Electromagnetic signal monitoring method and device based on deep learning

Abstract

The invention discloses an electromagnetic signal monitoring method and device based on deep learning in the technical field of electromagnetic signal monitoring. The electromagnetic signal monitoring method comprises the following steps: an electromagnetic signal monitoring process; carrying out deep learning on the electromagnetic signals; and the electromagnetic signal is subjected to an anti-electromagnetic interference process. According to the electromagnetic signal monitoring method and device based on deep learning, a proper high-performance device, a shielding scheme, an integrated design method and a deep learning model algorithm are adopted, monitoring and shielding strategies can be automatically adjusted according to real-time data through a deep learning model, manual intervention is reduced, the response speed and the processing efficiency are improved, and the electromagnetic signal monitoring method and device based on deep learning are suitable for popularization and application. Real-time data processing can be realized, a monitoring result can be quickly fed back, and potential interference can be timely dealt with.

Description

Technical Field

[0001] The present invention relates to the technical field of electromagnetic signal monitoring, and in particular to a method and device for electromagnetic signal monitoring based on deep learning. Background Art

[0002] At present, compared with the electromagnetic signal monitoring devices that introduce deep learning, the existing electromagnetic signal monitoring devices cannot effectively identify complex electromagnetic signals, especially in the interference environment with large noise, and the accuracy of signal detection is poor. The existing electromagnetic monitoring device algorithms are simple and cannot process complex signal data, which is likely to cause the omission and loss of important signal data. At the same time, due to the lack of advanced machine learning, it is impossible to perform intelligent analysis and prediction of data, and the processing efficiency is low. Ordinary monitoring devices are limited to recording basic data and cannot effectively manage and store large amounts of historical data. Finally, the current electromagnetic signal monitoring devices are limited to the monitoring of electromagnetic waves and cannot effectively shield and counter-interference electromagnetic waves. Summary of the invention

[0003] In order to solve the above problems, the purpose of the present invention is to provide an electromagnetic signal monitoring method and device based on deep learning, so as to solve the technical problems that the electromagnetic monitoring devices in the prior art cannot process complex signal data, cannot perform intelligent analysis and prediction of data, and cannot effectively shield and anti-interference of electromagnetic waves. Appropriate high-performance devices, shielding schemes, integrated design methods, and deep learning model algorithms are adopted to achieve high-precision, intelligent, real-time processing, self-adaptive capability, multi-source fusion comprehensive monitoring, long-term learning with low false alarm rate, electromagnetic shielding and active emission of "anti-phase" signals to offset interference.

[0004] In order to achieve the above object, the technical solution of the present invention is as follows:

[0005] The present invention provides an electromagnetic signal monitoring method based on deep learning, comprising the following steps:

[0006] S1. Electromagnetic signal monitoring process;

[0007] S2. Process of deep learning of electromagnetic signals;

[0008] S3. The process of electromagnetic signal anti-electromagnetic interference.

[0009] Furthermore, the S1 comprises the following steps:

[0010] S11. The signal display and control module sends a target monitoring command to the signal transceiver module through data processing;

[0011] S12. The signal transceiver module monitors the electromagnetic signals in the surrounding environment through the antenna, and collects and sends them to the sensor recognition module;

[0012] S13. The sensor recognition module captures electromagnetic signals from the surrounding environment, identifies the frequency and type of the signal, measures the signal strength, identifies potential fault signals and interference signals, and transmits the identified signal characteristics and processing results to the signal processing unit module;

[0013] S14. The signal processing unit module receives the electromagnetic signal from the sensor recognition module, amplifies the signal and converts it into a digital signal, analyzes the time domain and frequency domain, extracts the important features of the electromagnetic signal, and organizes and converts the results into a data format suitable for the deep learning model;

[0014] S15. The signal processing unit module receives the electromagnetic signal from the sensor identification module and transmits the data to the signal display and control module after processing, so as to facilitate the subsequent signal display and control module to issue instructions.

[0015] Further, the S2 comprises the following steps:

[0016] S21. Marking the electromagnetic signal after processing by the signal processing unit module as an interference signal and a normal signal, and manually marking the electromagnetic signal stored in the data storage module to distinguish between normal signals and interference signals;

[0017] S22. Based on the deep learning models of various architectures built into the deep learning module, the data is divided into a training set accounting for 70%, a validation set accounting for 15%, and a test set accounting for 15%;

[0018] S23. Select a loss function and optimization algorithm, perform multiple training and validation, and avoid overfitting by monitoring the performance of the validation set to select the best training model.

[0019] Furthermore, the built-in deep learning models of various architectures in S22 include:

[0020] Convolutional neural network (CNN) can effectively extract the time and frequency domain features of electromagnetic signals, automatically classify and identify signals, identify normal signals and interference signals, and make adjustments according to the changes of electromagnetic signals in different environments to improve the adaptability of the system; and

[0021] Recurrent Neural Network (RNN), which is used to process sequence data, analyze electromagnetic signals in time series, and capture the timing characteristics of signals; and

[0022] Long short-term memory network (LSTM), which can effectively predict and analyze changing trends; and

[0023] Generative adversarial networks (GANs) can learn generative tasks with manual data labeling.

[0024] Further, the S3 comprises the following steps:

[0025] S31. The deep learning model learns according to the received electromagnetic signal, and the deep learning model establishes an interference signal model according to the characteristics of the electromagnetic signal and predicts the interference characteristics of the future electromagnetic signal;

[0026] S32. The deep learning model emits an “anti-phase” signal that is opposite in phase to the interference signal to cancel the interference;

[0027] S33. The deep learning model evaluates the effectiveness of anti-interference measures and determines the anti-interference effect by monitoring the changes of interference signals in real time;

[0028] S34. Based on the new electromagnetic signal data collected by the sensor recognition module, it is fed back to the deep learning model to dynamically adjust the anti-interference strategy.

[0029] The present invention also provides an electromagnetic signal monitoring device based on deep learning, corresponding to the electromagnetic signal monitoring method based on deep learning, characterized in that it includes an external unit and an internal unit, and the internal unit includes:

[0030] A signal transceiver module, used for receiving and transmitting electromagnetic signals; and

[0031] a sensor recognition module that receives electromagnetic signal transmissions from the signal transceiver module and is used to capture and recognize electromagnetic signals; and

[0032] A signal processing unit module receives the electromagnetic signal from the sensor recognition module, performs front-end processing on the electromagnetic signal, and amplifies and converts the signal into a digital signal; and

[0033] A signal display and control module, interconnected with the signal processing unit module, receives and processes data from the signal processing unit module, and displays the collected electromagnetic signals; and

[0034] A deep learning module receives the electromagnetic signal from the signal processing unit module, extracts signal features through a variety of architectures and performs classification; and

[0035] The data storage module stores the extraction and classification results of the deep learning module for subsequent analysis.

[0036] Furthermore, the external unit includes:

[0037] The metal shielding shell is in the shape of a cuboid and has an electromagnetic shielding coating on the outside, which can effectively block external electromagnetic waves; and

[0038] A touch display screen, arranged on the front side of the metal shielding shell and electrically connected to the signal display and control module; and

[0039] A power module, disposed on the back of the metal shielding shell, using a rechargeable lithium battery to provide the electromagnetic signal monitoring device with the required power; and

[0040] A power control module, arranged on the upper side of the metal shielding shell, is control-connected to the power module and adopts any one of a toggle switch, a push switch or a rocker switch; and

[0041] A power indicator light, arranged on the upper side of the metal shielding shell, to display the status of the rechargeable lithium battery in real time; and

[0042] A charging interface, arranged on the upper side of the metal shielding shell, to provide a convenient way to replenish electric energy for the electromagnetic signal monitoring device; and

[0043] A USB external interface, provided on the upper side of the metal shielding shell, connects the internal unit with an external device; and

[0044] The antenna is arranged on the right side of the metal shielding shell, interconnected with the signal transceiver module, and is responsible for the transmission and reception of electromagnetic signals.

[0045] Furthermore, the external unit also includes a shielded transmitting module, one end of which is connected to the signal display and control module via a USB interface, and the other end of which is grounded, for controlling and managing the transmission of electromagnetic signals.

[0046] By adopting the above technical solution, the present invention has the following advantages:

[0047] The present invention provides an electromagnetic signal monitoring method and device based on deep learning. The deep learning model can extract complex features from a large amount of data, provide more accurate signal recognition and anomaly detection, and reduce the possibility of missed reports and false reports. Through the deep learning model, the monitoring and shielding strategies can be automatically adjusted according to real-time data, reducing manual intervention, improving reaction speed and processing efficiency, and can realize real-time data processing, quickly feedback monitoring results, and respond to potential interference in a timely manner. The deep learning model can carry out continuous learning and optimization, adapt to changes in the electromagnetic environment, and provide a more comprehensive monitoring perspective by integrating data from different sensors, and improve the ability to locate and identify electromagnetic interference sources. The deep learning model establishes an interference signal model according to the characteristics of the electromagnetic signal, and predicts the interference characteristics of the future electromagnetic signal and generates an "anti-phase" signal with the opposite phase to the interference signal to offset the interference. According to environmental changes and interference characteristics, the interference suppression strategy is dynamically adjusted to improve the suppression effect. Over time, the electromagnetic signal monitoring device can continuously optimize the interference identification and suppression strategy by accumulating data to adapt to the new electromagnetic environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1It is a structural block diagram of the internal units of the electromagnetic signal monitoring device based on deep learning;

[0049] Figure 2 A three-dimensional diagram of an external unit of an electromagnetic signal monitoring device based on deep learning;

[0050] Figure 3 It is a flow chart of the electromagnetic signal monitoring method based on deep learning;

[0051] Figure 4 Flowchart for deep learning of electromagnetic signals of the present invention. DETAILED DESCRIPTION

[0052] The technical solution of the present invention is described in detail below in conjunction with the drawings in the specification. It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0053] An electromagnetic signal monitoring device based on deep learning includes an external unit and an internal unit, wherein the internal unit is specifically as follows Figure 1 As shown, it includes a signal transceiver module, which monitors the electromagnetic signals in the surrounding environment through an antenna, and collects and sends them to a sensor identification module; and a sensor identification module, which captures electromagnetic signals from the surrounding environment to provide basic data and identify the frequency and type of the signal, measures the signal strength and identifies potential fault signals and interference signals, and finally passes the identified signal characteristics and processing results to the signal processing unit module to prepare for deep learning analysis; and a signal processing unit module, which receives the electromagnetic signal from the sensor identification module, performs front-end processing on the electromagnetic signal, and amplifies and converts the signal into a digital signal; and a signal display and control module, which is interconnected with the signal processing unit module, receives and processes the data of the signal processing unit module, and displays the collected electromagnetic signals; and a deep learning module, which receives the electromagnetic signal from the signal processing unit module, extracts and classifies the signal characteristics through a variety of architectures; and a data storage module, which stores the extraction and classification results of the deep learning module for subsequent analysis.

[0054] External units are as follows Figure 2As shown, it includes a metal shielding shell, a touch screen 1, a power module, a power control module 2, a power indicator 3, a charging interface 4, a charging interface 5 and an antenna 6, wherein the metal shielding shell is a rectangular parallelepiped with a width range of 240mm-250mm, a length range of 360mm-370mm, a height range of 25mm-35mm, and a thickness range of 0.5mm-2mm, and is made of a mixture of stainless steel and nickel silver. A layer of electromagnetic shielding coating is coated on the outside, which can effectively block external electromagnetic waves, prevent external electromagnetic interference from affecting internal signal processing, maintain signal integrity and clarity, and ensure the normal operation of the equipment;

[0055] The touch display screen 1 is installed on the front side of the metal shielding shell and is electrically connected to the signal display and control module of the internal unit, so as to facilitate the user to perform human-computer interaction and operation.

[0056] The power module is arranged on the back of the metal shielding shell. The power module adopts a rechargeable lithium battery to provide the necessary power for the electromagnetic signal monitoring device. This design is not only easy to carry, but also makes the charging process simple and efficient.

[0057] The power control module 2 is arranged on the upper side of the metal shielding shell and is controlled and connected to the power module, and adopts any one of a toggle switch, a push switch or a rocker switch; in a specific embodiment, the power control module is equipped with a push switch, which is installed on the upper side of the metal shielding shell to ensure the convenience of operation.

[0058] The power indicator light 3 is arranged on the upper side of the metal shielding shell to display the status of the rechargeable lithium battery in real time, reminding the user to pay attention to the power usage to prevent the device operation from being affected by insufficient power.

[0059] The charging interface 4 is arranged on the upper side of the metal shielding shell to provide a convenient way of replenishing electric energy for the electromagnetic signal monitoring device.

[0060] A USB external interface 5 is provided on the upper side of the metal shielding shell to connect the internal unit with an external device; and

[0061] Antenna 6 is installed on the right side of the metal shielding shell and is interconnected with the signal transceiver module of the internal unit. It is responsible for the transmission and reception of electromagnetic signals and is specifically used for signal reception and transmission to ensure the stability and effectiveness of communication. By optimizing the sensitivity and directivity of the antenna, the system can significantly improve the effect of signal capture and transmission, thereby enhancing the overall performance and stability.

[0062] The external unit also includes a shielded transmitting module, one end of which is connected to the signal display and control module of the internal unit through the USB external interface 5, and its transmission shielding effectiveness is between 30 and 100 decibels, and the other end is grounded to enhance the shielding effect. Used to control and manage the emission of electromagnetic signals, the shielded transmitting module can effectively reduce external electromagnetic wave interference, ensure the clarity and safety of the signal, and its operating frequency range is 0-6GHz, which is suitable for a variety of wireless communication applications and improves the overall performance of the system.

[0063] A method for monitoring electromagnetic signals based on deep learning, comprising the following steps: Figure 3 As shown:

[0064] S1. Electromagnetic signal monitoring process;

[0065] Among them, S1 includes the following specific steps:

[0066] S11. The signal display and control module sends a target monitoring command to the signal transceiver module through data processing;

[0067] S12. The signal transceiver module monitors the electromagnetic signals in the surrounding environment through the antenna, and collects and sends them to the sensor recognition module;

[0068] S13. The sensor recognition module captures electromagnetic signals from the surrounding environment, identifies the frequency and type of the signal, measures the signal strength, identifies potential fault signals and interference signals, and transmits the identified signal characteristics and processing results to the signal processing unit module;

[0069] S14. The signal processing unit module receives the electromagnetic signal from the sensor identification module, amplifies the signal and converts it into a digital signal to make it suitable for processing, analyzes the time domain and frequency domain at the same time, identifies the periodicity, spectrum and other characteristics of the signal, extracts the important features of the electromagnetic signal, organizes the results and converts them into a data format suitable for the deep learning model for subsequent analysis and use;

[0070] S15. The signal processing unit module receives the electromagnetic signal from the sensor identification module and transmits the data to the signal display and control module after processing, so as to facilitate the subsequent signal display and control module to issue instructions.

[0071] S2. Process of deep learning of electromagnetic signals;

[0072] S2 includes the following steps: Figure 4 As shown:

[0073] S21. The electromagnetic signal detected by the high-sensitivity electromagnetic sensor and processed by the signal processing unit module are marked as interference signals and normal signals, respectively, and the electromagnetic signal stored in the data storage module is manually marked to distinguish normal signals from interference signals;

[0074] S22. After the data is transferred to the deep learning module, the data is divided into a training set accounting for 70%, a validation set accounting for 15%, and a test set accounting for 15% according to the deep learning models of various architectures built into the deep learning module;

[0075] Deep learning models use a variety of architectures, and different architectures are suitable for different types of data and tasks, such as convolutional neural networks (CNN), recurrent neural networks (RNN), long short-term memory networks (LSTM), and generative adversarial networks (GAN). Convolutional neural networks (CNN) can effectively extract the time domain and frequency domain features of electromagnetic signals, automatically classify and identify signals, and identify normal signals and interference signals. And adjust according to the changes in electromagnetic signals in different environments to improve the adaptability of the system. Recurrent neural networks (RNN) are used to process sequence data, analyze electromagnetic signals in time series, and capture the timing characteristics of signals. Long short-term memory networks (LSTM) can effectively predict and analyze changing trends. Generative adversarial networks (GAN) can learn to generate tasks under the condition of manual data labeling.

[0076] S23. Deep learning models need to be trained with a large amount of labeled data, and the back propagation algorithm is used to adjust the weights between neurons. During the training process, an optimization algorithm is used to minimize the loss function and improve the prediction ability of the model. Select the loss function and optimization algorithm, conduct multiple training and verification, and avoid overfitting by monitoring the performance of the verification set. Select the best training model. During the training process, ensure that the format of the preprocessed electromagnetic signal is consistent with the format of the training data.

[0077] S3. The process of electromagnetic signal anti-electromagnetic interference.

[0078] S3 includes the following steps:

[0079] S31. Using the sensor recognition module to monitor the electromagnetic interference signal in the environment in real time, analyzing its frequency, amplitude and phase, the deep learning model learns according to the received electromagnetic signal, and the deep learning model establishes an interference signal model according to the characteristics of the electromagnetic signal and predicts the interference characteristics of the future electromagnetic signal;

[0080] S32. The deep learning model emits an “anti-phase” signal that is opposite in phase to the interference signal to cancel the interference;

[0081] S33. The deep learning model evaluates the effectiveness of anti-interference measures and determines the anti-interference effect by monitoring the changes of interference signals in real time;

[0082] S34. The sensor recognition module collects new data and trains the deep learning model to improve its accuracy and adaptability in identifying interference. Based on the monitoring results, a long-term monitoring system is established to continuously collect electromagnetic signal data and feed it back to the deep learning model to adjust its performance and anti-interference strategy, dynamically adjust anti-interference measures, and ensure the continued effectiveness of anti-interference.

[0083] The present invention integrates monitoring and shielding functions into the electromagnetic signal monitoring device to achieve efficient electromagnetic signal management. The monitoring data provides a basis for shielding decisions after deep learning. Conversely, the shielding effect can also be fed back to the deep learning model for optimization. Deep learning is used to optimize shielding decisions and improve interference recognition and response capabilities. It also supports manual adjustment of shielding levels and the ability to deal with interference.

[0084] Finally, it should be pointed out that although the present invention has been described with reference to the current specific embodiments, ordinary technicians in this technical field should realize that the above embodiments are only used to illustrate the present invention and are not used as a limitation of the present invention. Various equivalent changes or substitutions can be made without departing from the concept of the present invention. Therefore, as long as the changes and modifications to the above embodiments are within the essential spirit of the present invention, they will fall within the scope of the claims of the present invention.

Claims

1. A method for monitoring electromagnetic signals based on deep learning, characterized in that: The following steps are involved: S1. Electromagnetic signal monitoring process; S2. Process of deep learning of electromagnetic signals; S3. The process of electromagnetic signal anti-electromagnetic interference.

2. The electromagnetic signal monitoring method based on deep learning according to claim 1, characterized in that: The S1 comprises the following steps: S11. The signal display and control module sends a target monitoring command to the signal transceiver module through data processing; S12. The signal transceiver module monitors the electromagnetic signals in the surrounding environment through the antenna, and collects and sends them to the sensor recognition module; S13. The sensor recognition module captures electromagnetic signals from the surrounding environment, identifies the frequency and type of the signal, measures the signal strength, identifies potential fault signals and interference signals, and transmits the identified signal characteristics and processing results to the signal processing unit module; S14. The signal processing unit module receives the electromagnetic signal from the sensor recognition module, amplifies the signal and converts it into a digital signal, analyzes the time domain and frequency domain, extracts the important features of the electromagnetic signal, and organizes and converts the results into a data format suitable for the deep learning model; S15. The signal processing unit module receives the electromagnetic signal from the sensor identification module and transmits the data to the signal display and control module after processing, so as to facilitate the subsequent signal display and control module to issue instructions.

3. The electromagnetic signal monitoring method based on deep learning according to claim 1, characterized in that: The S2 comprises the following steps: S21. Marking the electromagnetic signal after processing by the signal processing unit module as an interference signal and a normal signal, and manually marking the electromagnetic signal stored in the data storage module to distinguish between normal signals and interference signals; S22. Based on the deep learning models of various architectures built into the deep learning module, the data is divided into a training set accounting for 70%, a validation set accounting for 15%, and a test set accounting for 15%; S23. Select a loss function and optimization algorithm, perform multiple training and validation, and avoid overfitting by monitoring the performance of the validation set to select the best training model.

4. The electromagnetic signal monitoring method based on deep learning according to claim 3 is characterized in that: The S22 has built-in deep learning models of various architectures, including: Convolutional neural network (CNN) can effectively extract the time and frequency domain features of electromagnetic signals, automatically classify and identify signals, identify normal signals and interference signals, and make adjustments according to the changes of electromagnetic signals in different environments to improve the adaptability of the system; and Recurrent Neural Network (RNN), which is used to process sequence data, analyze electromagnetic signals in time series, and capture the timing characteristics of signals; and Long short-term memory network (LSTM), which can effectively predict and analyze changing trends; and Generative adversarial networks (GANs) can learn generative tasks with manual data labeling.

5. The electromagnetic signal monitoring method based on deep learning according to claim 1, characterized in that: The S3 comprises the following steps: S31. The deep learning model learns according to the received electromagnetic signal, and the deep learning model establishes an interference signal model according to the characteristics of the electromagnetic signal and predicts the interference characteristics of the future electromagnetic signal; S32. The deep learning model emits an "anti-phase" signal that is opposite in phase to the interference signal to offset the interference; S33. The deep learning model evaluates the effectiveness of anti-interference measures and determines the anti-interference effect by monitoring the changes of interference signals in real time; S34. Based on the new electromagnetic signal data collected by the sensor recognition module, it is fed back to the deep learning model to dynamically adjust the anti-interference strategy.

6. An electromagnetic signal monitoring device based on deep learning, corresponding to the electromagnetic signal monitoring method based on deep learning, characterized in that: The invention comprises an external unit and an internal unit, wherein the internal unit comprises: A signal transceiver module, used for receiving and transmitting electromagnetic signals; and a sensor recognition module, receiving the electromagnetic signal transmission from the signal transceiver module, capturing and recognizing the electromagnetic signal; and A signal processing unit module receives the electromagnetic signal from the sensor recognition module, performs front-end processing on the electromagnetic signal, and amplifies and converts the signal into a digital signal; and A signal display and control module, interconnected with the signal processing unit module, receives and processes data from the signal processing unit module, and displays the collected electromagnetic signals; and A deep learning module receives the electromagnetic signal from the signal processing unit module, extracts signal features through a variety of architectures and performs classification; and The data storage module stores the extraction and classification results of the deep learning module for subsequent analysis.

7. The electromagnetic signal monitoring device based on deep learning according to claim 6, characterized in that: The external unit comprises: The metal shielding shell is in the shape of a cuboid and has an electromagnetic shielding coating on the outside, which can effectively block external electromagnetic waves; and A touch display screen, arranged on the front side of the metal shielding shell and electrically connected to the signal display and control module; and A power module, disposed on the back of the metal shielding shell, using a rechargeable lithium battery to provide the electromagnetic signal monitoring device with the required power; and A power control module, arranged on the upper side of the metal shielding shell, is control-connected to the power module and adopts any one of a toggle switch, a push switch or a rocker switch; and A power indicator light, arranged on the upper side of the metal shielding shell, to display the status of the rechargeable lithium battery in real time; and A charging interface, arranged on the upper side of the metal shielding shell, to provide a convenient way to replenish electric energy for the electromagnetic signal monitoring device; and A USB external interface, provided on the upper side of the metal shielding shell, for connecting the internal unit with an external device; and The antenna is arranged on the right side of the metal shielding shell, interconnected with the signal transceiver module, and is responsible for the transmission and reception of electromagnetic signals.

8. The electromagnetic signal monitoring device based on deep learning according to claim 7, characterized in that: The external unit also includes a shielded transmitting module, one end of which is connected to the signal display and control module through the USB external interface, and the other end of which is grounded, for controlling and managing the emission of electromagnetic signals.

Images