Deep learning method and system for electromagnetic signal classification and identification

Abstract

The invention discloses a deep learning method and system for classification and identification of electromagnetic signals, and the method comprises the steps: constructing electromagnetic signal data sets, each data set comprising multi-dimensional electromagnetic signal features and corresponding label information; selecting and configuring a corresponding deep learning model; label pollution is carried out on the obtained electromagnetic signal data set, pollution of Y individual levels and pollution of M matrix levels are included, the pollution of the Y individual levels is carried out by randomly selecting data points of a specific proportion and modifying labels of the data points, and the pollution of the M matrix levels is carried out by applying a predefined pollution matrix to carry out systematic modification on the labels of all or part of the data set; training an obtained deep learning model by using the obtained electromagnetic signal data set with the pollution label; and inputting an unpolluted electromagnetic signal data set into the obtained deep learning model, and outputting an identified electromagnetic signal. Data pollution of different levels can be effectively handled, and it is ensured that the model can achieve the optimal classification effect in various environments.

Description

Background Art

[0002] Electromagnetic signal classification is a key technology area that is widely used in a variety of fields such as communications, radar monitoring, and wireless spectrum management. The core of this technology lies in the ability to accurately identify and classify electromagnetic signals collected from different sources. Traditional signal classification methods usually rely on feature engineering and machine learning algorithms, such as support vector machines (SVM) and traditional neural networks. These methods perform well when the signal features are obvious and the environment is relatively stable. However, as the complexity of the signal environment increases, especially in the presence of interference and noise, the performance of these traditional methods often suffers.

[0003] In practical applications, especially in environments with severe noise or interference conditions, the performance of existing technologies often drops significantly due to the inability to effectively handle label errors or contamination. In addition, existing solutions have obvious deficiencies in the information utilization of multi-channel signals, dynamic adjustment of mislabeling, and adaptability to complex polluted environments, which limits their application effect in high-noise backgrounds.

[0004] Existing implementations: In recent years, the introduction of deep learning technology has brought new possibilities for electromagnetic signal classification. Deep neural networks, especially convolutional neural networks (CNNs) and recurrent neural networks (RNNs), have been widely studied and applied due to their powerful automatic feature extraction capabilities. These methods can achieve more accurate classification in complex signal environments by learning the underlying patterns in large amounts of data. However, although deep learning models have made significant progress in signal classification tasks, they still face challenges in dealing with label-contaminated data, especially when the training data contains mislabeled or noisy labels, the performance and robustness of the model usually decrease significantly. This shows that further technological innovation is needed to address emerging problems.

[0005] Existing electromagnetic signal classification technologies face several significant shortcomings, including low robustness to noise and label contamination, low classification efficiency, and insufficient ability to handle multi-channel data. First, existing methods are too sensitive to mislabeled or noisy data, which often leads to a significant performance degradation in practical applications, especially in environments with uneven data quality; Secondly, traditional signal processing algorithms often require complex feature engineering and long model training, which not only consumes a lot of computing resources but also prolongs the development cycle.

[0006] In addition, many models fail to effectively integrate information from multi-channel signal data, resulting in underutilization of important features, affecting classification accuracy and efficiency. Summary of the invention

[0007] The technical problem to be solved by the present invention is to provide a deep learning method and system for electromagnetic signal classification and recognition in view of the deficiencies in the above-mentioned prior art, so as to solve the technical problems of reduced recognition accuracy and insufficient model robustness caused by label contamination.

[0008] The present invention adopts the following technical solutions: A deep learning method for electromagnetic signal classification and recognition comprises the following steps: Construct electromagnetic signal datasets, each of which contains multi-dimensional electromagnetic signal features and corresponding label information; Select and configure the corresponding deep learning model; Conduct label contamination on the obtained electromagnetic signal dataset, including Y individual level contamination and M matrix level contamination. The Y individual level contamination is performed by randomly selecting a specific proportion of data points and modifying their labels. The M matrix level contamination is performed by systematically modifying the labels of all or part of the dataset by applying a predefined contamination matrix. A deep learning model is trained using the obtained electromagnetic signal dataset with pollution labels; The uncontaminated electromagnetic signal dataset is input into the obtained deep learning model, and the identified electromagnetic signals are output.

[0009] Preferably, the magnetic signal dataset includes an RML2016.10a dataset, a WiFi signal dataset and a Bluetooth individual dataset.

[0010] Preferably, the deep learning model includes MCLDNN and CNN5, MCLDNN is used to process the RML data set with multi-channel data input, and CNN5 is used to process the WiFi signal data set and the Bluetooth individual data set.

[0011] Preferably, the Y individual level pollution is specifically: Implement a label pollution function `add_symmetric_noise`, which systematically introduces symmetric noise into the label data; by randomly relabeling the sample labels in each category according to the specified noise ratio `noise_rate`; Introduce asymmetric noise through the function `add_asymmetric_noise` to modify the class label according to the predefined mislabeled pairs (source, target); Use `flip_labels` to perform label flipping, randomly select a certain proportion of data points and change their labels to any other labels except the current labels, so as to increase the diversity of data and test the robustness of the model to noise.

[0012] Preferably, use `flip_labels` to flip labels as follows:

[0013] in, for, for, is the number of signal categories.

[0014] Preferably, the M matrix level pollution is specifically: Construct the pollution matrix using the `construct_M_matrix_simple` method ; Implement `construct_M_matrix_block` method for total number of categories When it can be evenly divided, a block diagonal contamination matrix is ​​constructed; Construct a random pollution matrix through the `construct_M_matrix_random` method; Deploy the `LinearContaminationModel` class to take the total number of categories , a predefined pollution matrix , and the original probability of the category and the class probability after contamination Initialize the parameters; calculate the pollution weight of each category and make a random selection in the pollution matrix according to the weight to generate pollution labels that meet the settings of the pollution matrix, thereby achieving systematic pollution of the original data labels.

[0015] Preferably, in constructing the block diagonal contamination matrix, for even numbers , using a size of The all-ones matrix Build two identical blocks; for odd numbers , the last block size is adjusted to , ensuring that all categories are appropriately considered.

[0016] Preferably, constructing a random pollution matrix is ​​as follows: Generate a The random number matrix , and then normalize it by summing the rows to ensure that the sum of the elements in each row is 1; times the identity matrix and times the random matrix Linear combination to form the final random contamination matrix , random contamination matrix Can be used with probability Randomly contaminate the class label to any other class.

[0017] Preferably, the model robustness enhancement processing is performed at a specific contamination rate; the Huber loss function is introduced to optimize the model training; the Log-Cosh loss function is implemented; and the deep learning model is optimized.

[0018] In a second aspect, an embodiment of the present invention provides a deep learning system for electromagnetic signal classification and recognition, characterized in that it includes: Data module, constructs electromagnetic signal data sets, each of which contains multi-dimensional electromagnetic signal features and corresponding label information; Build modules, select and configure the corresponding deep learning model; The pollution module performs label pollution on the electromagnetic signal dataset, including Y individual level pollution and M matrix level pollution. The Y individual level pollution randomly selects a specific proportion of data points and modifies their labels. The M matrix level pollution systematically modifies the labels of all or part of the dataset by applying a predefined pollution matrix. The training module uses a deep learning model trained on an electromagnetic signal dataset with pollution labels; The recognition module inputs the uncontaminated electromagnetic signal dataset into the obtained deep learning model and outputs the recognized electromagnetic signal.

[0019] In a third aspect, a computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned deep learning method for electromagnetic signal classification and recognition when executing the computer program.

[0020] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, comprising a computer program, which, when executed by a processor, implements the steps of the above-mentioned deep learning method for electromagnetic signal classification and recognition.

[0021] Compared with the prior art, the present invention has at least the following beneficial effects: A deep learning method for electromagnetic signal classification and recognition introduces individual-level and matrix-level label contamination simulation methods; individual-level contamination simulates occasional errors that may occur in the data annotation process by randomly modifying the label of a single sample; matrix-level contamination systematically modifies the label through a predefined contamination matrix, which is suitable for simulating errors caused by device systematic deviations or algorithm recognition tendencies. This multi-level contamination simulation method effectively improves the robustness and accuracy of the model when facing complex pollution environments in the real world; secondly, dynamic learning strategies and adaptive loss functions are adopted, which are automatically adjusted according to the degree of data contamination to optimize the model training process and improve model performance. By dynamically adjusting learning parameters (such as learning rate) and selecting appropriate loss functions (such as Huber loss or Log-Cosh loss), the present invention can effectively deal with different levels of data contamination and ensure that the model can achieve the best classification effect in various environments.

[0022] Furthermore, RML2016.10a is one of the standard datasets for radio spectrum modulation recognition, covering 11 different modulation modes, such as AM, FM, QAM, PSK, etc. It contains radio signal samples collected under different signal-to-noise ratio (SNR) conditions, and is particularly suitable for signal modulation classification and recognition tasks; the WiFi signal dataset (NTU-HAR dataset) is designed for human activity recognition (HAR) research and uses channel state information (CSI) for wireless sensing. The dataset contains 3×114×500 CSI features, covering six different types of human activities (such as boxing, falling, walking, etc.), which represent human motion characteristics through reflected signals transmitted through wireless channels; the Bluetooth individual dataset (visualization effect is shown below) is mainly a dataset collected through Bluetooth signals. The dataset records signal samples from different Bluetooth devices at a sampling rate of 250Msps (mega samples per second). Each sample contains signal features from 150 Bluetooth devices, which is suitable for tasks such as device identification and identity authentication in Bluetooth communication.

[0023] Furthermore, MCLDNN and CNN5 were chosen to test the effects of the three label pollution strategies on different signal datasets. MCLDNN was used to process the complex RML2016.10a dataset because it combines spatiotemporal feature extraction capabilities, while CNN5 was used for the simpler WiFi signal dataset and Bluetooth individual dataset because of its lightweight structure and its ability to process local signal features. By using these two models, the impact of label pollution strategies in different signal environments can be more comprehensively evaluated, providing a basis for further optimizing label processing in signal processing.

[0024] Furthermore, the design purpose of Y individual-level contamination is to simulate individual label errors caused by individual labeling errors, collection noise or equipment interference in actual scenarios by randomly selecting certain samples in the data set and modifying their labels. This method helps the model improve its robustness to small-scale or local noise when facing imperfect data in the real world. By introducing individual-level label noise, the model can better learn immunity to noise, ensure high classification accuracy under imperfect labeling conditions, and adapt to diverse signal data characteristics.

[0025] Furthermore, the design purpose of M-matrix-level contamination is to simulate large-scale label noise caused by systematic errors, global interference or data transmission problems by systematically modifying the labels of the entire dataset on a large scale. This method helps the model to have stronger adaptability when facing more complex global noise. By introducing matrix-level contamination in multiple categories or regions in the dataset, the model can learn to effectively identify signal features in the presence of widespread noise interference, avoiding a significant drop in model performance due to widespread noise, thereby improving the generalization and robustness of the model.

[0026] Furthermore, by performing robustness enhancement processing at a specific contamination rate, the label contamination problem that may occur in real-life scenarios can be effectively simulated. The Huber loss function combines the advantages of mean square error and absolute error, and can provide a more stable gradient when confronting outliers, thereby reducing the impact of noise interference on the model. The Log-Cosh loss function further improves the performance of the model in the face of large-scale errors, retaining sufficient sensitivity while avoiding gradient explosions caused by outliers; through these robustness enhancement methods, the model can still maintain a high classification accuracy even at a higher contamination rate, ensuring more robust processing of contaminated data in practical applications. This method provides the model with stronger fault tolerance and generalization capabilities in complex real-world data environments.

[0027] It can be understood that the beneficial effects of the second aspect mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0028] In summary, the present invention can effectively simulate the label inaccuracy that may occur in the real environment by introducing the label contamination mechanism at the Y individual level and the M matrix level. By selecting a suitable deep learning model and optimization strategy, this method can significantly improve the robustness and generalization ability of electromagnetic signal classification. When processing complex multi-dimensional electromagnetic signal features, the model shows strong adaptability, ensuring that a high classification accuracy can still be achieved under different pollution conditions, thereby enhancing the stability and practicality of the system.

[0029] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings to be used in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0031] Figure 1 A schematic diagram of the process of generating pollution labels for the electromagnetic signal classification model based on the multi-level label pollution strategy of the present invention; Figure 2 A schematic flow chart of a method for electromagnetic signal processing and model robustness optimization based on deep learning provided in an embodiment of the present invention; Figure 3 A schematic diagram of an electromagnetic signal processing flow based on a multi-channel deep neural network (MCLDNN) provided in an embodiment of the present invention; Figure 4 It is the result of multi-gradient contamination of simple M matrix; Figure 5 It is the result of multi-gradient contamination of random M matrix; Figure 6 Comparison of the fitting results of the two methods; Figure 7 Comparison of results of various pollution modes on the WIFI dataset; Figure 8 To compare the results of various pollution methods on the Bluetooth brand individual data set; Fig. 9 Schematic diagram of introducing three label contamination strategies on three datasets; Fig.10 A schematic diagram of a computer device provided by an embodiment of the present invention; Fig.11 The present invention is a block diagram of an electronic device provided according to an embodiment. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] In the description of the present invention, it should be understood that the terms “include” and “comprises” indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0034] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0035] It should be further understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present invention generally indicates that the associated objects are in an "or" relationship.

[0036] It should be understood that, although the terms first, second, third, etc. may be used to describe preset ranges, etc. in the embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are only used to distinguish preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0037] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0038] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0039] The present invention provides a deep learning method for electromagnetic signal classification and recognition, which effectively reproduces pollution scenarios in the real world by simulating label pollution at the individual level and matrix level; individual level pollution simulates occasional data labeling errors by randomly modifying the labels of individual samples, while matrix level pollution simulates systematic errors caused by equipment system deviations or algorithm recognition tendencies. In addition, the present invention optimizes the performance of the model in various environments by dynamically adjusting the loss function and training strategy to adapt to different pollution levels; through these innovative measures, the present invention not only improves the accuracy of electromagnetic signal classification, but also significantly enhances the robustness of the model in the face of complex pollution conditions.

[0040] See also Figure 1 and Figure 2 The present invention provides a deep learning method for electromagnetic signal classification and recognition, comprising the following steps: S1. Provide multiple electromagnetic signal datasets, including but not limited to RML2016.10a dataset, WiFi signal dataset and Bluetooth individual dataset, each of which contains multi-dimensional electromagnetic signal features and corresponding label information; S2. Select and configure a suitable deep learning model for electromagnetic signal processing, wherein the model includes but is not limited to MCLDNN and CNN5, wherein MCLDNN is used for multi-channel data input processing of RML data sets, and CNN5 is used for processing WiFi signal data sets and Bluetooth individual data sets; See also Figure 3 , MCLDNN (Multi-Channel Learning Deep Neural Network) is a multi-channel learning deep neural network, mainly used to process and classify complex signals. MCLDNN can receive data input from multiple channels at the same time, and extract features from different scales through its hierarchical deep structure. Its design goal is to effectively capture cross-channel correlation and distinctive information in signal processing tasks, and is particularly suitable for processing data with multi-dimensional features such as electromagnetic signals.

[0041] S3, label contamination of the data set, including Y individual level contamination and M matrix level contamination, where individual level contamination is performed by randomly selecting a specific proportion of data points and modifying their labels, and matrix level contamination is performed by systematically modifying the labels of all or part of the data set by applying a predefined contamination matrix; Individual level pollution is specifically: S301, implement a label pollution function `add_symmetric_noise`, which systematically introduces symmetric noise into the label data; simulate mislabeling by randomly relabeling the sample labels in each category according to the specified noise ratio `noise_rate`, but excluding the original category; this process is expressed as conditional probability; The conditional probability is:

[0042] in, and is the total number of categories.

[0043] S302, introducing asymmetric noise through the function `add_asymmetric_noise`, which specifically modifies the category label according to the predefined mislabeled pair (source, target); This operation reflects prior knowledge based on false alarms, such as QAM16 being mislabeled as QAM64, reflecting possible bias in the label generation process. For each pair (s, t) in `noise_pairs`, the probability of mislabeling is defined as , thereby precisely controlling the label contamination dynamics during model training.

[0044] S303. Use `flip_labels` to perform label flipping. This step randomly selects a certain proportion of data points and changes their labels to any other labels except the current labels, so as to increase the diversity of the data and test the robustness of the model to noise.

[0045] This operation is defined as:

[0046] when Further simulations are performed to account for possible random disturbances in the data.

[0047] The M matrix level is specifically: S304, construct a simple M matrix; Construct the pollution matrix using the `construct_M_matrix_simple(K, epsilon)` method The matrix is ​​linearly combined to times the identity matrix and times the all-ones matrix Add together to create a pollution rate of The symmetric contamination matrix ensures that the contamination probability of each class is evenly distributed among all other classes.

[0048] S305, constructing a block M matrix; Implement `construct_M_matrix_block(K, epsilon)` method for total number of categories When it can be evenly divided, a block diagonal contamination matrix is ​​constructed; this matrix adopts different construction strategies according to the parity of the number of categories: For even numbers , using a size of The all-ones matrix Build two identical blocks; For odd numbers , the last block size is adjusted to , ensuring that all categories are appropriately considered.

[0049] Through this method, the contamination correlation within the category is enhanced, reflecting the possible intrinsic connection between categories in practical applications.

[0050] S306, constructing a random M matrix; Through the `construct_M_matrix_random(K, epsilon, random_state)` method, a random pollution matrix is ​​constructed as follows: First generate a The random number matrix , after row summing, normalization is performed to ensure that the sum of the elements in each row is 1; Then, times the identity matrix and times the random matrix Linear combination to form the final random contamination matrix , the matrix can be expressed as Randomly contaminate the class label to any other class.

[0051] S307. Apply the linear contamination model.

[0052] Deployment: `LinearContaminationModel` class, which is based on the total number of categories , a predefined pollution matrix , and the original probability of the category and the class probability after contamination Initialize the parameters; the model calculates the pollution weight of each category and randomly selects in the pollution matrix according to the weight to generate pollution labels that meet the settings of the pollution matrix, thereby achieving systematic pollution of the original data labels.

[0053] S4, training the deep learning model using a data set with contaminated labels, wherein the training process includes weight initialization, forward propagation, loss calculation, back propagation, and parameter update; S401. Initialize the deep learning model and select a network structure suitable for electromagnetic signal classification, such as MCLDNN or CNN5.

[0054] Configuration includes setting an appropriate learning rate, batch size, and number of training iterations.

[0055] Use an uncontaminated electromagnetic signal dataset to train the model normally to ensure that the model achieves optimal performance under baseline conditions.

[0056] This step provides a performance comparison baseline for subsequent contamination experiments.

[0057] S402, automated gradient contamination implementation.

[0058] Design and implement an automated script that automatically adjusts the contamination rate according to a preset gradient (from 5% to 90%, increasing by 5% each time) and gradually applies it to the labeled data. The script supports two contamination levels (individual level and M-matrix level) and three specific contamination methods under each level.

[0059] S5. Evaluate the performance of the model on an uncontaminated test data set, including but not limited to calculating the accuracy, recall, precision and F1 score of the model, and comparing the changes in model performance before and after contamination; Use the above script to generate datasets with different contamination rates. Retrain the model for each dataset generated by each contamination level and method, and record and analyze the performance changes of the model on the contaminated data. Collect key performance indicators such as accuracy, recall, and F1 score for each experiment. Analyze the impact of different contamination rates, levels, and methods on model performance to quantitatively evaluate the specific impact of contamination on recognition accuracy. Based on the collected data, draw a graph of the relationship between contamination rate and model performance. These graphs intuitively show the impact of different contamination strategies on model performance, revealing the potential threat of data quality issues that may be encountered in practical applications to model performance.

[0060] S6. Conduct a systematic evaluation of different pollution methods at multiple levels, and conduct pollution experiments with a gradient of 5% for a total of 18 times; through repeated experiments, accurately depict the quantitative relationship between the pollution ratio and the accuracy of electromagnetic signal classification; S7. At a specific contamination rate, such as 10%, the model robustness is enhanced; the Huber loss function is used to replace the traditional loss function to enhance the model's tolerance to abnormal labels; at the same time, the frequent misjudgments are counted as prior knowledge, and the network is adjusted to focus on more discriminative features, thereby improving the overall robustness and classification accuracy of the model.

[0061] S701. Based on the experimental results, evaluate and adjust the model parameters or introduce new algorithm strategies, such as introducing new loss functions or data enhancement techniques, to improve the robustness and reliability of the model in highly polluted environments.

[0062] S702, introduce Huber loss function to optimize model training; Huber loss provides a smooth transition between square loss and absolute loss by using a threshold Control the sensitivity of the loss function. For small errors, the function uses square loss, and for large errors it switches to linear loss, thereby reducing excessive penalties for abnormal labels. The specific expression is:

[0063] in, is the true value, is the predicted value. This method is particularly suitable for processing contaminated data and reduces the sensitivity of the model to abnormal data.

[0064] S703, implementing the Log-Cosh loss function; This loss function provides different penalty mechanisms for small and large errors. The Log-Cosh loss function exhibits continuous and progressive characteristics and is less sensitive to large errors, thereby improving the robustness of the model. The expression is as follows:

[0065] in, is the true value, is the predicted value, is the hyperbolic cosine function, which is an even function.

[0066] This loss function combines the characteristics of mean square error and the advantages of mean absolute error, which can effectively mitigate the impact of large errors and is suitable for use in contaminated label environments. The shape of this function is similar to square error loss, but the penalty for large errors grows more slowly, so that when facing extreme values ​​or outliers, it will not give too large a penalty.

[0067] S704: Based on the introduction of the above loss function, a comprehensive evaluation of the model performance is performed.

[0068] Adjust model training hyperparameters, such as learning rate and batch size, to adapt to the new loss function. Monitor the model's performance under different pollution levels and further optimize the algorithm accordingly to ensure that the model can maintain high recognition accuracy and robustness when facing polluted data.

[0069] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as systems, methods or program products. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, which may be collectively referred to herein as "circuits", "modules" or "platforms".

[0070] In another embodiment of the present invention, a deep learning system for electromagnetic signal classification and identification is provided. The system can be used to implement the above-mentioned deep learning method for electromagnetic signal classification and identification. Specifically, the deep learning system for electromagnetic signal classification and identification includes a data module, a construction module, a contamination module, a training module and an identification module.

[0071] Among them, the data module constructs an electromagnetic signal data set, each of which contains multi-dimensional electromagnetic signal features and corresponding label information; Build modules, select and configure the corresponding deep learning model; The pollution module performs label pollution on the electromagnetic signal dataset, including Y individual level pollution and M matrix level pollution. The Y individual level pollution randomly selects a specific proportion of data points and modifies their labels. The M matrix level pollution systematically modifies the labels of all or part of the dataset by applying a predefined pollution matrix. The training module uses a deep learning model trained on an electromagnetic signal dataset with pollution labels; The recognition module inputs the uncontaminated electromagnetic signal dataset into the obtained deep learning model and outputs the recognized electromagnetic signal.

[0072] In another embodiment of the present invention, a terminal device is provided, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the deep learning method for electromagnetic signal classification and recognition, including: Construct electromagnetic signal datasets, each of which contains multi-dimensional electromagnetic signal features and corresponding label information; select and configure the corresponding deep learning model; perform label contamination on the obtained electromagnetic signal datasets, including Y individual level and M matrix level contamination. The Y individual level contamination is performed by randomly selecting a specific proportion of data points and modifying their labels. The M matrix level contamination is performed by systematically modifying the labels of all or part of the datasets by applying a predefined contamination matrix; use the obtained electromagnetic signal dataset with contaminated labels to train the obtained deep learning model; input the uncontaminated electromagnetic signal dataset into the obtained deep learning model, and output the identified electromagnetic signal.

[0073] In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device, and can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that more specific examples (non-exhaustive list) of the computer-readable storage medium here include: an electrical connection with one or more wires, a portable 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.

[0074] Computer readable storage media also include data signals propagated in baseband or as part of a carrier wave, which carry readable program codes. Such propagated data signals can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the readable storage medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0075] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0076] One or more instructions stored in a computer-readable storage medium may be loaded and executed by a processor to implement the corresponding steps of the deep learning method for electromagnetic signal classification and recognition in the above embodiment; one or more instructions in the computer-readable storage medium may be loaded and executed by a processor as follows: Construct electromagnetic signal datasets, each of which contains multi-dimensional electromagnetic signal features and corresponding label information; select and configure the corresponding deep learning model; perform label contamination on the obtained electromagnetic signal datasets, including Y individual level and M matrix level contamination. The Y individual level contamination is performed by randomly selecting a specific proportion of data points and modifying their labels. The M matrix level contamination is performed by systematically modifying the labels of all or part of the datasets by applying a predefined contamination matrix; use the obtained electromagnetic signal dataset with contaminated labels to train the obtained deep learning model; input the uncontaminated electromagnetic signal dataset into the obtained deep learning model, and output the identified electromagnetic signal.

[0077] See also Fig.10 , the terminal device is a computer device, and the computer device 60 of this embodiment includes: a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the computer program 63 is executed by the processor 61, the deep learning method for electromagnetic signal classification and recognition in the embodiment is implemented. To avoid repetition, it is not described one by one here. Alternatively, when the computer program 63 is executed by the processor 61, the functions of each model / unit in the deep learning system for electromagnetic signal classification and recognition in the embodiment are implemented. To avoid repetition, it is not described one by one here.

[0078] The computer device 60 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art will appreciate that Fig.10 This is only an example of the computer device 60 and does not constitute a limitation of the computer device 60. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.

[0079] The processor 61 may be a central processing unit (CPU), or other general-purpose processors, central processing units, graphics processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, data processing logic based on quantum computing, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0080] The memory 62 may be an internal storage unit of the computer device 60, such as a hard disk or memory of the computer device 60. The memory 62 may also be an external storage device of the computer device 60, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the computer device 60.

[0081] Furthermore, the memory 62 may include both an internal storage unit of the computer device 60 and an external storage device. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 may also be used to temporarily store data that has been output or is to be output.

[0082] Any reference to a memory, database or other medium used in the embodiments provided in the present application may include at least one of a non-volatile and a volatile memory. Non-volatile memory may include a read-only memory (ROM), a magnetic tape, a floppy disk, a flash memory, an optical memory, a high-density embedded non-volatile memory, a resistive random access memory (ReRAM), a magnetic random access memory (MRAM), a ferroelectric random access memory (FRAM), a phase change memory (PCM), a graphene memory, etc. Volatile memory may include a random access memory (RAM) or an external cache memory, etc. As an illustration and not limitation, RAM may be in various forms, such as a static random access memory (SRAM) or a dynamic random access memory (DRAM), etc.

[0083] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0084] See also Fig.11 The terminal device 600 is an electronic device, and the electronic device is in the form of a general computing device. The components of the electronic device may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0085] The storage unit stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 performs the steps according to various exemplary embodiments of the present invention described in the above method section of this specification. For example, the processing unit 610 can perform the following steps: Figure 1 Follow the steps shown in .

[0086] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0087] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0088] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0089] The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 650. Furthermore, the electronic device 600 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 via a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0090] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings here can usually be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0091] See also Figure 4 ,This figure shows that when performing 18 repeated experiments, the contamination rate gradually increases with a gradient of 5%, ranging from 5% to 90%. The horizontal axis is the signal-to-noise ratio (SNR), ranging from -20 to 18, and the vertical axis is the accuracy of the model. These experiments simulate the impact of different label contamination ratios on model performance by applying a simple M matrix method.

[0092] See also Figure 5 ,This figure shows that when performing 18 repeated experiments, the contamination rate gradually increases with a gradient of 5%, ranging from 5% to 90%. The horizontal axis is the signal-to-noise ratio (SNR), ranging from -20 to 18, and the vertical axis is the accuracy of the model. These experiments simulate the impact of different label contamination ratios on model performance by applying the random M matrix method.

[0093] See also Figure 6 , This figure shows the relationship between the label contamination ratio and the model accuracy. The experiment was conducted 18 times, and the contamination rate gradually increased by 5%, covering a range from 5% to 90%. The figure contains two sets of data, which are the experimental results under the simple M matrix and random M matrix methods. The dataset used is the RML2016.10a dataset.

[0094] See also Figure 7 ,This figure shows the relationship between the label contamination ratio and the model accuracy. The experiment was conducted 18 times, and the contamination rate gradually increased by 5%, covering a range from 5% to 90%. The figure includes two sets of data, corresponding to the experimental results under the simple M matrix and random M matrix methods. The dataset used is the WiFi signal dataset.

[0095] See also Figure 8 ,This figure shows the relationship between the label pollution ratio and the model accuracy. The experiment was conducted 18 times, and the pollution rate gradually increased by 5%, covering the range from 5% to 90%. The figure includes two sets of data, corresponding to the experimental results of simple M matrix, random M matrix, Y individual symmetric method, Y individual asymmetric, and random flip method. The dataset used is the Bluetooth individual dataset.

[0096] See also Fig. 9By introducing three label pollution strategies on these three datasets, the model can achieve a wider range of signal coverage, especially when dealing with diverse signal patterns and complex signal environments, showing stronger robustness and generalization. Each dataset corresponds to a different communication scenario, including long-distance radio modulation recognition, close-range WiFi environment perception, and individual signal characteristics of Bluetooth devices. Based on these datasets, the model can not only process common wireless communication signals, but also accurately identify and classify signals in more complex and diverse environments through different model architectures; this integration method of multiple signal sources and different label pollution strategies improves the model's adaptability and anti-interference ability when dealing with complex signal problems.

[0097] In summary, the present invention provides a deep learning method and system for electromagnetic signal classification and recognition, aiming to improve the robustness and efficiency of electromagnetic signal classification; by simulating different levels of label contamination, the present invention can test and enhance the performance of the model in the face of different contamination conditions, thereby improving reliability in practical applications; in addition, the present invention adopts advanced deep learning technology, and optimizes the training process by dynamically adjusting the learning strategy and adaptive loss function, thereby reducing the demand for computing resources and shortening the model training time; this not only enhances the model's ability to process multi-channel data, but also greatly improves the overall classification efficiency.

[0098] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0099] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0100] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0101] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0102] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0103] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0104] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0105] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0106] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0107] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0108] The above contents are only for explaining the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A deep learning method for electromagnetic signal classification and recognition, characterized in that: The following steps are involved: S1. Construct electromagnetic signal datasets, each of which contains multi-dimensional electromagnetic signal features and corresponding label information; S2. Select and configure the corresponding deep learning model; S3, label contamination of the electromagnetic signal data set obtained in step S1, including Y individual level contamination and M matrix level contamination, wherein the Y individual level contamination is performed by randomly selecting a specific proportion of data points and modifying their labels, and the M matrix level contamination is performed by systematically modifying the labels of all or part of the data set by applying a predefined contamination matrix; S4, using the electromagnetic signal data set with pollution labels obtained in step S3 to train the deep learning model obtained in step S2; S5. Input the uncontaminated electromagnetic signal data set into the deep learning model obtained in step S4, and output the identified electromagnetic signal.

2. The deep learning method for electromagnetic signal classification and recognition according to claim 1, characterized in that: The magnetic signal datasets include the RML2016.10a dataset, the WiFi signal dataset, and the Bluetooth individual dataset.

3. The deep learning method for electromagnetic signal classification and recognition according to claim 1, characterized in that: The deep learning models include MCLDNN and CNN5. MCLDNN is used to process the RML dataset with multi-channel data input, and CNN5 is used to process the WiFi signal dataset and the Bluetooth individual dataset.

4. The deep learning method for electromagnetic signal classification and recognition according to claim 1, characterized in that: The specific pollution level of Y is: Implement a label pollution function `add_symmetric_noise`, which systematically introduces symmetric noise into the label data; by randomly relabeling the sample labels in each category according to the specified noise ratio `noise_rate`; Introduce asymmetric noise through the function `add_asymmetric_noise` to modify the class label according to the predefined mislabeled pairs (source, target); Use `flip_labels` to perform label flipping, randomly select a certain proportion of data points and change their labels to any other labels except the current labels, so as to increase the diversity of data and test the robustness of the model to noise.

5. The deep learning method for electromagnetic signal classification and recognition according to claim 4, characterized in that: Using `flip_labels` to flip labels is as follows: in, for, for, is the number of signal categories.

6. The deep learning method for electromagnetic signal classification and recognition according to claim 1, characterized in that: The M matrix level pollution is specifically: Construct the pollution matrix using the `construct_M_matrix_simple` method ; Implement `construct_M_matrix_block` method for total number of categories When it can be evenly divided, a block diagonal contamination matrix is ​​constructed; Construct a random pollution matrix through the `construct_M_matrix_random` method; Deploy the `LinearContaminationModel` class to take the total number of categories , a predefined pollution matrix , and the original probability of the category and the class probability after contamination Initialize the parameters; calculate the pollution weight of each category and make a random selection in the pollution matrix according to the weight to generate pollution labels that meet the settings of the pollution matrix, thereby achieving systematic pollution of the original data labels.

7. The deep learning method for electromagnetic signal classification and recognition according to claim 6, characterized in that: In constructing a block diagonal contamination matrix, for even numbers , using a size of The all-ones matrix Build two identical blocks; for odd numbers , the last block size is adjusted to , ensuring that all categories are appropriately considered.

8. The deep learning method for electromagnetic signal classification and recognition according to claim 6, characterized in that: Constructing a random pollution matrix is ​​as follows: Generate a The random number matrix , and then normalize it by summing the rows to ensure that the sum of the elements in each row is 1; times the identity matrix and times the random matrix Linear combination to form the final random contamination matrix , random contamination matrix Can be used with probability Randomly contaminate the class label to any other class.

9. The deep learning method for electromagnetic signal classification and recognition according to claim 1, characterized in that: Enhance model robustness at a specific contamination rate; introduce Huber loss function to optimize model training; implement Log-Cosh loss function; and optimize deep learning models.

10. A deep learning system for electromagnetic signal classification and recognition, characterized in that: include: Data module, constructs electromagnetic signal data sets, each of which contains multi-dimensional electromagnetic signal features and corresponding label information; Build modules, select and configure the corresponding deep learning model; The pollution module performs label pollution on the electromagnetic signal dataset, including Y individual level pollution and M matrix level pollution. The Y individual level pollution randomly selects a specific proportion of data points and modifies their labels. The M matrix level pollution systematically modifies the labels of all or part of the dataset by applying a predefined pollution matrix. The training module uses a deep learning model trained on an electromagnetic signal dataset with pollution labels; The recognition module inputs the uncontaminated electromagnetic signal dataset into the obtained deep learning model and outputs the recognized electromagnetic signal.

Images