An electromagnetic interference noise suppression method, device, equipment and storage medium

Abstract

The application discloses an electromagnetic interference noise suppression method and device, equipment and a storage medium, comprising: acquiring a time sequence signal detected by a target sensor in an electronic device; inputting the time sequence signal into an electromagnetic interference noise suppression model obtained by training an initial suppression model based on a recurrent neural network with a sensor signal dataset of the electronic device collected under laboratory conditions without electromagnetic interference noise and with electromagnetic interference noise, so as to suppress electromagnetic interference noise in the time sequence signal through the electromagnetic interference noise suppression model to obtain a target sensor signal. The application filters out electromagnetic interference noise in the sensor signal through the electromagnetic interference noise suppression model based on the recurrent neural network, that is, filters out the electromagnetic interference noise in a software manner, can effectively suppress electromagnetic interference, and saves time and cost.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to an electromagnetic interference noise suppression method, apparatus, device, and storage medium. Background Technology

[0002] Electromagnetic interference (EMI) refers to the electromagnetic waves emitted by electronic devices during their operation, which can interfere with other parts of the device or other external equipment. Therefore, it is necessary to suppress the noise generated by electromagnetic interference to prevent interference with the device itself or other equipment.

[0003] Currently, most mainstream electromagnetic interference (EMI) noise suppression methods focus on hardware design. For example, during the early and later stages of electronic product development and EMI testing, common-mode filters, differential-mode filters, capacitors, and resistors are added to the electronic circuitry to suppress EMI noise. However, these methods have problems or limitations. For instance, even if EMI factors are thoroughly considered in the early stages of development, various forms of EMI may still occur during prototype testing, leading to failures. Improving the product at this point is not only technically challenging but also results in significant time and cost wastage due to rework. Furthermore, defects in structural design and PCB (Printed Circuit Board) design may prevent the implementation of improvements, hindering the product's timely market launch. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide an electromagnetic interference noise suppression method, apparatus, device, and storage medium, which can effectively suppress electromagnetic interference noise and save time and cost. The specific solution is as follows:

[0005] In a first aspect, this application discloses an electromagnetic interference noise suppression method, comprising:

[0006] Acquire timing signals detected by target sensors in electronic devices;

[0007] The time-series signal is input into the trained electromagnetic interference noise suppression model so that the electromagnetic interference in the time-series signal can be suppressed by the electromagnetic interference noise suppression model to obtain the target sensor signal; wherein, the electromagnetic interference noise suppression model is a model obtained by training an initial suppression model based on a recurrent neural network using sensor signal datasets of the electronic device collected under laboratory conditions under both electromagnetic interference noise and non-electromagnetic interference noise conditions.

[0008] Optionally, acquiring the timing signal detected by the target sensor in the electronic device includes:

[0009] Acquire the capacitive touch signal detected by the target sensor when the capacitive touch switch is touched;

[0010] Accordingly, the step of inputting the time-series signal into the trained electromagnetic interference noise suppression model, so as to suppress the electromagnetic interference noise in the time-series signal through the electromagnetic interference noise suppression model to obtain the target sensor signal, includes:

[0011] The capacitive touch signal is input into the trained electromagnetic interference noise suppression model so that the electromagnetic interference noise in the capacitive touch signal can be suppressed by the electromagnetic interference noise suppression model to obtain the target sensor signal.

[0012] Optionally, the process of creating the electromagnetic interference noise suppression model includes:

[0013] Under laboratory conditions, the sensor signal of the capacitive touch switch when it is touched is obtained without electromagnetic interference noise, and a first capacitive touch signal dataset is obtained.

[0014] Under laboratory conditions, the sensor signal of the capacitive touch switch when it is not touched is obtained under electromagnetic interference noise, and a capacitive no-touch signal dataset is obtained.

[0015] The initial suppression model based on a recurrent neural network is trained using the first capacitive touch signal dataset and the capacitive non-touch signal dataset to obtain an electromagnetic interference noise suppression model.

[0016] Optionally, the step of training the initial suppression model based on a recurrent neural network using the first capacitive touch signal dataset and the capacitive non-touch signal dataset to obtain an electromagnetic interference noise suppression model includes:

[0017] The capacitive no-touch signal dataset is preprocessed to obtain a preprocessed capacitive no-touch signal dataset.

[0018] The first capacitive touch signal dataset is superimposed on the preprocessed capacitive non-touch signal dataset as a label to obtain a second capacitive touch signal dataset containing electromagnetic interference.

[0019] The initial suppression model based on a recurrent neural network was trained using the second capacitive touch signal dataset to obtain an electromagnetic interference noise suppression model.

[0020] Optionally, the preprocessing of the capacitive touchless signal dataset to obtain a preprocessed capacitive touchless signal dataset includes:

[0021] The capacitive no-touch signal dataset is regularized to obtain a regularized capacitive no-touch signal dataset.

[0022] The electromagnetic interference noise segments in the regularized capacitive touchless signal dataset are preprocessed according to a preset electromagnetic interference noise segment processing method to obtain a preprocessed capacitive touchless signal dataset. The preset electromagnetic interference noise segment processing method includes any one or more of the following: randomly selecting electromagnetic interference noise segments, scaling electromagnetic interference noise segments, and randomly superimposing electromagnetic interference noise segments.

[0023] Optionally, the electromagnetic interference noise suppression method further includes:

[0024] Under laboratory conditions, the sensor signal of the capacitive touch switch when it is touched under electromagnetic interference noise is obtained, and a third capacitive touch signal dataset containing electromagnetic interference is obtained.

[0025] The electromagnetic interference noise suppression model was tested using the third capacitive touch signal dataset as a test set.

[0026] Optionally, the network structure of the electromagnetic interference noise suppression model consists of an input layer, a fully connected layer, a GRU layer, and a fully connected layer.

[0027] Secondly, this application discloses an electromagnetic interference noise suppression device, comprising:

[0028] The signal acquisition module is used to acquire the timing signals detected by the target sensor in the electronic device;

[0029] The noise suppression module is used to input the time-series signal into the trained electromagnetic interference noise suppression model so as to suppress the electromagnetic interference noise in the time-series signal through the electromagnetic interference noise suppression model to obtain the target sensor signal; wherein, the electromagnetic interference noise suppression model is a model obtained by training an initial suppression model based on a recurrent neural network using sensor signal datasets of the electronic device collected under laboratory conditions under both electromagnetic interference noise-free and electromagnetic interference noise-affected conditions.

[0030] Thirdly, this application discloses an electronic device, including a processor and a memory; wherein, when the processor executes a computer program stored in the memory, it implements the aforementioned electromagnetic interference noise suppression method.

[0031] Fourthly, this application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned electromagnetic interference noise suppression method.

[0032] As can be seen, this application first acquires the timing signal detected by the target sensor in the electronic device, and then inputs the timing signal into an electromagnetic interference noise suppression model obtained by training an initial suppression model based on a recurrent neural network using a dataset of sensor signals from the electronic device collected under laboratory conditions, both under electromagnetic interference noise and free from it. This allows the electromagnetic interference noise in the timing signal to be suppressed using the electromagnetic interference noise suppression model, thus obtaining the target sensor signal. Therefore, this application filters out electromagnetic interference noise from the sensor signal using an electromagnetic interference noise suppression model based on a recurrent neural network, i.e., it filters out electromagnetic interference noise through software, effectively suppressing electromagnetic interference noise while saving time and cost. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0034] Figure 1 This is a flowchart of an electromagnetic interference noise suppression method disclosed in this application;

[0035] Figure 2 This is a flowchart of a specific electromagnetic interference noise suppression method disclosed in this application;

[0036] Figure 3 This is a diagram illustrating the electromagnetic interference suppression effect of an electromagnetic interference noise suppression model disclosed in this application.

[0037] Figure 4 This is a schematic diagram of the structure of an electromagnetic interference noise suppression device disclosed in this application;

[0038] Figure 5 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0039] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0040] This application discloses an electromagnetic interference noise suppression method. See [link to relevant documentation]. Figure 1 As shown, the method includes:

[0041] Step S11: Acquire the timing signal detected by the target sensor in the electronic device.

[0042] In this embodiment, the time-series signals detected by sensors in the electronic device to be suppressed for electromagnetic interference are first acquired. The electronic device refers to a sensitive device easily affected by electromagnetic interference noise; it can be a small component, a circuit board assembly, a standalone electrical device, or even a large system. For example, when suppressing electromagnetic interference in a switching power supply, the time-series signals detected by sensors in the switching power supply are acquired.

[0043] Step S12: Input the time-series signal into the trained electromagnetic interference noise suppression model so that the electromagnetic interference in the time-series signal can be suppressed by the electromagnetic interference noise suppression model to obtain the target sensor signal; wherein, the electromagnetic interference noise suppression model is a model obtained by training an initial suppression model based on a recurrent neural network using a dataset of sensor signals of the electronic device collected under laboratory conditions under both electromagnetic interference noise and non-electromagnetic interference noise conditions.

[0044] In this embodiment, after acquiring the time-series signal detected by the target sensor in the electronic device, the time-series signal is input into an electromagnetic interference noise suppression model obtained by training an initial suppression model based on a Recurrent Neural Network (RNN) using sensor signal datasets of the electronic device collected under laboratory conditions, both under electromagnetic interference noise and free from it. Then, the electromagnetic interference noise suppression model suppresses electromagnetic interference in the input time-series signal, outputting the noise-suppressed target sensor signal. The network structure of the electromagnetic interference noise suppression model consists of an input layer, a fully connected layer, a GRU (Gated Recurrent Unit) layer, and another fully connected layer.

[0045] It is understandable that, since the electromagnetic interference noise suppression model can reduce noise from electromagnetic interference experienced by different electronic devices, when creating the electromagnetic interference noise suppression model, sensor signal datasets of different electronic devices under laboratory conditions, both those unaffected by electromagnetic interference noise and those affected by electromagnetic interference noise, can be collected according to the actual application scenario. Then, the collected sensor signal datasets can be used as the training set to train the initial suppression model based on a recurrent neural network, thereby obtaining an electromagnetic interference noise suppression model capable of reducing noise from various different electronic devices.

[0046] As can be seen, this embodiment first acquires the timing signal detected by the target sensor in the electronic device, and then inputs the timing signal into an electromagnetic interference noise suppression model obtained by training an initial suppression model based on a recurrent neural network using sensor signal datasets of the electronic device collected under laboratory conditions, both under electromagnetic interference noise and free from it. This allows the electromagnetic interference noise in the timing signal to be suppressed using the electromagnetic interference noise suppression model, thus obtaining the target sensor signal. Therefore, this embodiment filters out electromagnetic interference noise from the sensor signal using an electromagnetic interference noise suppression model based on a recurrent neural network, i.e., filtering out electromagnetic interference noise through software, which effectively suppresses electromagnetic interference noise and saves time and cost.

[0047] This application discloses a specific method for suppressing electromagnetic interference noise. (See also...) Figure 2 As shown, the method includes:

[0048] Step S21: Obtain the capacitive touch signal detected by the target sensor when the capacitive touch switch is touched.

[0049] In this embodiment, a capacitive touch switch is used as an example. The capacitive touch switch is used as an electronic device to suppress electromagnetic interference. First, the signal detected by the target sensor when the capacitive touch switch is touched is collected, that is, the capacitive touch signal.

[0050] Step S22: Input the capacitive touch signal into the trained electromagnetic interference noise suppression model so that the electromagnetic interference noise in the capacitive touch signal can be suppressed by the electromagnetic interference noise suppression model to obtain the target sensor signal; wherein, the electromagnetic interference noise suppression model is a model obtained by training an initial suppression model based on a recurrent neural network using a dataset of sensor signals of the capacitive touch switch collected under laboratory conditions under both electromagnetic interference noise and non-electromagnetic interference noise conditions.

[0051] In this embodiment, after obtaining the capacitive touch signal detected by the target sensor when the capacitive touch switch is touched, the capacitive touch signal is input into an electromagnetic interference noise suppression model obtained by training an initial suppression model based on a recurrent neural network using a dataset of sensor signals of the capacitive touch switch collected under laboratory conditions under both electromagnetic interference noise and no electromagnetic interference noise. Then, the electromagnetic interference noise in the capacitive touch signal is suppressed by the electromagnetic interference noise suppression model to obtain the noise-reduced target sensor signal.

[0052] In this embodiment, the creation process of the electromagnetic interference noise suppression model may specifically include: under laboratory conditions, acquiring sensor signals when the capacitive touch switch is touched without electromagnetic interference noise to obtain a first capacitive touch signal dataset; under laboratory conditions, acquiring sensor signals when the capacitive touch switch is not touched under electromagnetic interference noise to obtain a capacitive no-touch signal dataset; and training an initial suppression model based on a recurrent neural network using the first capacitive touch signal dataset and the capacitive no-touch signal dataset to obtain the electromagnetic interference noise suppression model. That is, under laboratory conditions, sensor signals when the capacitive touch switch is touched without external electromagnetic interference and sensor signals when the capacitive touch switch is not touched under external electromagnetic interference are collected to obtain the first capacitive touch signal dataset and the capacitive no-touch signal dataset, respectively. Then, the first capacitive touch signal dataset and the capacitive no-touch signal dataset are divided into a training set and a validation set according to a preset ratio. The training set is input into the initial suppression model based on a recurrent neural network for training to obtain the electromagnetic interference noise suppression model. Finally, the noise reduction effect of the electromagnetic interference noise suppression model is verified using the validation set.

[0053] In this embodiment, training the initial suppression model based on a recurrent neural network using the first capacitive touch signal dataset and the capacitive non-touch signal dataset to obtain an electromagnetic interference noise suppression model may specifically include: preprocessing the capacitive non-touch signal dataset to obtain a preprocessed capacitive non-touch signal dataset; superimposing the first capacitive touch signal dataset as a label with the preprocessed capacitive non-touch signal dataset to obtain a second capacitive touch signal dataset containing electromagnetic interference; and training the initial suppression model based on a recurrent neural network using the second capacitive touch signal dataset to obtain the electromagnetic interference noise suppression model. In this embodiment, after obtaining the first capacitive touch signal dataset and the capacitive non-touch signal dataset, the capacitive non-touch signal dataset can be preprocessed accordingly. Then, the first capacitive touch signal dataset can be superimposed as a label with the preprocessed capacitive non-touch signal dataset to obtain a synthesized second capacitive touch signal dataset containing electromagnetic interference. The second capacitive touch signal dataset is then divided into a training set and a test set according to a preset ratio, and the initial suppression model based on a recurrent neural network is trained using the training set to obtain the electromagnetic interference noise suppression model.

[0054] Specifically, the preprocessing of the capacitive touchless signal dataset to obtain a preprocessed capacitive touchless signal dataset may include: performing regularization on the capacitive touchless signal dataset to obtain a regularized capacitive touchless signal dataset, and then preprocessing the electromagnetic interference noise segments in the regularized capacitive touchless signal dataset according to a preset electromagnetic interference noise segment processing method to obtain a preprocessed capacitive touchless signal dataset; wherein, the preset electromagnetic interference noise segment processing method includes, but is not limited to, any one or more of the following: randomly selecting electromagnetic interference noise segments, scaling electromagnetic interference noise segments, and randomly superimposing electromagnetic interference noise segments.

[0055] After obtaining the electromagnetic interference noise suppression model, to verify its noise reduction effect, the following further steps can be taken: Under laboratory conditions, acquire the sensor signal when the capacitive touch switch is touched under electromagnetic interference noise to obtain a third capacitive touch signal dataset containing electromagnetic interference. Then, use this third capacitive touch signal dataset as a test set to test the electromagnetic interference noise suppression model, and use the test results to determine the noise reduction effect of the model. Evaluation metrics include, but are not limited to, SNR (Signal-to-Noise Ratio) and loss value. For details, see [link to relevant documentation]. Figure 3 As shown, Figure 3 The diagram shows the noise reduction effect of the electromagnetic interference noise suppression model on a capacitive signal dataset subjected to external electromagnetic interference and touch during an electromagnetic interference experiment. It can be seen that the electromagnetic interference noise in the capacitive signal dataset is removed after noise reduction by the electromagnetic interference noise suppression model.

[0056] It should be noted that during the training of the initial suppression model based on the recurrent neural network, the loss value decreases as the number of training iterations increases. The synthesized second capacitive touch signal dataset containing electromagnetic interference noise will increasingly resemble the first capacitive touch signal dataset without external electromagnetic interference noise under the influence of the model, resulting in a higher SNR value. Therefore, it is necessary to adjust the training parameters and loss function accordingly. In one specific implementation, the input data dimension of the input layer of the electromagnetic interference noise suppression model is set to 30, the output feature dimension to 18, the number of hidden units in the GRU layer to 12, the feature dimension of the fully connected layer to 12, and the number of network solution parameters to 1775. The training parameters are set as follows: the solver is set to Adam, the maximum number of iterations is set to 500, and the learning rate is set to 0.001.

[0057] For more detailed processing procedures regarding the above steps, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0058] As can be seen, this application takes a capacitive touch switch as an example. The capacitive touch signal when the capacitive touch switch is touched is input into an electromagnetic interference noise suppression model obtained by training an initial suppression model based on a recurrent neural network using a dataset of sensor signals of the capacitive touch switch collected under laboratory conditions when it is free from electromagnetic interference noise and when it is subject to electromagnetic interference noise. This suppresses electromagnetic interference noise by filtering out the electromagnetic interference noise received by the capacitive touch switch through software, and at the same time solves the baseline drift problem caused by electromagnetic interference noise.

[0059] Accordingly, this application also discloses an electromagnetic interference noise suppression device, see [link to relevant documentation]. Figure 4 As shown, the device includes:

[0060] The signal acquisition module 11 is used to acquire the timing signal detected by the target sensor in the electronic device;

[0061] The noise suppression module 12 is used to input the time-series signal into the trained electromagnetic interference noise suppression model so as to suppress the electromagnetic interference noise in the time-series signal through the electromagnetic interference noise suppression model to obtain the target sensor signal; wherein, the electromagnetic interference noise suppression model is a model obtained by training an initial suppression model based on a recurrent neural network using sensor signal datasets of the electronic device collected under laboratory conditions under both electromagnetic interference noise and non-electromagnetic interference noise conditions.

[0062] The specific workflow of each of the above modules can be found in the relevant content disclosed in the foregoing embodiments, and will not be repeated here.

[0063] As can be seen, in this embodiment, the timing signal detected by the target sensor in the electronic device is first acquired. Then, the timing signal is input into an electromagnetic interference noise suppression model obtained by training an initial suppression model based on a recurrent neural network using a dataset of sensor signals from the electronic device collected under laboratory conditions, both under electromagnetic interference noise and free from it. This allows the electromagnetic interference noise in the timing signal to be suppressed using the electromagnetic interference noise suppression model, thus obtaining the target sensor signal. Therefore, this embodiment uses an electromagnetic interference noise suppression model based on a recurrent neural network to filter out electromagnetic interference noise from the sensor signal, i.e., filtering out electromagnetic interference noise through software. This effectively suppresses electromagnetic interference noise and saves time and cost.

[0064] In some specific embodiments, the signal acquisition module 11 may specifically include:

[0065] The first signal acquisition unit is used to acquire the capacitive touch signal detected by the target sensor when the capacitive touch switch is touched;

[0066] Accordingly, the noise suppression module 12 may specifically include:

[0067] The noise suppression unit is used to input the capacitive touch signal into the trained electromagnetic interference noise suppression model so that the electromagnetic interference noise in the capacitive touch signal can be suppressed by the electromagnetic interference noise suppression model to obtain the target sensor signal.

[0068] In some specific embodiments, the process of creating the electromagnetic interference noise suppression model may specifically include:

[0069] The second signal acquisition unit is used to acquire the sensor signal when the capacitive touch switch is touched under laboratory conditions, without electromagnetic interference noise, to obtain the first capacitive touch signal dataset.

[0070] The third signal acquisition unit is used to acquire the sensor signal of the capacitive touch switch when it is not touched under laboratory conditions and is subject to electromagnetic interference noise, so as to obtain a capacitive no-touch signal dataset.

[0071] The first model training unit is used to train the initial suppression model based on the recurrent neural network using the first capacitive touch signal dataset and the capacitive non-touch signal dataset to obtain the electromagnetic interference noise suppression model.

[0072] In some specific embodiments, the first model training unit may specifically include:

[0073] The first data preprocessing unit is used to preprocess the capacitive non-touch signal dataset to obtain the preprocessed capacitive non-touch signal dataset.

[0074] The signal superposition unit is used to superimpose the first capacitive touch signal dataset as a label onto the preprocessed capacitive non-touch signal dataset to obtain a second capacitive touch signal dataset containing electromagnetic interference.

[0075] The second model training unit is used to train the initial suppression model based on the recurrent neural network using the second capacitive touch signal dataset to obtain the electromagnetic interference noise suppression model.

[0076] In some specific embodiments, the first data preprocessing unit may specifically include:

[0077] The data regularization unit is used to perform regularization processing on the capacitive non-touch signal dataset to obtain a regularized capacitive non-touch signal dataset.

[0078] The second data preprocessing unit is used to preprocess the electromagnetic interference noise segments in the normalized capacitive touchless signal dataset according to a preset electromagnetic interference noise segment processing method to obtain a preprocessed capacitive touchless signal dataset; the preset electromagnetic interference noise segment processing method includes any one or more of the following: randomly selecting electromagnetic interference noise segments, scaling electromagnetic interference noise segments, and randomly superimposing electromagnetic interference noise segments.

[0079] In some specific embodiments, the electromagnetic interference noise suppression method may further include:

[0080] The fourth signal acquisition unit is used to acquire the sensor signal of the capacitive touch switch when it is touched under laboratory conditions and is subject to electromagnetic interference noise, so as to obtain a third capacitive touch signal dataset containing electromagnetic interference.

[0081] The model testing unit is used to test the electromagnetic interference noise suppression model using the third capacitive touch signal dataset as a test set.

[0082] In some specific embodiments, the network structure of the electromagnetic interference noise suppression model consists of an input layer, a fully connected layer, a GRU layer, and a fully connected layer.

[0083] Furthermore, embodiments of this application also disclose an electronic device, Figure 5 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0084] Figure 5 This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the electromagnetic interference noise suppression method disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0085] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0086] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0087] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the electromagnetic interference noise suppression method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include a computer program capable of performing other specific tasks.

[0088] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned electromagnetic interference noise suppression method. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0089] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0090] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0091] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0092] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0093] The electromagnetic interference noise suppression method, apparatus, device, and storage medium provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for suppressing electromagnetic interference noise, characterized in that, include: Acquire timing signals detected by target sensors in electronic devices; The time-series signal is input into the trained electromagnetic interference noise suppression model so that the electromagnetic interference noise in the time-series signal can be suppressed by the electromagnetic interference noise suppression model to obtain the target sensor signal; wherein, the electromagnetic interference noise suppression model is a model obtained by training an initial suppression model based on a recurrent neural network using a dataset of sensor signals of the electronic device collected under laboratory conditions under both electromagnetic interference noise and non-electromagnetic interference noise conditions. The acquisition of timing signals detected by the target sensor in the electronic device includes: acquiring the capacitive touch signal detected by the target sensor when the capacitive touch switch is touched; The process of creating the electromagnetic interference noise suppression model includes: under laboratory conditions, acquiring sensor signals when the capacitive touch switch is touched without electromagnetic interference noise to obtain a first capacitive touch signal dataset; under laboratory conditions, acquiring sensor signals when the capacitive touch switch is not touched under electromagnetic interference noise to obtain a capacitive no-touch signal dataset; and using the first capacitive touch signal dataset and the capacitive no-touch signal dataset to train an initial suppression model based on a recurrent neural network to obtain an electromagnetic interference noise suppression model. The step of training an initial suppression model based on a recurrent neural network using the first capacitive touch signal dataset and the capacitive non-touch signal dataset to obtain an electromagnetic interference noise suppression model includes: preprocessing the capacitive non-touch signal dataset to obtain a preprocessed capacitive non-touch signal dataset; superimposing the first capacitive touch signal dataset as a label with the preprocessed capacitive non-touch signal dataset to obtain a second capacitive touch signal dataset containing electromagnetic interference; and training the initial suppression model based on a recurrent neural network using the second capacitive touch signal dataset to obtain the electromagnetic interference noise suppression model. The method further includes: acquiring sensor signals when the capacitive touch switch is touched under laboratory conditions and subjected to electromagnetic interference noise, to obtain a third capacitive touch signal dataset containing electromagnetic interference; using the third capacitive touch signal dataset as a test set to test the electromagnetic interference noise suppression model, and judging the noise reduction effect of the electromagnetic interference noise suppression model by the test results and preset evaluation indicators; the evaluation indicators include SNR and loss value. The step of inputting the timing signal into the trained electromagnetic interference noise suppression model so as to suppress the electromagnetic interference noise in the timing signal through the electromagnetic interference noise suppression model and obtain the target sensor signal includes: inputting the capacitive touch signal into the trained electromagnetic interference noise suppression model so as to suppress the electromagnetic interference noise in the capacitive touch signal through the electromagnetic interference noise suppression model and obtain the target sensor signal; The preprocessing of the capacitive touchless signal dataset to obtain a preprocessed capacitive touchless signal dataset includes: performing regularization on the capacitive touchless signal dataset to obtain a regularized capacitive touchless signal dataset; and preprocessing the electromagnetic interference noise segments in the regularized capacitive touchless signal dataset according to a preset electromagnetic interference noise segment processing method to obtain a preprocessed capacitive touchless signal dataset; the preset electromagnetic interference noise segment processing method includes any one or more of the following: randomly selecting electromagnetic interference noise segments, scaling electromagnetic interference noise segments, and randomly superimposing electromagnetic interference noise segments; The network structure of the electromagnetic interference noise suppression model consists of an input layer, a fully connected layer, a GRU layer, and a fully connected layer. The electromagnetic interference noise suppression model has an input data dimension of 30 for the input layer, an output feature dimension of 18, a hidden unit number of 12 for the GRU layer, a feature dimension of 12 for the fully connected layer, and a network solution parameter number of 1775. The training parameters are: the solver is Adam, the maximum number of iterations is 500, and the learning rate is 0.

001.

2. An electromagnetic interference noise suppression device, characterized in that, include: The signal acquisition module is used to acquire the timing signals detected by the target sensor in the electronic device; The noise suppression module is used to input the time-series signal into the trained electromagnetic interference noise suppression model so as to suppress the electromagnetic interference noise in the time-series signal through the electromagnetic interference noise suppression model to obtain the target sensor signal; wherein, the electromagnetic interference noise suppression model is a model obtained by training an initial suppression model based on a recurrent neural network using a dataset of sensor signals of the electronic device collected under laboratory conditions under both electromagnetic interference noise and non-electromagnetic interference noise conditions. The signal acquisition module is specifically used to acquire the capacitive touch signal detected by the target sensor when the capacitive touch switch is touched; The device is further configured to, under laboratory conditions, acquire sensor signals when the capacitive touch switch is touched without electromagnetic interference noise, to obtain a first capacitive touch signal dataset; under laboratory conditions, acquire sensor signals when the capacitive touch switch is not touched under electromagnetic interference noise, to obtain a capacitive no-touch signal dataset; and use the first capacitive touch signal dataset and the capacitive no-touch signal dataset to train an initial suppression model based on a recurrent neural network to obtain an electromagnetic interference noise suppression model. The device is further configured to preprocess the capacitive non-touch signal dataset to obtain a preprocessed capacitive non-touch signal dataset; superimpose the first capacitive touch signal dataset as a label onto the preprocessed capacitive non-touch signal dataset to obtain a second capacitive touch signal dataset containing electromagnetic interference; and train an initial suppression model based on a recurrent neural network using the second capacitive touch signal dataset to obtain an electromagnetic interference noise suppression model. The device is further configured to acquire, under laboratory conditions, the sensor signal of the capacitive touch switch when it is touched under electromagnetic interference noise, to obtain a third capacitive touch signal dataset containing electromagnetic interference; use the third capacitive touch signal dataset as a test set to test the electromagnetic interference noise suppression model, and judge the noise reduction effect of the electromagnetic interference noise suppression model by the test results and preset evaluation indicators; the evaluation indicators include SNR and loss value. The noise suppression module is specifically used to input the capacitive touch signal into the trained electromagnetic interference noise suppression model, so as to suppress the electromagnetic interference noise in the capacitive touch signal through the electromagnetic interference noise suppression model and obtain the target sensor signal. The preprocessing of the capacitive touchless signal dataset to obtain a preprocessed capacitive touchless signal dataset includes: performing regularization on the capacitive touchless signal dataset to obtain a regularized capacitive touchless signal dataset; and preprocessing the electromagnetic interference noise segments in the regularized capacitive touchless signal dataset according to a preset electromagnetic interference noise segment processing method to obtain a preprocessed capacitive touchless signal dataset; the preset electromagnetic interference noise segment processing method includes any one or more of the following: randomly selecting electromagnetic interference noise segments, scaling electromagnetic interference noise segments, and randomly superimposing electromagnetic interference noise segments; The network structure of the electromagnetic interference noise suppression model consists of an input layer, a fully connected layer, a GRU layer, and a fully connected layer. The electromagnetic interference noise suppression model has an input data dimension of 30 for the input layer, an output feature dimension of 18, a hidden unit number of 12 for the GRU layer, a feature dimension of 12 for the fully connected layer, and a network solution parameter number of 1775. The training parameters are: the solver is Adam, the maximum number of iterations is 500, and the learning rate is 0.

001.

3. An electronic device, characterized in that, It includes a processor and a memory; wherein, when the processor executes a computer program stored in the memory, it implements the electromagnetic interference noise suppression method as described in claim 1.

4. A computer-readable storage medium, characterized in that, Used to store computer programs; wherein, when the computer programs are executed by a processor, they implement the electromagnetic interference noise suppression method as described in claim 1.

Images