CN117912440A - Active noise reduction method and system for automobile and storable medium - Google Patents
Active noise reduction method and system for automobile and storable medium Download PDFInfo
- Publication number
- CN117912440A CN117912440A CN202410146570.5A CN202410146570A CN117912440A CN 117912440 A CN117912440 A CN 117912440A CN 202410146570 A CN202410146570 A CN 202410146570A CN 117912440 A CN117912440 A CN 117912440A
- Authority
- CN
- China
- Prior art keywords
- signal
- vehicle
- noise
- noise reduction
- original
- 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
- 230000009467 reduction Effects 0.000 title claims abstract description 94
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000013145 classification model Methods 0.000 claims abstract description 32
- 239000013598 vector Substances 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 12
- 230000005236 sound signal Effects 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 9
- 238000012706 support-vector machine Methods 0.000 claims description 8
- 230000005284 excitation Effects 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 238000005316 response function Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 abstract description 3
- 238000012360 testing method Methods 0.000 description 10
- 238000012549 training Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000013461 design Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000011900 installation process Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Soundproofing, Sound Blocking, And Sound Damping (AREA)
- Fittings On The Vehicle Exterior For Carrying Loads, And Devices For Holding Or Mounting Articles (AREA)
Abstract
The invention discloses an active noise reduction method, an active noise reduction system and a storage medium for an automobile, wherein the method comprises the steps that an error microphone is arranged in the automobile, the error microphone is used for acquiring an in-automobile noise signal in real time, and the method further comprises the steps of establishing a required model in an off-line mode, wherein the model comprises a secondary path identification model, a road classification model and an in-automobile target acoustic signal model; and acquiring an in-vehicle noise signal acquired by the error microphone in real time. And calling a secondary path identification model to calculate an original noise signal in the vehicle according to the noise signal in the vehicle. Extracting the characteristic vector of the original noise signal in the vehicle, and calling the sound frequency of the target signal by combining the acquired vehicle speed. And processing the original noise signal in the vehicle, and predicting the noise reduction signal of the error microphone position by combining the target signal sound frequency. And inputting the predicted noise reduction signal into a secondary path identification model to obtain a loudspeaker output signal. The method can realize the fusion and noise reduction of various types of noise without a reference sensor, and can control the quality of noise in the vehicle after noise reduction.
Description
Technical Field
The invention relates to the technical field of noise control of automobiles, in particular to an active noise reduction method and system for automobiles and a storable medium.
Background
The active noise control technology is an active control method, and is to send out a cancellation signal through a loudspeaker to realize noise attenuation at a target position. The active noise reduction system of the automobile generally uses a feedforward active noise reduction algorithm to process a reference signal and an error signal in real time, generate a noise cancellation signal to drive a loudspeaker to sound, and reduce the noise level of a target point.
In the traditional active noise reduction system, the target noise sources are different, and the mode of acquiring the reference signals is also different. For example, the engine actively reduces noise, obtains an engine rotating speed signal through CAN, and synthesizes a reference signal; the road noise actively reduces noise, and an acceleration sensor is arranged on a vehicle body or a chassis to acquire a reference acceleration signal; the wind noise actively reduces noise, and a microphone is arranged outside or inside the vehicle body to acquire a reference sound signal. Therefore, different reference signals need to be used for different noise sources, which means that more types of noise in the vehicle are reduced, and the number and types of reference sensors are increased. The existence of the reference sensor not only makes the system more complex, but also causes higher cost of the active noise reduction system, which hinders the popularization of the system. In addition, the existing active noise reduction system is designed according to the objective of reducing the noise in the vehicle to 0, but in practical engineering application, due to the influence of factors such as system hardware, a secondary path and the like, the active noise reduction system cannot reduce the noise in the vehicle to 0, and only a part of the noise in the vehicle can be reduced, so that the quality of the noise in the vehicle after noise reduction is difficult to control. Therefore, under the condition of reducing or not using a reference sensor, the method has extremely important significance for the development of an active noise reduction system, and can fusion noise reduction of various types of noise and control the quality of the noise in the vehicle after noise reduction.
Disclosure of Invention
In order to overcome the defects, the invention aims to provide an active noise reduction method, an active noise reduction system and a storage medium for an automobile, which can realize fusion noise reduction of various types of noise without using a reference sensor and can control the quality of noise in the automobile after noise reduction compared with the traditional self-adaptive FxLMS noise reduction system.
In order to achieve the above purpose, the invention adopts the following technical scheme: an active noise reduction method for an automobile, wherein at least one error microphone is arranged in the automobile and used for acquiring an in-automobile noise signal in real time, and the method comprises the following steps:
The method comprises the steps of establishing a required model offline, wherein the model comprises a secondary path identification model, a road classification model and an in-vehicle target acoustic signal model, and the in-vehicle target acoustic signal model is an in-vehicle target acoustic signal data set which is mapped one by vehicle speed, road surface type and target signal acoustic frequency;
Acquiring an in-vehicle noise signal acquired by the error microphone in real time;
calling the secondary path identification model to calculate an original noise signal in the vehicle according to the noise signal in the vehicle;
Extracting a feature vector of the original noise signal in the vehicle, inputting the road classification model by combining the acquired vehicle speed to obtain road classification, and calling the target signal sound frequency in the vehicle target sound signal data set according to the road classification and the current vehicle speed;
Processing the original noise signal in the vehicle, and predicting a noise reduction signal of the position of the error microphone by combining the sound frequency of the target signal;
and inputting the noise reduction signal of the predicted error microphone position into a secondary path identification model to obtain an output signal of the loudspeaker.
The invention has the beneficial effects that:
1. the noise reduction signal is predicted directly by using the noise signal in the vehicle, so that the use of an additional reference sensor is avoided, the hardware cost of the system can be reduced, and the design and the installation process of the system are simplified.
2. The current road surface type is classified through the road classification model, and corresponding target in-vehicle signals are called according to the classification result, so that corresponding noise reduction strategies can be provided for different types of noise, and simultaneous noise reduction of various types of noise such as engine noise, road noise, wind noise and the like is realized. By effectively controlling different types of noise, the quality of the sound environment in the vehicle can be remarkably improved.
3. The in-car target acoustic signal model is added, so that the active control of in-car noise can be realized, and the subjective feeling of a user is met by adjusting the in-car target acoustic signal model, so that a more comfortable and pleasant in-car acoustic environment is obtained.
4. The noise reduction signal is predicted directly by using the in-vehicle noise signal, so that complex self-adaptive algorithm and a large amount of real-time calculation are avoided. The complexity of the method is reduced, and the real-time performance and response speed of the method are improved.
Further, processing the original noise signal in the vehicle, and predicting the noise reduction signal at the position of the error microphone by combining the target signal acoustic frequency specifically includes:
Collecting in-vehicle original noise signals at N time and in-vehicle original noise signals of front N-1 sampling points to form a sequence Y (N), wherein Y (N) = [ Y (N-N), Y (N-n+1),. The Y (N) ], and Y (N) is the in-vehicle original noise signal at the N time;
Performing time-frequency conversion on each in-car original noise signal in the sequence Y (N) to obtain an amplitude A (f) and an initial phase Ph (f) of each frequency f of Y (N);
After predicting the delay time α seconds, the phase Ph' (f) of the original noise signal y Original, original (f) in the vehicle at each frequency f, where α is the signal delay time between the error microphone and the speaker;
calculating the phase Ph Counteracting each other (f) of the cancellation signal y Counteracting each other (f) at each frequency at the error microphone location from the target signal acoustic frequency f Target object and the phase Ph' (f);
From the phase Ph Counteracting each other (f) of the cancellation signal y Counteracting each other (f), a noise reduction signal y Counteracting each other (n) of the error microphone position is calculated.
Because the output signal of the current N-time loudspeaker is noise-reduced, the noise signal in the automobile is collected by the error microphone after alpha seconds, the noise signal in the automobile, collected by the error microphone after alpha seconds, is predicted through the current N-time and N points before the N-time, and then the output signal of the current N-time loudspeaker is reversely deduced through the predicted noise signal in the automobile after alpha seconds.
Further, the calculation formula of the phase Ph Counteracting each other (f) is:
Ph Counteracting each other (f)＝arccos[y Original, original (f)-D(f Target object ,vn,typen)/A(f)]；
Wherein y Original, original (f)＝A(f)*cos(Ph'(f)),D(f Target object ,vn,typen) is an in-vehicle target sound signal dataset, v n is the vehicle speed at time n, type n is the eigenvector of an in-vehicle original noise signal at time n, and f Target object is the target signal sound frequency at time n.
Because the original noise signal y Original, original (f) is set according to the noise drop in the cabin as 0, but in actual use, the application defines the corresponding target signal frequency according to the road surface type of the vehicle speed. The data set D (f Target object , v, type) is thus called upon to find the desired target signal acoustic frequency f Target object to calculate the phase Ph Counteracting each other (f) of the cancellation signal y Counteracting each other (f) at the error microphone location.
Further, the calculation formula of the noise reduction signal y Counteracting each other (n) is as follows:
Wherein f max and f min are the maximum and minimum, respectively, of frequency f in sequence Y (N).
Further, the calculation formula of the output signal of the speaker is:
Wherein S (n) is an output signal of the speaker at the nth moment, L is a positive integer and not more than L, L is the length of the time domain FIR filter in the secondary path recognition model, and h (L) is a coefficient of the time domain FIR filter.
Since the noise reduction signal is a noise reduction signal obtained at the error microphone position after the signal output from the speaker passes through the secondary path, it is also necessary to convert the noise reduction signal y Counteracting each other (n) at the error microphone position into the output signal of the speaker through the secondary path.
Further, the original noise signal in the vehicle at the nth moment is equal to the convolution of the noise signal in the vehicle at the nth moment and the secondary path in the secondary path identification model of the output signal of the loudspeaker at the nth-alpha moment;
the calculation formula of the original noise signal y (n) in the vehicle at the nth time is as follows:
Wherein E (n) is an in-vehicle noise signal at the nth time.
Since the in-vehicle noise signal E (n) is the original noise signal y (n) minus the signal S (n- α) output by the speaker, the original noise signal y (n) is equal to the convolution of the in-vehicle noise signal E (n) plus the signal S (n- α) output by the speaker via the secondary path.
Further, establishing the secondary path identification model includes using a white noise signal as excitation, performing offline identification on an impulse response function of the secondary path by using an LMS algorithm, and identifying a time domain of a finite unit impulse response to be a coefficient h (L) of a time domain FIR filter, wherein L is a positive integer and not more than L, and L is the length of the time domain FIR filter.
And carrying out secondary path identification by using an LMS algorithm, and finally obtaining a finite unit impulse response of the secondary path by iteratively calculating the error between the loudspeaker excitation signal and the error microphone receiving signal for subsequent design and control of the active noise reduction system.
Furthermore, the road classification model is used for predicting the type of the vehicle surface where the vehicle is located according to the vehicle speed and the feature vector, and a support vector machine is adopted to construct the road classification model.
The noise levels in the vehicles on different roads are different, so that the road types can be identified according to the noise signals in the vehicles, and under the off-line state, a road classification model is established to realize the mapping relationship of the vehicle speed, the feature vectors of the noise signals in the vehicles and the road types.
The invention also discloses an active noise reduction system of the automobile, which comprises:
the error microphone is used for acquiring in-vehicle noise signals in real time;
the digital signal processor adopts the noise reduction method to generate an output signal of the loudspeaker;
a power amplifier for power amplifying an output signal;
and the loudspeaker sounds under the drive of the power amplifier.
Only the error microphone is needed, noise reduction can be carried out on the space in the vehicle, and the cost is reduced. Meanwhile, the system is compatible with various types of noise and can reduce noise simultaneously, and the sound quality can be controlled, so that the active noise reduction system is more practical and efficient, and can provide a better in-vehicle sound environment.
The invention also discloses a computer storage medium, wherein the computer storage medium is stored with a computer program, and the computer program realizes the active noise reduction method of the automobile when being executed by a processor.
Drawings
FIG. 1 is a flowchart showing a first embodiment of the present invention;
FIG. 2 is a schematic diagram of a secondary path recognition by a secondary path recognition model according to an embodiment of the present invention;
FIG. 3 is a second flowchart of a first embodiment of the present invention;
FIG. 4 is a third flowchart of a first embodiment of the present invention;
FIG. 5 is a system block diagram of a second embodiment of the present invention;
fig. 6 is a block diagram of a digital signal processor according to a second embodiment of the present invention.
Detailed Description
The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, but not all embodiments, and all other embodiments obtained by those skilled in the art without making any inventive effort based on the embodiments of the present invention are within the scope of protection of the present invention. Examples
Fig. 1 is a schematic flow chart of an embodiment of an active noise reduction method for an automobile according to the present application.
The first embodiment of the invention provides an active noise reduction method for an automobile, wherein at least one error microphone is arranged in the automobile, the error microphone is used for acquiring noise signals in the automobile in real time, and the active noise reduction method for the automobile comprises the following steps:
And 100, establishing a required model offline, wherein the model comprises a secondary path identification model, a road classification model and an in-vehicle target acoustic signal model.
The method comprises the steps of establishing a secondary path identification model, namely taking a white noise signal as excitation, adopting an LMS algorithm to perform off-line identification on an impact response function of a secondary path to obtain a coefficient h (L) of a time domain FIR (Finite Impulse Response finite length unit impulse response) filter, wherein L is a positive integer and not more than L, and L is the length of the time domain FIR filter, namely L is 1,2,3. h (L) is a finite length sequence of L points long, and the output of the time domain FIR filter is the linear convolution of the input sequence with the coefficient h (L).
The secondary path is the air path from the speaker to the error microphone, and in this embodiment, a conventional LMS (LEAST MEAN square least mean square) algorithm is used for secondary path identification.
FIG. 2 shows a schematic diagram of secondary path recognition by a stage path recognition model, in which the white noise signal v (n) is the excitation signal of the loudspeaker at time n, and the signal at the time of transmission of the white noise signal v (n) to the error microphone position is d (n), the transfer function of v (n) through the secondary pathAnd if the output is y v (n), the identification error e 0(n)＝d(n)-yv (n) of the system is processed by continuous iteration of the system, so that the power of the identification error e 0 (n) is close to 0, and the secondary path identification is completed, so that a secondary path identification model is formed. The identified finite unit impulse response corresponds to the estimated function/>The time domain of which is denoted as the coefficient h (l) of the FIR filter.
Referring to fig. 2, the offline identification of the impact response function of the secondary path by using the LMS algorithm specifically includes:
estimating function of secondary channel transfer function Initializing;
Processing the white noise signal v (n) through a secondary channel S (z) to obtain a desired signal d (n);
Passing white noise signal v (n) through secondary channel estimation function Processing to obtain an estimated signal y v (n);
Performing difference processing on the expected signal d (n) and the estimated signal y v (n) to obtain an identification error e 0 (n);
Judging whether the identification error e 0 (n) converges to a set minimum value, wherein the set minimum value approaches zero;
if the estimated signal y v (n) and the desired signal d (n) approach the same, the estimated function is known Approaching the secondary channel S (z), an off-line identification of the secondary channel is achieved.
Otherwise, the identification error e 0 (n) and the white noise signal v (n) are processed together by the LMS algorithm, and the processed result is output to the estimation function
Estimating a functionThe updated estimation function/>, obtained by using the result processed by the LMS algorithm and the white noise signal v (n)Until the recognition error e 0 (n) converges to the set minimum value.
In this embodiment, the LMS algorithm is used to perform off-line identification of the secondary path, and the error between the speaker excitation signal and the error microphone receiving signal is calculated through iteration, so as to finally obtain the finite unit impulse response of the secondary path, which is used for the subsequent design and control of the active noise reduction system. Those skilled in the art should know how to perform the secondary path recognition in the offline state through the LMS algorithm, and detailed description thereof will be omitted herein. The offline identification is separated from the active noise reduction system, and the system identification is independently carried out, so that the operation load is not increased on the active noise reduction system, and the robustness of the active noise reduction system is not damaged.
In the running process of the vehicle, the noise levels in the vehicles passing through different roads are different, so that the recognition of the road types can be completed according to the noise signals in the vehicles, namely, a road classification model is established, and the road classification model is used for realizing the mapping relation between the vehicle speed, the characteristic vectors of the noise signals in the vehicles and the road types.
There are many methods for building the road classification model, and in one embodiment, a support vector machine (SVM support vector machine classification model) is used to build the road classification model, where the support vector machine is a classification model, and the objective of the support vector machine is to find an optimal hyperplane to separate data samples of different classes. Here, a support vector machine is used to learn and establish the relationship between the vehicle speed and the feature vector and the road surface type.
Fig. 3 shows a flow of building a road classification model, which specifically includes:
and acquiring a data set, and acquiring different vehicle speeds, road surfaces and corresponding noise signals in the vehicle to form the data set.
The data set is divided into a training set and a test set.
Extracting signal characteristics of the training set and the testing set, and forming the signal characteristics into characteristic vectors.
And inputting the training set into a support vector machine for model training.
And adopting the test set to perform accuracy test of the model, and judging whether the accuracy test passes or not.
And when the accuracy test passes, fixing the road classification model.
The road classification model is Mode (v, vec, type), v is the vehicle speed, vec is the feature vector, and type is the road surface type. The elements in the feature vector vec are time domain signal feature values, such as average values, peak values and the like, and the determination of the feature vector vec can be selected according to the model classification accuracy.
When the accuracy test of the model by the test set fails, the model parameters are required to be adjusted or new feature vectors are required to be used for model training again until the accuracy test of the model by the test set meets the requirements.
In this embodiment, a road classification model is pre-established in an offline state, and the type of the vehicle surface where the vehicle is located is predicted according to the vehicle speed and the feature vector.
The target sound signal model in the car is designed artificially, and the target sound frequency refers to the processed ideal noise signal in the car after the active noise reduction system works. The in-vehicle target sound signal model is an in-vehicle target sound signal data set which is mapped one by the speed, the road surface type and the target signal sound frequency. The target signal presents amplitude values corresponding to different frequencies in a frequency domain, and is required to be designed artificially to ensure that sound has good sense in hearing.
Different vehicle speeds and road types may correspond to different in-vehicle target sounds. This means that the spectral characteristics of the target sound in the vehicle will vary at different vehicle speeds and road conditions. Therefore, the data set D (f Target object , v, type) of the target sound in the vehicle can be constructed according to the vehicle speed, the road surface type and the target signal sound frequency, and one target signal sound frequency f Target object can be determined according to the vehicle speed and the road surface type to generate the noise signal in the vehicle which is suitable for hearing feeling, so that the optimization and the performance improvement of the active noise reduction system are realized.
And 200, acquiring an in-vehicle noise signal acquired by the error microphone in real time.
The error microphone may be provided according to the condition of the cabin, at least one being provided, typically near the human ear. In one embodiment, an error microphone may be provided for each seat near the ear.
And 300, calling a secondary path identification model to calculate an original noise signal in the vehicle according to the noise signal in the vehicle.
At the nth time, the in-vehicle noise signal is E (n), the vehicle speed at this time is v n, and the speaker output signal is S (n) at this time. The output signal of the loudspeaker reaches the position of the error microphone after being delayed by the secondary path, so that the noise in the vehicle is reduced, and the delay time of the whole process is alpha seconds, namely the signal delay time between the error microphone and the loudspeaker. The delay time alpha is a definite value after the system is fixed, and is not changed and can be known in advance. The in-vehicle noise signal E (n) is thus generated after the output signal S (t) of the loudspeaker before a second, where t=n-a.
The in-vehicle noise signal E (n) is a signal after noise reduction by the noise reduction system, and therefore, the in-vehicle original noise signal y (n) needs to be obtained according to the noise-reduced in-vehicle noise signal E (n), and the in-vehicle original noise signal y (n) is a signal before noise reduction. Since the in-vehicle noise signal E (n) is the original noise signal y (n) minus the signal S (n- α) output by the speaker, the original noise signal y (n) is equal to the convolution of the in-vehicle noise signal E (n) plus the signal S (n- α) output by the speaker via the secondary path. That is:
wherein is a positive integer and L is not greater than L, i.e., L is 1,2,3.
And 400, extracting the feature vector of the original noise signal in the vehicle, and respectively inputting a road classification model and a target sound signal model in the vehicle to obtain the sound frequency of the target signal by combining the acquired vehicle speed. The method specifically comprises the following steps:
Firstly extracting feature vectors of original noise signals in a vehicle, combining the acquired vehicle speeds, and inputting the feature vectors into a road classification model to obtain road classification.
And then, according to the road classification and the current car speed, the target signal sound frequency in the car interior target sound signal data set is called.
Extracting the characteristic vector Vec n of the original noise signal y (n) to form the characteristic vector Vec n of the nth moment, acquiring the vehicle speed v n of the current nth moment by the vehicle-mounted CAN bus, inputting Vec n and v n into a road surface classification model Mode (v, vec, type) in the step 100, and determining the road surface type n of the current nth moment because Vec n and v n are already determined.
After the road surface type n at the current time n is determined, the corresponding data set D (f Target object , v, type) can be called by combining the vehicle speed v n at the current time n, and the target signal sound frequency f Target object at the current time n is determined. After this step, the original noise signal y (n) at time n corresponds to the target signal sound frequency f Target object one by one.
And 500, processing original noise signals in the vehicle, and predicting noise reduction signals of the position of the error microphone by combining the sound frequency of the target signals.
Fig. 4 shows a specific flow of step 500, where processing the original noise signal in the vehicle, and predicting the noise reduction signal of the error microphone position in combination with the target signal acoustic frequency specifically includes:
And 51, collecting in-vehicle original noise signals at N moments and forming a sequence Y (N) by the in-vehicle original noise signals of the previous N-1 sampling points. Wherein Y (N) = [ Y (N-N), Y (N-n+1),. The term, Y (N) ].
And 52, performing time-frequency conversion on each in-vehicle original noise signal in the sequence Y (N) to obtain an amplitude A (f) and an initial phase Ph (f) of each frequency f of Y (N).
Step 53, predicting the phase Ph' (f) of the original noise signal y Original, original (f) in the vehicle at each frequency f after the delay time α seconds.
Ph' (f) =2pi f (α+β) +ph (f), where β is the sampling time corresponding to the nth sampling point.
Step 54, calculating the phase Ph Counteracting each other (f) of the cancellation signal y Counteracting each other (f) at each frequency f at the error microphone location based on the target signal acoustic frequency f Target object and the phase Ph' (f).
Because of y Original, original (f)-y Counteracting each other (f)＝D(f Target object ,vn,typen), then the phase Ph Counteracting each other (f)＝arccos[y Original, original (f)-D(f Target object ,vn,typen)/a (f) of the signal y Counteracting each other (f), where y Original, original (f) =a (f) cos (Ph' (f)).
Because the original noise signal y Original, original (f) is set according to the noise drop in the cabin as 0, but in actual use, the application defines the corresponding target signal frequency according to the road surface type of the vehicle speed. The data set D (f Target object , v, type) is thus called upon to find the desired target signal acoustic frequency f Target object to calculate the phase Ph Counteracting each other (f) of the cancellation signal y Counteracting each other (f) at the error microphone location.
Step 55, calculating the noise reduction signal y Counteracting each other (n) of the error microphone position according to the phase Ph Counteracting each other (f) of the cancellation signal y Counteracting each other (f).
Where f max and f min are the upper and lower limits of the control frequency, f max and f min, respectively, i.e. the maximum and minimum values of frequency f in sequence Y (N). The noise reduction signal y Counteracting each other (n) is a noise reduction signal of the error microphone position after n+α predicted at the nth time.
Step 600, inputting the noise reduction signal of the predicted error microphone position into a secondary path identification model to obtain an output signal of the loudspeaker.
Since the noise reduction signal obtained in step 500 is a noise reduction signal obtained at the error microphone position after the signal output from the speaker passes through the secondary path, it is also necessary to convert the noise reduction signal y Counteracting each other (n) at the error microphone position in step 500 into the output signal of the speaker through the secondary path.
Because ofY Counteracting each other (n) has been calculated in step 55, so the output signal of the speaker at the current time n is calculated as S (n).
Because the output signal of the current N-time loudspeaker is noise-reduced, the noise signal in the automobile is collected by the error microphone after alpha seconds, the noise signal in the automobile, collected by the error microphone after alpha seconds, is predicted through the current N-time and N points before the N-time, and then the output signal of the current N-time loudspeaker is reversely deduced through the predicted noise signal in the automobile after alpha seconds. The output signal of the speaker at the current n time will produce a noise reduction effect at the error microphone after a second.
The output signal of the speaker in step 500 is amplified by the power amplifier and then outputted.
Conventional active noise reduction systems typically require the use of a reference sensor to obtain a reference value of the noise signal in order to generate the noise reduction signal. However, in the present embodiment, the prediction of the noise reduction signal is performed directly using the in-vehicle noise signal, avoiding the use of an additional reference sensor. This reduces the hardware cost of the system and simplifies the design and installation process of the system. Meanwhile, in the embodiment, an in-vehicle target acoustic signal model is added, so that the active control of in-vehicle noise can be realized, and the subjective feeling of a user is met by adjusting the in-vehicle target acoustic signal model, so that a more comfortable and pleasant in-vehicle acoustic environment is obtained. The noise reduction signal is predicted directly by using the in-vehicle noise signal, so that complex self-adaptive algorithm and a large amount of real-time calculation are avoided. The complexity of the method is reduced, and the real-time performance and response speed of the method are improved. The active noise reduction method in the embodiment is more practical and efficient, and can provide a better in-vehicle sound environment.
The second embodiment of the application provides an active noise reduction system for an automobile. Fig. 5 is a block diagram of an embodiment of an active noise reduction system for an automobile according to the present application.
An active noise reduction system for an automobile includes an error microphone, a Digital Signal Processor (DSP), a power Amplifier (AMP), and a speaker. The error microphone is used for collecting in-vehicle noise signals in real time, the digital signal processor is used for executing the method steps in the first embodiment to generate output signals of the loudspeaker, and the power amplifier is used for amplifying the output signals and driving the loudspeaker to generate noise cancellation.
The error microphone and the loudspeaker are at least one, and the error microphone is arranged near the human ear.
Fig. 6 is a block diagram of an embodiment of a digital signal processor according to the present application. The digital signal processor comprises a stage path identification module, a pavement classification model training module, an in-car target sound signal design module and an active noise reduction algorithm main body module. The method comprises the steps of performing off-line processing on a stage path identification module, a pavement classification model training module and an in-vehicle target sound signal design module, namely storing the designed model in a corresponding module in an off-line state, performing on-line execution on an active noise reduction algorithm main body module, collecting in-vehicle noise signals of an error microphone and acquired output signals of a loudspeaker in real time, and predicting the output signals of the loudspeaker at the current moment.
The system comprises a secondary path identification module, a road classification model training module, an in-vehicle target sound signal design module and an in-vehicle target sound signal design module, wherein the secondary path identification model which is built offline is stored in the secondary path identification module, the road classification model training module is used for training the road classification model and storing the trained road classification model, and the in-vehicle target sound signal design module is used for storing the designed in-vehicle target sound signal model.
The main body module of the active noise reduction algorithm is a core improvement point of the system and is used for acquiring an in-vehicle noise signal acquired by an error microphone in real time; according to the noise signal in the vehicle, invoking a secondary path identification model to calculate an original noise signal in the vehicle; extracting a feature vector of an original noise signal in a vehicle, inputting a road classification model by combining the acquired vehicle speed to obtain road classification, and calling a target signal sound frequency in a target sound signal data set in the vehicle according to the road classification and the current vehicle speed; processing original noise signals in the vehicle, and predicting noise reduction signals of the position of the error microphone by combining the sound frequency of the target signals; and inputting the noise reduction signal of the predicted error microphone position into a secondary path identification model to obtain an output signal of the loudspeaker.
A third embodiment of the present invention provides a storable medium having stored thereon a computer program which, when executed by a processor, implements steps such as in an active noise reduction method for an automobile.
In the foregoing embodiments of the present application, it should be understood that the disclosed method, apparatus, computer readable storage medium and electronic device may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be additional divisions when actually implemented, for example, multiple components or modules may be combined or integrated into another apparatus, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with respect to each other may be an indirect coupling or communication connection via some interfaces, devices or components or modules, which may be in electrical, mechanical, or other forms.
The integrated modules, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a read-only memory (ROM), a random access memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and to implement the same, but are not intended to limit the scope of the present invention, and all equivalent changes or modifications made according to the spirit of the present invention should be included in the scope of the present invention.
Claims (10)
1. An automobile active noise reduction method is characterized in that at least one error microphone is arranged in an automobile and used for acquiring an in-automobile noise signal in real time, and the method is characterized in that: the method comprises the following steps:
The method comprises the steps of establishing a required model offline, wherein the model comprises a secondary path identification model, a road classification model and an in-vehicle target acoustic signal model, and the in-vehicle target acoustic signal model is an in-vehicle target acoustic signal data set which is mapped one by vehicle speed, road surface type and target signal acoustic frequency;
Acquiring an in-vehicle noise signal acquired by the error microphone in real time;
calling the secondary path identification model to calculate an original noise signal in the vehicle according to the noise signal in the vehicle;
Extracting a feature vector of the original noise signal in the vehicle, inputting the road classification model by combining the acquired vehicle speed to obtain road classification, and calling the target signal sound frequency in the vehicle target sound signal data set according to the road classification and the current vehicle speed;
Processing the original noise signal in the vehicle, and predicting a noise reduction signal of the position of the error microphone by combining the sound frequency of the target signal;
and inputting the noise reduction signal of the predicted error microphone position into a secondary path identification model to obtain an output signal of the loudspeaker.
2. The method of active noise reduction for an automobile of claim 1, wherein: processing original noise signals in the vehicle, and predicting noise reduction signals of the position of the error microphone by combining the sound frequency of the target signals specifically comprises the following steps:
Collecting in-vehicle original noise signals at N time and in-vehicle original noise signals of front N-1 sampling points to form a sequence Y (N), wherein Y (N) = [ Y (N-N), Y (N-n+1),. The Y (N) ], and Y (N) is the in-vehicle original noise signal at the N time;
Performing time-frequency conversion on each in-car original noise signal in the sequence Y (N) to obtain an amplitude A (f) and an initial phase Ph (f) of each frequency f of Y (N);
After predicting the delay time α seconds, the phase Ph' (f) of the original noise signal y Original, original (f) in the vehicle at each frequency f, where α is the signal delay time between the error microphone and the speaker;
calculating the phase Ph Counteracting each other (f) of the cancellation signal y Counteracting each other (f) at each frequency at the error microphone location from the target signal acoustic frequency f Target object and the phase Ph' (f);
From the phase Ph Counteracting each other (f) of the cancellation signal y Counteracting each other (f), a noise reduction signal y Counteracting each other (n) of the error microphone position is calculated.
3. The method of active noise reduction for an automobile of claim 2, wherein: the calculation formula of the phase Ph Counteracting each other (f) is as follows:
Ph Counteracting each other (f)＝arccos[y Original, original (f)-D(f Target object ,vn,typen)/A(f)]；
Wherein y Original, original (f)＝A(f)*cos(Ph'(f)),D(f Target object ,vn,typen) is an in-vehicle target sound signal dataset, v n is the vehicle speed at time n, type n is the eigenvector of an in-vehicle original noise signal at time n, and f Target object is the target signal sound frequency at time n.
4. The method of active noise reduction for an automobile of claim 2, wherein: the calculation formula of the noise reduction signal y Counteracting each other (n) is as follows:
Wherein f max and f min are the maximum and minimum, respectively, of frequency f in sequence Y (N).
5. The method of active noise reduction for an automobile of claim 2, wherein: the calculation formula of the output signal of the loudspeaker is as follows:
Wherein S (n) is an output signal of the speaker at the nth moment, L is a positive integer and not more than L, L is the length of the time domain FIR filter in the secondary path recognition model, and h (L) is a coefficient of the time domain FIR filter.
6. The method of active noise reduction for a vehicle of claim 5, wherein: the original noise signal in the vehicle at the nth moment is equal to the convolution of the noise signal in the vehicle at the nth moment and the secondary path in the secondary path identification model of the output signal of the loudspeaker at the nth-alpha moment;
the calculation formula of the original noise signal y (n) in the vehicle at the nth time is as follows:
wherein E (n) is the noise signal in the vehicle acquired by the error microphone at the nth moment.
7. The method for actively reducing noise in an automobile according to any one of claims 1-6, wherein: the method comprises the steps of establishing a secondary path identification model, namely taking a white noise signal as excitation, adopting an LMS algorithm to carry out off-line identification on an impact response function of a secondary path, and identifying a time domain of a finite unit impulse response to obtain a coefficient h (L) which is expressed as a time domain FIR filter, wherein L is a positive integer and is not more than L, and L is the length of the time domain FIR filter.
8. The method of active noise reduction for an automobile of claim 1, wherein: the road classification model is used for predicting the type of the vehicle surface where the vehicle is located according to the vehicle speed and the feature vector, and a support vector machine is adopted to construct the road classification model.
9. An active noise reduction system for an automobile is characterized in that: comprising
The error microphone is used for acquiring in-vehicle noise signals in real time;
A digital signal processor employing the noise reduction method of any one of claims 1-8 to generate an output signal of a speaker;
a power amplifier for power amplifying an output signal;
and the loudspeaker sounds under the drive of the power amplifier.
10. A computer-storable medium, characterized by: the computer-storable medium has stored thereon a computer program which, when executed by a processor, implements the method of actively noise reduction of a vehicle according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410146570.5A CN117912440A (en) | 2024-02-01 | 2024-02-01 | Active noise reduction method and system for automobile and storable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410146570.5A CN117912440A (en) | 2024-02-01 | 2024-02-01 | Active noise reduction method and system for automobile and storable medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117912440A true CN117912440A (en) | 2024-04-19 |
Family
ID=90683740
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410146570.5A Pending CN117912440A (en) | 2024-02-01 | 2024-02-01 | Active noise reduction method and system for automobile and storable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117912440A (en) |
-
2024
- 2024-02-01 CN CN202410146570.5A patent/CN117912440A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4283212B2 (en) | Noise removal apparatus, noise removal program, and noise removal method | |
| KR102487160B1 (en) | Audio signal quality enhancement based on quantitative signal-to-noise ratio analysis and adaptive wiener filtering | |
| Sano et al. | Active control system for low-frequency road noise combined with an audio system | |
| EP1580882B1 (en) | Audio enhancement system and method | |
| CN109747575B (en) | Multi-mode in-vehicle sound quality optimization system based on order optimization | |
| CN110120217B (en) | Audio data processing method and device | |
| CN112509546B (en) | Active noise reduction system for dynamic change sound field of automobile | |
| JP7186375B2 (en) | Speech processing device, speech processing method and speech processing system | |
| US20150010162A1 (en) | Systems and methods for optimizing an audio communication system | |
| CN111833841A (en) | Active control system and method for automobile road noise and vehicle system | |
| US5293425A (en) | Active noise reducing | |
| CN111724761B (en) | Vehicle-mounted active noise control device and method | |
| CN114582312B (en) | Active control method and system for anti-interference adaptive road noise in vehicle | |
| CN114677997A (en) | Real vehicle active noise reduction method and system based on acceleration working condition | |
| CN114464157A (en) | Active noise reduction method and device for vehicle and storage medium | |
| CN117912440A (en) | Active noise reduction method and system for automobile and storable medium | |
| CN116052630A (en) | Active control system and method for automobile road noise | |
| JP2003216163A (en) | Noise controller | |
| JP2002236497A (en) | Noise reduction system | |
| CN111833840A (en) | Noise reduction method and device, system, electronic equipment and storage medium | |
| JP2010202073A (en) | Apparatus and method for controlling noise | |
| KR20180102914A (en) | Infotainment system with noise canceling using neural network learning algorithm and control method thereof | |
| WO2019241657A1 (en) | Concurrent fxlms system with common reference and error signals | |
| JP3617079B2 (en) | Active noise control device and active vibration control device | |
| CN111418003B (en) | Active noise control method and system |
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 |