In-vehicle noise reduction method, device, equipment, storage medium and product

Through adaptive filtering algorithms and feature extraction mapping technology, filter weight coefficients are generated and dynamically adjusted, which solves the problem of poor low-frequency noise control effect of traditional in-vehicle noise reduction technology, and achieves efficient noise reduction in complex noise environments. Improves the quietness and ride comfort of the interior environment.

CN119049444AActive Publication Date: 2024-11-29DONGFENG MOTOR GRP

Patent Information

Application Number
CN202411343221.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-25
Publication Date
2024-11-29
Estimated Expiration
2044-09-25

AI Technical Summary

Technical Problem

The traditional in-vehicle noise reduction technology relies on passive sound insulation materials with limited effects, especially for low-frequency noise control. In addition, the adaptive filtering algorithm of the active noise reduction system is slow to converge and poorly adaptable, making it difficult to be in complex and variable in-vehicle noise environments. Implement ideal noise reduction effect.

Method used

Adaptive filtering algorithm is used to generate multiple control filter weight coefficients, combine the preset feature extraction algorithm to determine the mapping relationship between the training reference signal and the error signal and the filter weight coefficient, and extract the weight coefficient corresponding to the current signal as the noise reduction coefficient. And generate the target noise reduction signal through convolution.

Benefits of technology

It significantly improves the speed and applicability of noise reduction in the car, enables the active noise reduction system to maintain efficient and stable noise reduction effect in complex and variable noise environments, and improves riding experience and comfort.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119049444A_ABST
    Figure CN119049444A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of automobile noise reduction, in particular to an in-vehicle noise reduction method and device, equipment, a storage medium and a product. The method comprises the following steps: generating a plurality of control filter weight coefficients based on an adaptive filtering algorithm; based on a preset feature extraction algorithm, determining a mapping relation between a preset training reference signal and a control filter weight coefficient and between a preset training error signal and the control filter weight coefficient; based on the mapping relation, extracting control filter weight coefficients corresponding to the current reference signal and the current error signal, and taking the control filter weight coefficients as noise reduction coefficients; and carrying out convolution on the noise reduction coefficient and the current reference signal to generate a target noise reduction signal. According to the method, rapid matching and dynamic adjustment of the weight coefficient of the filter are achieved through the adaptive filtering algorithm and the feature extraction mapping technology, the noise reduction speed and applicability are effectively improved, and the in-vehicle active noise reduction system can still keep the efficient and stable noise reduction effect in the complex and changeable noise environment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of automobile noise reduction, and in particular to a method, device, equipment, storage medium and product for reducing noise in a vehicle. Background Art

[0002] With the development of the automobile industry, the impact of in-car noise on driving comfort has become more and more significant, and reducing in-car noise has become an important factor in improving user experience. Traditional in-car noise reduction technology mainly relies on the application of passive sound insulation materials, such as sound insulation cotton and sound insulation pads. Although these methods can reduce noise to a certain extent, the effect is limited, and the control effect on low-frequency noise is not good. In addition, passive noise reduction methods increase the weight of the car body, which is not conducive to the design trend of energy conservation and environmental protection of automobiles. Active noise reduction (ANC) technology has become a research hotspot in recent years due to its effective suppression of low-frequency noise. Traditional active noise reduction systems mostly use adaptive filtering algorithms to generate noise reduction signals, but there are problems such as slow convergence speed and poor adaptability, and it is difficult to achieve ideal noise reduction effects in complex and changeable in-car noise environments. Therefore, how to improve the speed and applicability of in-car noise reduction has become an urgent problem to be solved. Summary of the invention

[0003] The main purpose of this application is to provide a method, device, equipment, storage medium and product for in-vehicle noise reduction, aiming to solve the technical problem of how to improve the speed and applicability of in-vehicle noise reduction.

[0004] To achieve the above objectives, the present application provides a method for reducing noise in a vehicle, the method comprising:

[0005] Based on the adaptive filtering algorithm, multiple control filter weight coefficients are generated;

[0006] Based on a preset feature extraction algorithm, determining a mapping relationship between a preset training reference signal and a preset training error signal and the weight coefficient of the control filter;

[0007] Based on the mapping relationship, extracting the control filter weight coefficients corresponding to the current reference signal and the current error signal and using them as noise reduction coefficients;

[0008] The noise reduction coefficient is convolved with the current reference signal to generate a target noise reduction signal.

[0009] In one embodiment, the step of generating a plurality of control filter weight coefficients based on an adaptive filtering algorithm comprises:

[0010] Using noise information as input to a preset filter;

[0011] Initializing the weight coefficients and length of the preset filter;

[0012] Based on the adaptive filtering algorithm, the weight coefficients of the preset filter are adjusted to obtain the multiple control filter weight coefficients.

[0013] In one embodiment, the step of determining the mapping relationship between the preset training reference signal and the preset training error signal and the control filter weight coefficient based on the preset feature extraction algorithm includes:

[0014] Preprocessing the preset training reference signal and the preset training error signal to obtain a preprocessed signal, wherein the preprocessing includes one or more of a direct current component removal process, a noise removal process, and a normalization process;

[0015] Based on the preset feature extraction algorithm, feature extraction is performed on the preprocessed signal to obtain noise features;

[0016] The noise characteristics are respectively combined with corresponding control filter weight coefficients to obtain the mapping relationship.

[0017] In one embodiment, after the step of combining the noise features with the corresponding control filter weight coefficients to obtain the mapping relationship, the step further includes:

[0018] Constructing a mapping model based on the noise characteristics, the plurality of control filter weight coefficients and the mapping relationship;

[0019] Evaluating the training effect of the mapping model through a loss function;

[0020] The mapping model that passes the evaluation is validated, and based on the validation result, the hyperparameters of the mapping model are adjusted.

[0021] In one embodiment, the step of extracting the control filter weight coefficients corresponding to the current reference signal and the current error signal based on the mapping relationship and using them as noise reduction coefficients includes:

[0022] Performing the preprocessing on the current reference signal and the current error signal to obtain a current processed signal;

[0023] Based on the preset feature extraction algorithm, feature extraction is performed on the currently processed signal to obtain a current feature matrix;

[0024] The current feature matrix is ​​input into the mapping model to obtain corresponding control filter weight coefficients, which are used as the noise reduction coefficients.

[0025] In one embodiment, the step of convolving the noise reduction coefficient with the current reference signal to generate a target noise reduction signal includes:

[0026] Aligning the initial position of the noise reduction coefficient with the starting point of the current reference signal;

[0027] Determine a weighted sum at each position by moving the noise reduction coefficient point by point along the current reference signal;

[0028] generating a noise reduction signal based on the weighted sum;

[0029] The noise reduction signal is verified based on the current error signal, and the noise reduction signal that passes the verification is used as the target noise reduction signal.

[0030] In addition, to achieve the above-mentioned purpose, the present application also proposes a vehicle interior noise reduction device, the vehicle interior noise reduction device comprising:

[0031] A coefficient determination module, used for generating a plurality of control filter weight coefficients based on an adaptive filtering algorithm;

[0032] A mapping determination module, used to determine a mapping relationship between a preset training reference signal and a preset training error signal and the weight coefficient of the control filter based on a preset feature extraction algorithm;

[0033] A coefficient extraction module, used to extract the control filter weight coefficients corresponding to the current reference signal and the current error signal based on the mapping relationship, and use them as noise reduction coefficients;

[0034] The target module is used to convolve the noise reduction coefficient with the current reference signal to generate a target noise reduction signal.

[0035] In addition, to achieve the above-mentioned purpose, the present application also proposes an in-vehicle noise reduction device, which includes: a memory, a processor, and an in-vehicle noise reduction program stored in the memory and executable on the processor, wherein the in-vehicle noise reduction program is configured to implement the steps of the in-vehicle noise reduction method described above.

[0036] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, on which an in-vehicle noise reduction program is stored. When the in-vehicle noise reduction program is executed by a processor, the steps of the in-vehicle noise reduction method described above are implemented.

[0037] In addition, to achieve the above-mentioned purpose, the present application also proposes a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of the in-vehicle noise reduction method as described above are implemented.

[0038] This application generates multiple control filter weight coefficients based on an adaptive filtering algorithm; determines the mapping relationship between a preset training reference signal and a preset training error signal and the control filter weight coefficients based on a preset feature extraction algorithm; extracts the control filter weight coefficients corresponding to the current reference signal and the current error signal based on the mapping relationship, and uses them as noise reduction coefficients; convolves the noise reduction coefficients with the current reference signal to generate a target noise reduction signal. This application uses an adaptive filtering algorithm and feature extraction mapping technology to achieve rapid matching and dynamic adjustment of filter weight coefficients, effectively improving the speed and applicability of noise reduction, so that the in-vehicle active noise reduction system can still maintain efficient and stable noise reduction effects in complex and changeable noise environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a flow chart of the first embodiment of the in-vehicle noise reduction method of the present application;

[0040] Figure 2 This is a schematic diagram of a sub-process in the second embodiment of the in-vehicle noise reduction method of the present application;

[0041] Figure 3 This is a schematic diagram of a sub-process in the third embodiment of the in-vehicle noise reduction method of the present application;

[0042] Figure 4 The framework diagram for establishing the denoising weight coefficient set and CNN model training;

[0043] Figure 5 Schematic diagram of reference signal and error signal sub-band division;

[0044] Figure 6 This is the block diagram of the 2D CNN model;

[0045] Figure 7 This is a schematic diagram of the module structure of the in-vehicle noise reduction device according to an embodiment of the present application;

[0046] Figure 8 Schematic diagram of the device structure of the hardware operating environment involved in the in-vehicle noise reduction method in the embodiment of the present application.

[0047] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0048] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0049] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0050] It should be noted that with the development of the automobile industry, the impact of in-car noise on driving comfort has become more and more significant, and reducing in-car noise has become an important factor in improving user experience. Traditional in-car noise reduction technology mainly relies on the application of passive sound insulation materials, such as sound insulation cotton, sound insulation pads, etc. Although these methods can reduce noise to a certain extent, the effect is limited, and the control effect on low-frequency noise is not good. In addition, passive noise reduction methods increase the weight of the car body, which is not conducive to the design trend of energy conservation and environmental protection of automobiles. Active noise reduction (Active Noise Control, ANC) technology has become a research hotspot in recent years due to its effective suppression of low-frequency noise. Traditional active noise reduction systems mostly use adaptive filtering algorithms to generate noise reduction signals, but there are problems such as slow convergence speed and poor adaptability, and it is difficult to achieve ideal noise reduction effects in complex and changeable in-car noise environments. Therefore, how to improve the speed and applicability of in-car noise reduction has become an urgent problem to be solved.

[0051] The main solution of the present application is: based on an adaptive filtering algorithm, the present application generates multiple control filter weight coefficients; based on a preset feature extraction algorithm, the mapping relationship between a preset training reference signal and a preset training error signal and the control filter weight coefficients is determined; based on the mapping relationship, the control filter weight coefficients corresponding to the current reference signal and the current error signal are extracted and used as noise reduction coefficients; the noise reduction coefficients are convolved with the current reference signal to generate a target noise reduction signal.

[0052] This application realizes the rapid matching and dynamic adjustment of filter weight coefficients by utilizing adaptive filtering algorithm and feature extraction mapping technology, effectively improving the noise reduction speed and applicability, so that the active noise reduction system in the car can still maintain efficient and stable noise reduction effect in complex and changeable noise environments.

[0053] It should be noted that the execution subject of the method of this embodiment can be a computing service device with data processing, network communication and program running functions, or it can be the above-mentioned in-vehicle noise reduction device with the same or similar functions. This embodiment and the following embodiments will be described by taking the in-vehicle noise reduction device as an example.

[0054] Based on this, the first embodiment of the in-vehicle noise reduction method of the present application is proposed, please refer to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the in-vehicle noise reduction method of the present application.

[0055] In this embodiment, the method includes the following steps:

[0056] S1: Generate multiple control filter weight coefficients based on adaptive filtering algorithm;

[0057] It should be noted that the adaptive filtering algorithm is an algorithm that dynamically adjusts the filter weight coefficients to minimize the error signal. It updates the filter parameters in real time according to the changes in the input signal, so that the system can adaptively adjust in different environments and maintain the best filtering effect. Common adaptive filtering algorithms include the normalized least mean square algorithm (NLMS) and the recursive least squares algorithm (RLS). These algorithms adjust the filter weight coefficients with different update rules to ensure that the system has a good tracking ability for the changing noise environment. Controlling the filter weight coefficients is an important parameter of the filter, which is used to control the output signal shape of the filter. The size and arrangement of the weight coefficients directly determine the way the filter processes the input signal. By adjusting the weight coefficients, the filter can optimize the reference signal to generate a noise reduction signal suitable for the current noise environment. Multiple sets of control filter weight coefficients can cope with different noise patterns and improve the adaptability and response speed of the system.

[0058] Specifically, the initial reference signal and error signal are first collected from the in-vehicle sensor and input into the adaptive filter. The initial weight coefficient of the filter is set to zero or a small random value, which ensures that the system will not cause excessive interference to the input signal in the initial state. Adaptive filtering algorithms, such as NLMS or RLS, will use these initial signals to start adjusting the weight coefficients of the filter. The length of the filter (i.e., the number of weight coefficients) is determined by the design requirements of the system, usually depending on the complexity of the noise and the system computing power.

[0059] Furthermore, the adaptive filter processes each set of input reference signals and error signals through iterative calculations, and adjusts the current filter weight coefficients according to the size and direction of the error signal. Through gradual adjustment, the filter will output a set of updated weight coefficients after each iteration, and these weight coefficients will better match the current noise characteristics. The system will continue to make such adjustments to generate multiple sets of control filter weight coefficients, which represent the optimal noise reduction strategy in different noise scenarios. These multiple sets of weight coefficients can be stored for different noise environments or for subsequent weight coefficient mapping and optimization processes.

[0060] By using the adaptive filtering algorithm to generate multiple control filter weights, the system can prepare sets of weights suitable for a variety of noise environments in advance. In actual operation, the system can quickly select the weights that best suit the current noise characteristics from these sets, avoiding the complexity and delay of real-time calculations and greatly improving the response speed of the system. Different noise environments have different spectral characteristics. The adaptive filtering algorithm can generate multiple sets of weights to match various situations according to the changes in the noise environment. This dynamic adjustment capability enables the system to flexibly respond to the changing noise sources in the car, such as road bumps, engine noise, and external environmental sounds, significantly improving the applicability and stability of the in-car noise reduction system. Through repeated adjustment and optimization, the generated control filter weights can minimize the error signal, so that the system is in the optimal noise reduction state at every moment. This not only improves the effect of noise cancellation, but also makes the in-car environment quieter, improving the riding experience and comfort. This step provides a basis for the rapid matching of subsequent filter weights with noise characteristics, greatly improving the overall performance of the in-car noise reduction system.

[0061] S2: Determine a mapping relationship between a preset training reference signal and a preset training error signal and a weight coefficient of the control filter based on a preset feature extraction algorithm;

[0062] It should be noted that the preset feature extraction algorithm refers to a technology for feature analysis and extraction of signals, which is used to convert signals in the time domain or frequency domain into a data form that can reflect its essential characteristics. Common feature extraction algorithms include Fourier transform (FFT), short-time Fourier transform (STFT), Mel Spectrum Analysis, etc. These algorithms can convert complex signals into feature parameters that are easy to analyze and process, providing a basis for subsequent signal processing. The mapping relationship refers to the association between input signal features and output control parameters (such as filter weight coefficients) established through a model or algorithm. This relationship reflects how signal characteristics affect or determine the adjustment method of the filter, and is the key to achieving intelligent noise reduction. The mapping relationship can be trained and optimized through data-driven methods such as machine learning and deep learning.

[0063] Specifically, first, a large number of preset training reference signals and error signals are collected from multiple noise environments. These signals can be data collected during actual driving or data generated in a laboratory simulation environment. Then, these signals are processed using a preset feature extraction algorithm and converted into feature matrices, such as spectral features, energy distribution, phase information, etc. The extracted features can accurately describe the time-frequency characteristics of the signal. These features are used as input data and together with the corresponding control filter weight coefficients form a training data set.

[0064] Furthermore, by selecting appropriate machine learning or deep learning models (such as convolutional neural networks (CNNs), support vector machines (SVMs), etc.), the extracted signal features are associated with the control filter weight coefficients for training. The model learns the complex relationship between input features and output weight coefficients through a large amount of training data, and continuously adjusts its own parameters to optimize the prediction accuracy. After training, the model can accurately predict the optimal control filter weight coefficients corresponding to the current signal features, achieve real-time and fast mapping, and ensure that the system can automatically adjust to the optimal state in different noise environments.

[0065] Through the establishment of feature extraction algorithms and mapping relationships, the system can quickly identify the characteristics of the current noise environment and accurately match the most suitable control filter weight coefficients. Compared with the traditional successive optimization method, the use of mapping relationships greatly shortens the selection time of weight coefficients, thereby improving the response speed of the noise reduction system. The mapping relationship enables the system to automatically select the optimal filter weight coefficients according to the current environment without human intervention or manual adjustment. This intelligent adjustment capability enables the system to flexibly respond to different noise scenarios. Whether it is high-frequency noise or low-frequency noise, the system can respond in real time to the best effect, improving the effectiveness and stability of noise reduction. Through precise mapping relationships, the system can always maintain the working state of the optimal weight coefficient, ensuring that the output noise reduction signal can offset the actual noise to the greatest extent. Such optimization can significantly reduce the error signal in the car, making the car environment quieter and more comfortable, thereby improving the passenger experience and the overall quality of the vehicle. This step realizes the adaptive optimization of the noise reduction system by establishing an intelligent association between signal characteristics and control parameters, significantly improving the efficiency and effect of noise control in the car.

[0066] S3: Based on the mapping relationship, extract the control filter weight coefficients corresponding to the current reference signal and the current error signal, and use them as noise reduction coefficients;

[0067] It should be noted that the current reference signal is the latest noise signal collected by the active noise reduction system in the car during real-time operation, which is usually captured in real time by multiple sensors in the car (such as microphones). This signal reflects the characteristics of the current noise source in the car, including noise from the engine, tires, road surface, etc. The current error signal is the actual noise residual signal detected when the noise reduction system is running, which is generated by superimposing the noise reduction signal and the noise signal in the car. The error signal is used to evaluate the current noise reduction effect and guide the system to adjust the noise reduction strategy to continuously optimize the filter weight coefficients. Controlling the filter weight coefficients is a key parameter in the active noise reduction system. By adjusting these coefficients, the generation of the noise reduction signal can be optimized to offset the current noise in the car as much as possible to achieve the best noise reduction effect.

[0068] Specifically, the current reference signal and error signal are first collected through sensors, and these signals are preprocessed, such as filtering, denoising and normalization. Then, the current signal is feature analyzed using a preset feature extraction algorithm to extract parameters that can reflect the essential characteristics of the signal (such as spectral characteristics, energy distribution, etc.). These extracted feature data are used as input and sent to the previously trained mapping model. The mapping model is trained based on historical data and can identify the relationship between the input signal characteristics and the optimal control filter weight coefficients.

[0069] Furthermore, after the mapping model receives the input current signal characteristics, it will quickly analyze and match the corresponding control filter weight coefficients. The weight coefficients output by the model are the optimal noise reduction coefficients under the current environment. These noise reduction coefficients are used to control the adjustment of the filter to ensure that the generated noise reduction signal can offset the current noise to the greatest extent. Since this process is a fast match based on the mapping relationship, there is no need for complex real-time calculations. The system can quickly adapt to the changing noise environment and maintain the best noise reduction performance.

[0070] By extracting the control filter weight coefficient corresponding to the current signal based on the mapping relationship, the system can quickly locate the optimal noise reduction coefficient. This fast matching based on model prediction greatly shortens the filter adjustment time, enables the system to respond quickly when the noise environment changes, and greatly improves the noise reduction efficiency. The use of mapping relationships ensures a close match between the filter weight coefficient and the current noise characteristics, and the system can always remain in the optimal state even if the noise environment changes frequently. This adaptability enables the system to cope with variable noise sources, such as different road conditions, vehicle speed changes, or external interference, to ensure the consistency and stability of the noise reduction effect. By quickly extracting the optimal noise reduction coefficient corresponding to the current environment, the system can generate a noise reduction signal that is inversely proportional to the current noise, effectively reducing the error signal in the car. This ability of precise matching and dynamic adjustment makes the noise reduction effect in the car more significant, and passengers can enjoy a quieter and more comfortable car environment, improving the overall driving experience. This step achieves fast and accurate matching of the control filter weight coefficient through the efficient application of mapping relationships, greatly improving the response speed and adaptability of the in-car noise reduction system.

[0071] S4: Convolve the noise reduction coefficient with the current reference signal to generate a target noise reduction signal.

[0072] It should be noted that convolution is a mathematical operation that combines two signals to generate a new signal. The main function of this operation in signal processing is filtering, that is, combining a reference signal with a filter (noise reduction coefficient) through convolution to generate an output signal that can perform specific processing on the original signal.

[0073] Specifically, first prepare the extracted noise reduction coefficient and the current reference signal, use the noise reduction coefficient as the weight input of the filter, and the reference signal as the input signal. Before the convolution operation begins, both the reference signal and the filter coefficient must be normalized to ensure signal amplitude adaptability and avoid distortion during the operation.

[0074] Furthermore, the convolution operation uses the noise reduction coefficient as a sliding window, slides point by point across the entire reference signal, and calculates the weighted sum at each position to generate the output signal. Specifically, at each step of the convolution, the noise reduction coefficient is multiplied and summed with the corresponding part of the current reference signal to generate the output value at that position. Through this point-by-point calculation method, a complete target noise reduction signal is finally formed. This signal has an opposite phase and adjusted amplitude to the original noise signal, so it can effectively cancel the noise when physically superimposed.

[0075] The noise reduction coefficient is combined with the current reference signal through convolution operation, and the generated target noise reduction signal can accurately offset the current noise. This is because the convolution process adjusts the phase and amplitude of the signal so that the generated signal can form an opposite waveform with the original noise, thereby physically offsetting each other and achieving effective noise reduction. Since the convolution calculation is based on the reference signal collected in real time and the dynamically adjusted noise reduction coefficient, the generated target noise reduction signal can quickly adapt to the noise changes in the in-vehicle environment. Whether it is engine noise, road vibration or external interference noise, the system can generate a suitable noise reduction signal in real time to ensure the consistency and stability of the noise reduction effect. After the generated target noise reduction signal is superimposed with the in-vehicle noise in reverse phase, the noise level in the car is effectively reduced, creating a quieter and more comfortable environment for passengers. This not only improves the riding experience, but also reduces the fatigue caused by long-term exposure to noise, and improves the overall comfort and user satisfaction of the vehicle. This step combines the noise reduction coefficient with the reference signal through convolution operation to achieve accurate cancellation of in-vehicle noise, which is the key operation to achieve noise suppression in the entire noise reduction system.

[0076] This embodiment generates multiple control filter weight coefficients based on an adaptive filtering algorithm; determines the mapping relationship between a preset training reference signal and a preset training error signal and the control filter weight coefficients based on a preset feature extraction algorithm; extracts the control filter weight coefficients corresponding to the current reference signal and the current error signal based on the mapping relationship and uses them as noise reduction coefficients; convolves the noise reduction coefficients with the current reference signal to generate a target noise reduction signal. This embodiment achieves fast matching and dynamic adjustment of filter weight coefficients by utilizing an adaptive filtering algorithm and feature extraction mapping technology, effectively improving the noise reduction speed and applicability, so that the in-vehicle active noise reduction system can still maintain an efficient and stable noise reduction effect in a complex and changeable noise environment.

[0077] Based on the above first embodiment, a second embodiment of the vehicle interior noise reduction method of the present application is proposed. Figure 2 , Figure 2 This is a schematic diagram of a sub-process in the second embodiment of the in-vehicle noise reduction method of the present application.

[0078] like Figure 2 As shown, in this embodiment, step S1 includes:

[0079] S11: using the noise information as input of a preset filter;

[0080] S12: Initialize the weight coefficient and length of the preset filter;

[0081] S13: Based on the adaptive filtering algorithm, adjust the weight coefficient of the preset filter to obtain the multiple control filter weight coefficients.

[0082] It should be noted that the preset filter is a dynamic filter used for signal processing, and its weight coefficients can be continuously adjusted according to the input signal to optimize the filtering effect. The preset filter is usually initialized to zero or a small value, and dynamically adjusted through an adaptive algorithm as the system runs to achieve optimal processing of noise signals.

[0083] Specifically, noise information is collected from sensors inside the vehicle and used as input signals for the preset filter. This noise information may include noise characteristics from different sources such as the engine, road surface, and wind noise. Next, the weight coefficient and length of the preset filter are initialized. The initial value of the weight coefficient is generally set to zero or a random small value close to zero, and the length of the filter is set according to the design requirements of the system and the complexity of the input noise. This initialization ensures that the filter does not interfere too much with the noise in the initial state, waiting for further adjustment of the adaptive algorithm.

[0084] Furthermore, after the filter is initialized, the adaptive filtering algorithm begins to dynamically adjust the preset filter. Through the adaptive algorithm, the system continuously adjusts the weight coefficients according to the noise information collected in real time and the output error signal of the filter. The goal of the adjustment is to minimize the error signal, that is, the difference between the filter output and the expected output. The algorithm gradually adjusts the weight coefficients through iterative optimization. After multiple iterations, the preset filter can generate multiple control filter weight coefficients. These weight coefficients are the best control parameter sets under different noise environments, which can be used for rapid matching and application in the subsequent noise reduction process.

[0085] By dynamically adjusting the weight coefficients of the preset filter through the adaptive algorithm, the system can quickly adapt to different noise environments and generate multiple control filter weight coefficients. This adjustment mechanism enables the filter to respond to noise changes in a timely manner, improves the efficiency of generating noise reduction signals, and reduces the response delay of the system. The weight coefficients of the preset filter are automatically adjusted under different noise conditions, allowing the system to flexibly respond to various noise sources, such as engine noise, road vibration noise, external interference noise, etc. Different sets of control weight coefficients can cover various noise modes, enhance the adaptability of the system, and enable it to maintain the best noise reduction effect in complex and changing environments. The goal of the adaptive algorithm adjustment is to minimize the error signal, which means that the generated control filter weight coefficients are the optimized optimal parameter set. The optimization of the weight coefficients enables the filter to generate a noise reduction signal that is opposite to the noise more accurately, effectively reducing the noise level in the car and providing passengers with a quieter and more comfortable environment. This step realizes the generation of multiple control filter weight coefficients by initializing and adaptively adjusting the preset filter, providing a flexible parameter basis for the subsequent noise reduction process and improving the overall noise reduction performance of the system.

[0086] Based on the above first embodiment, in this embodiment, step S2 includes:

[0087] S21: preprocessing the preset training reference signal and the preset training error signal to obtain a preprocessed signal, wherein the preprocessing includes one or more of a DC component removal process, a noise removal process, and a normalization process;

[0088] S22: Based on the preset feature extraction algorithm, extract features from the preprocessed signal to obtain noise features;

[0089] S23: Combining the noise characteristics with corresponding control filter weight coefficients respectively to obtain the mapping relationship.

[0090] It should be noted that the preset training reference signal is a reference signal sample collected from historical data or experimental environment, which is used to train and optimize the control filter of the noise reduction system. This signal usually reflects the main characteristics of the noise source and provides a basis for the system to establish a mapping relationship between the characteristics and the filter weight coefficients. The preset training error signal is a residual noise signal that matches the reference signal, which is used to measure the noise reduction effect and guide the optimization of the filter weight coefficients. It represents the difference between the noise reduction signal and the actual noise, and is the feedback data in the control filter adjustment process. Feature extraction is the process of converting the preprocessed signal into feature data that is easier to analyze and model. Through feature extraction, key information in the signal, such as spectral characteristics, energy distribution, etc., can be extracted to provide a basis for establishing a mapping relationship between noise characteristics and control filter weight coefficients. The mapping relationship refers to combining the signal characteristics with the corresponding control filter weight coefficients through the feature extraction algorithm, thereby establishing an association model between the input signal characteristics and the output weight coefficients. This relationship enables the system to quickly select and adjust the weight coefficients in actual operation to optimize the noise reduction effect.

[0091] Specifically, the preset training reference signal and error signal are preprocessed to improve the quality and consistency of the signal. Preprocessing includes DC component removal, noise removal and normalization. Among them, DC component removal is to eliminate signal offset by subtracting the average value of the signal to ensure that the signal is symmetrical around zero. Noise removal removes unnecessary background noise through filtering and other methods to improve clarity. Normalization adjusts the amplitude of the signal to a uniform range for subsequent feature extraction and model training. These processing steps can eliminate interference factors in the signal and ensure the stability and robustness of the system under different signal conditions.

[0092] Furthermore, after preprocessing, the system uses a preset feature extraction algorithm (such as Fourier transform, short-time Fourier transform or Mel spectrum analysis) to extract features from the preprocessed signal. The goal of feature extraction is to convert the signal into a data form that can accurately describe its essential characteristics, such as spectral energy, frequency distribution, time-frequency structure, etc. These noise features are combined with the corresponding control filter weights to form training samples. Through these samples, the system can train a mapping model that can identify the relationship between features and weights, so that the optimal control filter weights can be quickly selected according to the current noise features during the actual noise reduction process.

[0093] By preprocessing the preset training signal, the system can effectively remove the DC component and noise interference in the signal, ensuring that the input training signal has good quality and stability. The normalization of the signal ensures that the feature extraction algorithm can operate within a uniform amplitude range, reducing the extraction deviation caused by the difference in signal amplitude. The preprocessed signal can more accurately reflect the essential characteristics of the noise in the feature extraction process. These accurate feature data are combined with the corresponding weight coefficients, so that the mapping relationship model can better learn the relationship between the features and the weight coefficients, thereby providing more accurate noise reduction parameter selection in practical applications. By establishing a mapping relationship between the preset training signal features and the control filter weight coefficients, the system can quickly identify the current noise environment and select the weight coefficients, significantly improving the system's adaptive ability and noise reduction effect. Even in complex and changeable noise scenes, the system can still dynamically adjust the weight coefficients to maintain the stability and consistency of the noise reduction performance. This step lays a solid foundation for the dynamic adjustment of the control filter through signal preprocessing, feature extraction and the establishment of mapping relationships, making the noise reduction system have higher accuracy and adaptability.

[0094] This embodiment generates multiple control filter weight coefficients based on an adaptive filtering algorithm; determines the mapping relationship between a preset training reference signal and a preset training error signal and the control filter weight coefficients based on a preset feature extraction algorithm; extracts the control filter weight coefficients corresponding to the current reference signal and the current error signal based on the mapping relationship and uses them as noise reduction coefficients; convolves the noise reduction coefficients with the current reference signal to generate a target noise reduction signal. This embodiment achieves fast matching and dynamic adjustment of filter weight coefficients by utilizing an adaptive filtering algorithm and feature extraction mapping technology, effectively improving the noise reduction speed and applicability, so that the in-vehicle active noise reduction system can still maintain an efficient and stable noise reduction effect in a complex and changeable noise environment.

[0095] Based on the above second embodiment, a third embodiment of the vehicle interior noise reduction method of the present application is proposed. Figure 3 , Figure 3 This is a schematic diagram of a sub-process in the third embodiment of the in-vehicle noise reduction method of the present application.

[0096] In this embodiment, after step S23, the method further includes:

[0097] S23a: constructing a mapping model based on the noise characteristics, the multiple control filter weight coefficients and the mapping relationship;

[0098] S23b: Evaluating the training effect of the mapping model through a loss function;

[0099] S23c: Validate the evaluated mapping model, and adjust hyperparameters of the mapping model based on the validation result.

[0100] It should be noted that the noise feature is the key information extracted from the preset training signal, which reflects the main characteristics of the noise signal, such as spectral characteristics, energy distribution, time-frequency domain characteristics, etc. These features are used to establish an association with the control filter weight coefficients to optimize the noise reduction effect. The mapping model is a data-driven model (such as a neural network, regression model, etc.) used to predict the optimal filter weight coefficients under the current noise environment. The model outputs the corresponding filter weight coefficients according to the noise characteristics to achieve intelligent control and adaptive adjustment of the system. The loss function is a standard for evaluating the training effect of the mapping model. It measures the difference between the model's predicted output and the true value. Common loss functions include mean square error (MSE), cross entropy loss, etc. The training results of the model are optimized by minimizing the loss function value. Hyperparameters are adjustable parameters in the model training process, such as learning rate, number of neural network layers, number of neurons, etc. Different hyperparameter settings will affect the training effect and prediction performance of the model. Optimizing hyperparameters helps improve the generalization ability and prediction accuracy of the model.

[0101] Specifically, through the aforementioned feature extraction and establishment of mapping relationships, the system combines noise features with multiple control filter weights as training data to construct a mapping model. The mapping model can be constructed using machine learning or deep learning techniques such as neural networks and support vector regression (SVR). During the training process, the model continuously adjusts internal parameters to learn the relationship between noise features and filter weights. After the training is completed, the model can output the corresponding optimal control filter weights based on the input noise features to provide support for the subsequent noise reduction process.

[0102] Furthermore, after the mapping model is trained, the system evaluates the prediction effect of the model through the loss function. The loss function calculates the error between the model's predicted output and the true weight coefficient, such as the mean square error (MSE) used to measure the difference between the predicted value and the actual value. After the evaluation is completed, the system verifies the performance of the model and adjusts the hyperparameters based on the verification results, such as adjusting the learning rate, the number of neural network layers, the batch size, etc. The optimization of hyperparameters can improve the training effect of the model, reduce the risk of overfitting, and improve the generalization ability of the model in different noise environments.

[0103] By constructing a mapping model and evaluating and optimizing it, the system can accurately predict the optimal control filter weight coefficient corresponding to the current noise characteristics. This high-precision mapping model enables the noise reduction system to respond quickly in complex and changeable noise environments, significantly improving the system's adaptability and noise reduction effect. By evaluating the mapping model through the loss function, the system can identify the source of errors in the model and optimize it by adjusting the hyperparameters. The optimized mapping model can more accurately match the noise characteristics with the filter weight coefficients, reduce the prediction error, make the generated noise reduction signal more accurate, and significantly improve the noise suppression effect in the car. Through verification and hyperparameter adjustment, the system ensures that the mapping model not only performs well on the training data, but also effectively copes with different noise scenarios in practical applications. This optimization process improves the generalization ability of the model, enabling the system to maintain efficient noise reduction in a variety of complex noise environments, ensuring the consistency and stability of the user experience. This step achieves accurate matching of noise characteristics and control filter weight coefficients by constructing, evaluating and optimizing the mapping model, providing the core intelligent adjustment capability for the noise reduction system.

[0104] Based on the above second embodiment, in this embodiment, step S3 includes:

[0105] S31: performing the preprocessing on the current reference signal and the current error signal to obtain a current processed signal;

[0106] S32: Based on the preset feature extraction algorithm, extract features from the currently processed signal to obtain a current feature matrix;

[0107] S33: Input the current feature matrix into the mapping model to obtain corresponding control filter weight coefficients as the noise reduction coefficients.

[0108] It should be noted that the current reference signal is the noise signal collected in real time by the active noise reduction system during actual operation, which reflects the characteristics of the noise source in the current environment in the car, including engine noise, road noise and wind noise. These signals are the basic input of the noise reduction process and are used to generate a noise reduction signal opposite to the noise. The current error signal is the actual noise signal remaining after the system generates the noise reduction signal. It represents the feedback information of the noise reduction effect and is used to adjust and optimize the weight coefficients of the control filter to further reduce the noise in the car. The feature matrix is ​​the result of the feature extraction algorithm and contains the multi-dimensional feature data of the current noise signal. The matrix shows the time-frequency characteristics of the signal in a structured manner and provides rich information input for the mapping model. The mapping model is a prediction model trained based on the association between the noise characteristics and the control filter weight coefficients. It can output the corresponding optimal control filter weight coefficients according to the input noise characteristics to optimize the noise reduction effect.

[0109] Specifically, the current reference signal and the current error signal are first preprocessed to remove the DC component, eliminate noise interference and normalize them to obtain a clearer and more stable current processed signal. This processing ensures the consistency and high quality of the input signal. Then, the system uses a preset feature extraction algorithm (such as Fourier transform, short-time Fourier transform, etc.) to extract features from the current processed signal and convert the signal into a multi-dimensional feature matrix. The feature matrix captures key information such as the frequency, amplitude, and energy of the current noise signal, and can accurately reflect the essential characteristics of the current noise.

[0110] Furthermore, after feature extraction is completed, the current feature matrix is ​​input into the previously constructed mapping model. The mapping model is trained based on the mapping relationship between noise features and control filter weight coefficients, and can quickly analyze the input feature matrix and predict the optimal control filter weight coefficients. These weight coefficients output by the model are the optimal noise reduction coefficients in the current environment, which are used to control the filter to generate a noise reduction signal that is opposite to the noise. This real-time prediction and application mechanism ensures that the noise reduction system maintains an efficient and stable noise reduction effect in a changing noise environment.

[0111] Through preprocessing and feature extraction, the system can quickly obtain high-quality feature data of the current noise environment, and directly input these features into the mapping model for weight coefficient prediction. This process reduces calculation delays, achieves rapid response to noise changes, and improves the real-time and adaptability of the noise reduction system. The mapping model can quickly match the optimal control filter weight coefficients according to the input feature matrix to generate accurate noise reduction coefficients. These optimized weight coefficients ensure that the noise reduction signal generated by the filter has the best anti-matching effect with the current noise, effectively reducing the residual noise in the car and significantly improving the noise reduction effect. Through precise analysis of the feature matrix and real-time adjustment of the weight coefficients, the system can continuously optimize the output signal of the filter and reduce the amplitude of the error signal. This process ensures that the noise reduction system can operate stably under various noise conditions and provide passengers with a quieter in-car environment. This step achieves rapid adaptation to the noise environment and accurate noise reduction through signal preprocessing, feature extraction and intelligent prediction of the mapping model, providing the core real-time adjustment capability for the in-car noise reduction system.

[0112] Based on the above second embodiment, in this embodiment, step S4 includes:

[0113] S41: aligning the initial position of the noise reduction coefficient with the starting point of the current reference signal;

[0114] S42: Determine a weighted sum at each position by moving the noise reduction coefficient point by point along the current reference signal;

[0115] S43: generating a noise reduction signal based on the weighted sum;

[0116] S44: verifying the noise reduction signal based on the current error signal, and using the noise reduction signal that passes the verification as the target noise reduction signal.

[0117] It should be noted that the weighted sum refers to the sum of the product of the noise reduction coefficient and each point of the current reference signal. In the convolution operation, the weighted sum represents the superposition effect of the signal and the filter coefficient at each position, which is a key step in generating the noise reduction signal. The target noise reduction signal is a proven effective noise reduction signal that can significantly reduce the noise in the car and provide a quiet environment for passengers. This signal is the noise reduction effect of the final output of the system.

[0118] Specifically, the initial position of the noise reduction coefficient is aligned with the starting point of the current reference signal to ensure that the convolution operation starts from the starting point of the signal. Then, by moving the noise reduction coefficient point by point along the current reference signal, the system calculates the product of the noise reduction coefficient and the signal value at each position and finds the weighted sum. Specifically, at each convolution position, the noise reduction coefficient is multiplied by a subsequence of the current reference signal one by one and then summed. This weighted sum is the output value of that position. Through point-by-point movement and cumulative calculation, a complete noise reduction signal is finally generated, which is the inverted cancellation result of the current noise signal.

[0119] Furthermore, after generating the noise reduction signal, the system verifies the generated noise reduction signal based on the current error signal. By analyzing the size and change trend of the error signal, the system determines whether the currently generated noise reduction signal is effective, that is, whether it can significantly reduce the noise. If the verification is successful, it means that the noise reduction signal has a good offset effect with the current noise signal, and the signal is confirmed and output as the final target noise reduction signal. If the verification fails, the system may need to adjust the noise reduction coefficient or recalculate to ensure that the final output target signal achieves the expected noise reduction effect.

[0120] By calculating the weighted sum of the noise reduction coefficient and the reference signal point by point, the system can generate a noise reduction signal that is highly matched to the current noise signal. This precise calculation method ensures that the phase and amplitude of the noise reduction signal are opposite to the noise signal, effectively improving the noise cancellation effect and reducing the residual noise in the car. By performing error signal verification on the generated noise reduction signal, the system can dynamically adjust the noise reduction strategy to ensure that the output signal remains effective in different noise environments. This verification mechanism enhances the stability of the system and makes the noise reduction effect more reliable and consistent in practical applications. The generation and verification process of the target noise reduction signal ensures the system's precise control and dynamic adjustment of the noise in the car, providing passengers with a quiet and comfortable in-car environment. This not only improves the driving and riding experience, but also reduces the fatigue and discomfort caused by long-term exposure to noise. This step ensures that the generated noise reduction signal can accurately and effectively cancel the current noise through point-by-point convolution calculation and dynamic verification, which is a key link in achieving efficient noise reduction.

[0121] This embodiment generates multiple control filter weight coefficients based on an adaptive filtering algorithm; determines the mapping relationship between a preset training reference signal and a preset training error signal and the control filter weight coefficients based on a preset feature extraction algorithm; extracts the control filter weight coefficients corresponding to the current reference signal and the current error signal based on the mapping relationship and uses them as noise reduction coefficients; convolves the noise reduction coefficients with the current reference signal to generate a target noise reduction signal. This embodiment achieves fast matching and dynamic adjustment of filter weight coefficients by utilizing an adaptive filtering algorithm and feature extraction mapping technology, effectively improving the noise reduction speed and applicability, so that the in-vehicle active noise reduction system can still maintain an efficient and stable noise reduction effect in a complex and changeable noise environment.

[0122] See also Figure 4-Figure 6 , Figure 4 The framework diagram for establishing the denoising weight coefficient set and CNN model training; Figure 5 Schematic diagram of reference signal and error signal sub-band division; Figure 6 This is the block diagram of the 2D CNN model.

[0123] In one embodiment, the method for reducing noise in a vehicle includes:

[0124] The reference signal x(n) passes through the secondary path After the filtering module, a filtered reference signal x′(n) is obtained.

[0125]

[0126] The filtered reference signal x′(n) is decomposed into m sub-band reference signals x′ after passing through the bandpass filter bank 1. i (n),i=1,2…m. Bandpass filter: i=1,2…m (m is an even number).

[0127] The error signal e(n) is decomposed into m sub-band signals e after passing through the bandpass filter bank 2. i (n), i = 1, 2…m, the bandwidth and frequency range of each sub-band are consistent with the sub-band vibration reference signal x i (n) Same.

[0128] The order of the noise reduction control filter W(n) is set to N, and each sub-band noise reduction control filter W i The order N i =N / m.

[0129] Subband reference signal x′ i (n) and the error signal e i (n) Input the weight coefficient W of each sub-band adaptive noise reduction control filter i (n) Optimization system, using the Normalized Least Mean Square (NLMS) algorithm to solve the optimal weight coefficients of each sub-band noise reduction control filter:

[0130]

[0131] Where: W i (n+1)-the current moment control filter weight coefficient, W i (n)-the weight coefficient of the control filter at the last moment, x′ i (n) - reference signal, - transposed reference signal, e i (n)-error signal, α-convergence factor, β-constant.

[0132] The weight coefficient W of the sub-band adaptive noise reduction control filter i (n) Perform a multi-point Fourier transform (FFT) to convert the sub-band adaptive noise reduction control filter weight coefficient W i (n) Transform from time domain to frequency domain W i (z).

[0133] W i (z) = FFT(W i (n))

[0134] In the frequency domain, the weight coefficient W of the sub-band adaptive noise reduction control filter is i (z) are stacked to obtain the full-band adaptive noise reduction control filter weight coefficient W(z) in the frequency domain.

[0135]

[0136] The weight coefficients W(z) of the full-band adaptive noise reduction control filter in the frequency domain are subjected to an inverse Fourier transform (IFFT) to obtain the weight coefficients W(n) of the full-band adaptive noise reduction control filter in the time domain.

[0137] W(n)=IFFT(W(z))

[0138] The reference vibration signal x(n) is convolved with the time-domain full-band adaptive noise reduction control filter weight coefficient W(n) to generate the noise reduction control filter output signal u(n).

[0139] u(n)=x(n)*W(n)

[0140] u(n) emits a noise reduction wave y(n) through a secondary path S(n) (including digital-to-analog conversion (DA), power amplification (AMP), etc.), which is superimposed in reverse phase with the original sound d(n) inside the car, thereby achieving the purpose of eliminating / reducing the noise inside the car.

[0141] e(n)=d(n)-y(n)

[0142] After the system achieves stable noise reduction, start 2D CNN model training; the weight coefficient W of the noise reduction control filter for each sub-band i Tagging (r i ,s i ) → W i (n); for each subband reference signal x′ i (n) and the error signal e i (n) Frame division; solve the Mel feature spectrum of each frame signal respectively as the input of the 2D CNN network, and use the corresponding sub-band noise reduction control filter weight coefficient label as the output, and establish the mapping relationship between the input layer and the output layer (x′ i (n),e i (n))→(r i ,s i ).

[0143] Repeat the above steps on different road surfaces to establish sub-band noise reduction control filter weight coefficient sets under different road surfaces, and the mapping relationship between the corresponding reference signal, error signal and output sub-band noise reduction control filter weight coefficient labels.

[0144] After the model training is completed, it is applied to the denoising process. The reference signal and the error signal are divided into subbands, and each subband signal is framed. The Mel feature spectrum of each frame signal is solved, and the solved Mel feature spectrum is output to the input layer of the 2D CNN network. The 2D CNN network recognizes the Mel feature spectrum of the reference signal and the error signal, and outputs the label of the corresponding subband denoising control filter weight coefficient according to the mapping relationship established previously. The corresponding subband denoising control filter weight coefficient is searched and extracted through the label.

[0145] (x′ i (n),e i (n))→(r i ,s i ) → W i (n)

[0146] The weight coefficient W of the filter for the extracted sub-band noise reduction control i (n) Perform a multi-point Fourier transform (FFT) to convert the sub-band noise reduction control filter weight coefficient W i (n) Transform from time domain to frequency domain W i (z).

[0147] W i (z) = FFT(W i (n))

[0148] In the frequency domain, the weight coefficient W of the sub-band noise reduction control filter is i (z) are stacked to obtain the full-band noise reduction control filter weight coefficient W(z) in the frequency domain.

[0149]

[0150] The full-band noise reduction control filter weight coefficient W(z) in the frequency domain is subjected to an inverse Fourier transform (IFFT) to obtain the full-band noise reduction control filter weight coefficient W(n) in the time domain.

[0151] W(n)=IFFT(W(z))

[0152] The reference vibration signal x(n) is convolved with the time-domain full-band noise reduction control filter weight coefficient W(n) to obtain the noise reduction control filter output signal u(n).

[0153] u(n)=x(n)*W(n)

[0154] u(n) emits a noise reduction wave y(n) through a secondary path S(n) (including digital-to-analog conversion (DA), power amplification (AMP), etc.), which is superimposed in reverse phase with the original sound d(n) inside the car, thereby achieving the purpose of eliminating / reducing the noise inside the car.

[0155] e(n)=d(n)-y(n)

[0156] Subband reference signal x′ i (n) and the error signal e i (n) Input the weight coefficient W of each sub-band adaptive noise reduction control filter i (n) Optimization system, using the Normalized Least Mean Square (NLMS) algorithm to solve the optimal noise reduction control filter weight coefficients for each subband:

[0157]

[0158] Where: W i (n+1)-current moment control filter coefficient, W i (n) - the control filter coefficient at the previous moment, x′ i (n) - reference signal, - transposed reference signal, e i (n)-error signal, α-convergence factor, β-constant.

[0159] The solved sub-band noise reduction control filter weight coefficient W i (n) Replace the weight coefficients in the corresponding control filter weight coefficient set library and modify the weight coefficients.

[0160] The present application also provides a vehicle interior noise reduction device. Figure 7 , Figure 7 This is a schematic diagram of the module structure of the in-vehicle noise reduction device according to an embodiment of the present application, and the in-vehicle noise reduction device includes:

[0161] A coefficient determination module 701 is used to generate a plurality of control filter weight coefficients based on an adaptive filtering algorithm;

[0162] A mapping determination module 702, configured to determine a mapping relationship between a preset training reference signal and a preset training error signal and a weight coefficient of the control filter based on a preset feature extraction algorithm;

[0163] A coefficient extraction module 703 is used to extract the control filter weight coefficients corresponding to the current reference signal and the current error signal based on the mapping relationship and use them as noise reduction coefficients;

[0164] The target module 704 is used to convolve the noise reduction coefficient with the current reference signal to generate a target noise reduction signal.

[0165] The in-vehicle noise reduction device provided in the embodiment of the present application adopts the in-vehicle noise reduction method in the above embodiment, which can solve the technical problem of how to improve the speed and applicability of in-vehicle noise reduction. Compared with the prior art, the beneficial effects of the in-vehicle noise reduction device provided in the embodiment of the present application are the same as the beneficial effects of the in-vehicle noise reduction method provided in the above embodiment, and other technical features of the in-vehicle noise reduction device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0166] The present application provides an in-vehicle noise reduction device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the in-vehicle noise reduction method in the above-mentioned embodiment.

[0167] Reference below Figure 8 , which shows a schematic diagram of the structure of an in-vehicle noise reduction device suitable for implementing the embodiment of the present application. The in-vehicle noise reduction device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 8 The in-vehicle noise reduction device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0168] like Figure 8 As shown, the in-vehicle noise reduction device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. Various programs and data required for the operation of the in-vehicle noise reduction device are also stored in the RAM 1004. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 may allow the in-vehicle noise reduction device to communicate with other devices wirelessly or by wire to exchange data. Although the in-vehicle noise reduction device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0169] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0170] The in-vehicle noise reduction device provided by the present application adopts the in-vehicle noise reduction method in the above embodiment, which can solve the technical problem of how to improve the speed and applicability of in-vehicle noise reduction. Compared with the prior art, the beneficial effects of the in-vehicle noise reduction device provided by the present application are the same as the beneficial effects of the in-vehicle noise reduction method provided by the above embodiment, and other technical features of the in-vehicle noise reduction device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0171] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0172] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0173] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the in-vehicle noise reduction method in the above-mentioned embodiment.

[0174] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above. The computer-readable storage medium may be included in the in-vehicle noise reduction device; or may exist independently without being assembled into the in-vehicle noise reduction device.

[0175] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the in-vehicle noise reduction device, the in-vehicle noise reduction device: generates multiple control filter weight coefficients based on an adaptive filtering algorithm; determines the mapping relationship between a preset training reference signal and a preset training error signal and the control filter weight coefficients based on a preset feature extraction algorithm; extracts the control filter weight coefficients corresponding to the current reference signal and the current error signal based on the mapping relationship, and uses them as noise reduction coefficients; convolves the noise reduction coefficients with the current reference signal to generate a target noise reduction signal. The computer program code for performing the operations of the present application can be written in one or more programming languages ​​or a combination thereof, and the programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or it can be connected to an external computer (for example, through the Internet using an Internet service provider).

[0176] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0177] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0178] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned in-vehicle noise reduction method, and can solve the technical problem of how to improve the speed and applicability of in-vehicle noise reduction. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the in-vehicle noise reduction method provided in the above-mentioned embodiment, and will not be repeated here.

[0179] An embodiment of the present application provides a computer program product, including a computer program, which implements the steps of the above-mentioned in-vehicle noise reduction method when executed by a processor.

[0180] The computer program product provided in this application can solve the technical problem of how to improve the speed and applicability of noise reduction in a vehicle. Compared with the prior art, the beneficial effects of the computer program product provided in the embodiment of this application are the same as the beneficial effects of the noise reduction method in a vehicle provided in the above embodiment, and will not be elaborated here.

[0181] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent processing scope of the present application.

Claims

1. A method for reducing noise in a vehicle, characterized in that: The method comprises: Based on the adaptive filtering algorithm, multiple control filter weight coefficients are generated; Based on a preset feature extraction algorithm, determining a mapping relationship between a preset training reference signal and a preset training error signal and the weight coefficient of the control filter; Based on the mapping relationship, extracting the control filter weight coefficients corresponding to the current reference signal and the current error signal and using them as noise reduction coefficients; The noise reduction coefficient is convolved with the current reference signal to generate a target noise reduction signal.

2. The method according to claim 1, characterized in that The step of generating a plurality of control filter weight coefficients based on the adaptive filtering algorithm comprises: Using noise information as input to a preset filter; Initializing the weight coefficients and length of the preset filter; Based on the adaptive filtering algorithm, the weight coefficients of the preset filter are adjusted to obtain the multiple control filter weight coefficients.

3. The method according to claim 1, characterized in that The step of determining the mapping relationship between the preset training reference signal and the preset training error signal and the control filter weight coefficient based on the preset feature extraction algorithm comprises: Preprocessing the preset training reference signal and the preset training error signal to obtain a preprocessed signal, wherein the preprocessing includes one or more of a direct current component removal process, a noise removal process, and a normalization process; Based on the preset feature extraction algorithm, feature extraction is performed on the preprocessed signal to obtain noise features; The noise characteristics are respectively combined with corresponding control filter weight coefficients to obtain the mapping relationship.

4. The method according to claim 3, characterized in that After the step of combining the noise characteristics with the corresponding control filter weight coefficients to obtain the mapping relationship, the method further includes: Constructing a mapping model based on the noise characteristics, the plurality of control filter weight coefficients and the mapping relationship; Evaluating the training effect of the mapping model through a loss function; The mapping model that passes the evaluation is validated, and based on the validation result, the hyperparameters of the mapping model are adjusted.

5. The method according to claim 4, characterized in that The step of extracting the control filter weight coefficients corresponding to the current reference signal and the current error signal based on the mapping relationship and using them as noise reduction coefficients includes: Performing the preprocessing on the current reference signal and the current error signal to obtain a current processed signal; Based on the preset feature extraction algorithm, feature extraction is performed on the currently processed signal to obtain a current feature matrix; The current feature matrix is ​​input into the mapping model to obtain corresponding control filter weight coefficients, which are used as the noise reduction coefficients.

6. The method according to claim 1, characterized in that The step of convolving the noise reduction coefficient with the current reference signal to generate a target noise reduction signal comprises: Aligning the initial position of the noise reduction coefficient with the starting point of the current reference signal; Determine a weighted sum at each position by moving the noise reduction coefficient point by point along the current reference signal; generating a noise reduction signal based on the weighted sum; The noise reduction signal is verified based on the current error signal, and the noise reduction signal that passes the verification is used as the target noise reduction signal.

7. A noise reduction device in a vehicle, characterized in that: The device comprises: A coefficient determination module, used for generating a plurality of control filter weight coefficients based on an adaptive filtering algorithm; A mapping determination module, used to determine a mapping relationship between a preset training reference signal and a preset training error signal and the weight coefficient of the control filter based on a preset feature extraction algorithm; A coefficient extraction module, used to extract the control filter weight coefficients corresponding to the current reference signal and the current error signal based on the mapping relationship, and use them as noise reduction coefficients; The target module is used to convolve the noise reduction coefficient with the current reference signal to generate a target noise reduction signal.

8. A computer device, characterized in that: The device comprises: a memory, a processor, and an in-vehicle noise reduction program stored in the memory and executable on the processor, wherein the in-vehicle noise reduction program is configured to implement the steps of the in-vehicle noise reduction method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium stores an in-vehicle noise reduction program, and when the in-vehicle noise reduction program is executed by the processor, the steps of the in-vehicle noise reduction method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the in-vehicle noise reduction method according to any one of claims 1 to 6 are implemented.

Citation Information

Patent Citations

  • Car active noise reduction method based on heuristic dynamic programming

    CN107791970A

  • Automobile active noise reduction method without feedback mechanism

    CN110992924A

  • Noise reduction method and device, test method and device, electronic equipment and storage medium

    CN115767359A

  • Active noise reduction method for automobile engine

    CN116052628A

  • Active noise control system and method based on DSP

    CN117037761A

Cited By

  • Filter coefficient matching model training method, active noise reduction method and system

    CN119864002A

  • Filter coefficient matching model training method, active noise reduction method and system

    CN119864002B

  • Automobile road noise processing method and device

    CN120089120A

  • Automobile road noise processing method, device, equipment, storage medium and product

    CN120089121A