Representing the convergence factor, μ, updated by the noise control filter in step two h Represents the convergence factor, μ, of the update of the auxiliary filter in step four w2 Representing the convergence factor updated by the noise control filter in step five, and x represents the reference signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211186487.8A CN115565515A (en) | 2022-09-27 | 2022-09-27 | Step-by-step virtual sensing noise reduction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211186487.8A CN115565515A (en) | 2022-09-27 | 2022-09-27 | Step-by-step virtual sensing noise reduction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115565515A true CN115565515A (en) | 2023-01-03 |
Family
ID=84742881
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211186487.8A Pending CN115565515A (en) | 2022-09-27 | 2022-09-27 | Step-by-step virtual sensing noise reduction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115565515A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116959472A (en) * | 2023-09-20 | 2023-10-27 | 武汉海微科技有限公司 | Active noise reduction method, device, equipment and storage medium |
-
2022
- 2022-09-27 CN CN202211186487.8A patent/CN115565515A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116959472A (en) * | 2023-09-20 | 2023-10-27 | 武汉海微科技有限公司 | Active noise reduction method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| George et al. | Advances in active noise control: A survey, with emphasis on recent nonlinear techniques | |
| Kuo et al. | Active noise control: Open problems and challenges | |
| Chang et al. | Active noise control in headsets by using a low-cost microcontroller | |
| Shi et al. | Block coordinate descent based algorithm for computational complexity reduction in multichannel active noise control system | |
| CN111627415A (en) | Active noise reduction device based on self-adaptive MFxLMS algorithm and FPGA implementation | |
| CN101345783B (en) | System and method for noise elimination | |
| Das et al. | A computationally efficient frequency-domain filtered-X LMS algorithm for virtual microphone | |
| Mostafavi et al. | Deep learning-based active noise control on construction sites | |
| Zhang et al. | Deep MCANC: A deep learning approach to multi-channel active noise control | |
| CN105913836B (en) | A kind of fixed point real-time noise-reducing method based on DSP | |
| Shi et al. | Computation-efficient solution for fully-connected active noise control window: Analysis and implementation of multichannel adjoint least mean square algorithm | |
| CN115565515A (en) | Step-by-step virtual sensing noise reduction method | |
| Hu et al. | Invisible speakers in home windows for simultaneous auxiliary audio playback and active noise cancellation | |
| Kuo et al. | Adaptive algorithms and experimental verification of feedback active noise control systems | |
| Ren et al. | An improved artificial bee colony algorithm for model-free active noise control: algorithm and implementation | |
| Sahib et al. | Comparison of performance and computational complexity of nonlinear active noise control algorithms | |
| Roy et al. | Active noise control using filtered-xLMS and feedback ANC filter algorithms | |
| CN110808025A (en) | Active noise control system modular design method based on FPGA | |
| Xie et al. | A feedforward and feedback composite active noise reduction headset based on inverse filter frequency equalization and its DSP system implementation | |
| CN113096629A (en) | Relative path virtual sensing method for single-channel feedback active noise control system | |
| Kuo et al. | Active noise control systems with optimized secondary path | |
| Luo et al. | EC-ANC: Edge Case-Enhanced Active Noise Cancellation for True Wireless Stereo Earbuds | |
| Behuria et al. | A novel variable step-size feedback Filtered-X LMS algorithm for acoustic noise removal | |
| Yuan et al. | Design and hardware realization of active noise control system based on momentum LMS algorithm | |
| Yuan et al. | Design of Active Noise Control System Based on FPGA |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |