Impulse noise suppression method for power line carrier communication system

Abstract

The invention relates to the technical field of power line data communication, in particular to an impulse noise suppression method for a power line carrier communication system, the method is applied to a noise suppression system, and the method comprises the following steps: establishing digital communication with the power line carrier communication system, and acquiring a communication data set within a preset time period based on the digital communication; the communication data set comprises data points acquired according to a preset acquisition frequency in a preset time period; each data point comprises a plurality of features; establishing a communication feature set based on the communication data set; optimizing a preset deep learning network model through the communication feature set to obtain an optimized deep learning network model; reading a current data point, and inputting the current data point into the optimized deep learning network model to obtain a predicted configuration parameter; and performing noise suppression processing on the power line carrier signal in the current data point according to the predicted configuration parameter.

Description

Technical Field

[0001] The present invention relates to the technical field of power line data communication, and in particular to a method for suppressing pulse noise in a power line carrier communication system. Background Art

[0002] Power line carrier communication (PLC) is a communication technology that transmits data through power lines and is widely used in smart grids, home automation and industrial control. However, there are various noise interferences in the power line communication environment, especially impulse noise, which has a significant impact on the communication quality. The randomness and transient characteristics of this noise make it difficult for traditional filtering and averaging methods to completely and effectively suppress it, resulting in reduced communication efficiency and increased data transmission error rate.

[0003] Existing noise suppression methods usually rely on fixed rules or simple statistical characteristics, which are difficult to adapt to complex and changing communication environments. With the development of artificial intelligence and deep learning technology, combining machine learning methods with communication noise suppression has become a new research direction, which can dynamically analyze the communication environment and optimize it in a targeted manner. However, how to efficiently extract communication features and build a reliable noise suppression mechanism in a complex power line environment is still a key difficulty in the current technological development. Summary of the invention

[0004] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method for suppressing pulse noise in a power line carrier communication system.

[0005] In order to achieve the above object, the main technical solutions adopted by the present invention include: An embodiment of the present invention provides a method for suppressing impulse noise in a power line carrier communication system, the method being applied to a noise suppression system, the method comprising: Establishing digital communication with a power line carrier communication system, and acquiring a communication data set within a preset time period based on the digital communication; The communication data set includes data points collected at a preset collection frequency within a preset time period; each data point includes multiple features; Based on the communication data set, establishing a communication feature set; Optimizing a preset deep learning network model through the communication feature set to obtain an optimized deep learning network model; The deep learning network model includes a deep learning network and an additional attention layer embedded in the deep learning network for optimizing the deep learning network during the training process of the deep learning network model; Read the current data points and input them into the optimized deep learning network model to obtain the predicted configuration parameters; The configuration parameters include precoding parameters, adaptive spectrum and subcarrier mapping; Noise suppression processing is performed on the power line carrier signal in the current data point according to the predicted configuration parameters.

[0006] Preferably, establishing a communication feature set based on the communication data set specifically includes: Preprocessing the data points in the communication data set to obtain a preprocessed communication data set; Based on the features in the preprocessed communication data set, obtain the corresponding covariance matrix; Performing eigenvalue decomposition on the covariance matrix to obtain a plurality of data groups, wherein each data group includes an eigenvalue and its corresponding eigenvector; Selecting eigenvectors corresponding to N eigenvalues ​​in a plurality of data groups according to a preset strategy, and projecting each data point in the preprocessed communication data set onto each of the screened eigenvectors based on the screened eigenvectors to obtain a reduced-dimensional data representation corresponding to the data point; The dimensionality-reduced data representations corresponding to all data points in the preprocessed communication data set are used to form a communication feature set.

[0007] Preferably, selecting the feature vectors corresponding to the N feature values ​​in the plurality of data groups according to a preset strategy specifically includes: For each eigenvalue and the corresponding eigenvector in each data group, the target value of the data group is obtained using formula (1); Wherein, formula (1) is: ; Q is the target value of the data set; is the characteristic value in the data set; is the eigenvalue in the data set The variance of the corresponding eigenvector; M is the sum of the eigenvalues ​​in all data sets; is the preset weight parameter; Based on the target value of each data group, the eigenvectors corresponding to the eigenvalues ​​in the first N data groups with the largest target values ​​are selected.

[0008] Preferably, preprocessing the communication data set to obtain a preprocessed communication data set specifically includes: Screening out data points whose time series attenuation coefficients meet a preset threshold value from the communication data set to form an initial communication data set; Screening the initial communication data set according to a preset screening strategy to obtain a preprocessed communication data set; Among them, the time series attenuation coefficient of the data point is calculated using formula (2); The formula (2) is: ; in, is a preset first decay rate; is a preset first weight coefficient; is a preset second decay rate; is a preset second weight coefficient; is a preset third weight coefficient; t is the time from the collection moment of the data point corresponding to the time series attenuation coefficient to the current moment; It is the time series attenuation coefficient corresponding to the data point when the time between the collection moment and the current moment is t.

[0009] Preferably, the screening of the initial communication data set according to a preset screening strategy to obtain a preprocessed communication data set specifically includes: Screening out data points that meet similar conditions in the initial communication data set, and marking the screened data points that meet similar conditions with screening identifiers; Using a step window, the trend adaptation analysis is performed on the data points with the filter mark in the initial communication data set, so that each data point with the filter mark gets a corresponding fitness score; Filtering the data points with the filter identifier in the initial communication data set to obtain a first data point set; The first data point set consists of data points with screening identifiers whose fitness scores are greater than a preset fitness value; Obtaining a comprehensive score corresponding to each data point in the first data point set according to the fitness score corresponding to the data point and a preset credibility score; The data points in the first data point set whose comprehensive scores are greater than a preset threshold and the preset configuration parameters corresponding to the data points one by one form a preprocessed communication data set.

[0010] Preferably, a step window is used to perform trend adaptation analysis on the data points with screening identifiers in the initial communication data set, so that each data point with screening identifiers obtains a corresponding fitness score, specifically including: Using stepped windows, the initial communication data set is divided into multiple consecutive or overlapping time windows, each time window contains a preset number of data points; In each step window, a predetermined fitting algorithm is used to perform trend fitting on the designated features of the data points with the screening mark, and a fitting trend line corresponding to the designated features is generated. The predicted value of the designated feature of each data point with the screening mark is calculated according to the fitting trend line, and the fitness score of the data point is further calculated based on the predicted value of the designated feature. The specified characteristics include: signal strength, noise level, signal-to-noise ratio; Wherein, the predetermined fitting algorithm is a linear fitting algorithm.

[0011] Preferably, the step of calculating the fitness score of the data point based on the predicted value of the specified feature specifically includes: Based on the predicted value of the specified feature, the fitness score of the data point is calculated using formula (3); The formula (3) is: ; ; For the first i The fitness score of each data point; For the first i The true value of the signal strength in the data points; For the first i The predicted value of the signal strength in the data points; For the first i The true value of the noise level in the data points; For the first i The predicted value of the noise level in the data points; For the first i The true value of the signal-to-noise ratio in the data points; For the first i The predicted value of the signal-to-noise ratio in the data points; is the preset maximum deviation value of signal strength; is the preset maximum deviation value of the noise level; is the preset maximum deviation value of the signal-to-noise ratio; is the time position weight, which indicates the importance of the i-th data point with the screening mark in the step window; is the collection time corresponding to the i-th data point with the screening mark; is the center time point of the step window; is a preset parameter value.

[0012] Preferably, obtaining a comprehensive score corresponding to each data point in the first data point set according to the fitness score corresponding to the data point and a preset credibility score specifically includes: According to the fitness score and the preset credibility score corresponding to each data point in the first data point set, the comprehensive score corresponding to the data point is obtained using formula (4); The formula (4) is: ; is the comprehensive score corresponding to the i-th data point with the screening mark; is the average of the fitness scores corresponding to all data points with the screening mark; is the standard deviation of the fitness scores corresponding to all data points with the screening mark; is the preset credibility score corresponding to the i-th data point with the screening mark; is the average of the credibility scores corresponding to all data points with the screening mark; is the standard deviation of the credibility scores corresponding to all data points with the screening mark; is the pre-set fitness weight coefficient; is a pre-set credibility weight coefficient.

[0013] Preferably, Among them, the deep learning network model is obtained by embedding an additional attention layer between the feature extraction layer and the fully connected layer in the convolutional neural network; The attention addition layer adopts any one of the self-attention mechanism, the soft attention mechanism and the hard attention mechanism.

[0014] The beneficial effects of the present invention are as follows: a method for suppressing pulse noise in a power line carrier communication system of the present invention analyzes environmental characteristics through a deep learning network model combined with a communication feature set, optimizes the parameter configuration of noise suppression, and suppresses pulse noise in power line communication in a targeted manner. Compared with the prior art, it can significantly improve the suppression effect. At the same time, the configuration parameters (such as precoding parameters, adaptive spectrum, and subcarrier mapping) are dynamically acquired, so that the communication system can adapt to the complex and changeable power line environment, effectively reduce the bit error rate, and improve communication stability. In addition, the current data point is input and the configuration parameters are predicted according to the optimized model to ensure that the communication system can quickly respond to environmental changes and maintain high-quality data transmission. The noise in the power line carrier communication environment is complex and changeable. The prediction ability of the deep learning model can accurately respond to various noise characteristics, making the system more suitable for actual application scenarios, and the deep learning network model embedded with the additional attention layer enhances the focus on key features, can adaptively optimize, and adapt to the noise characteristics in different environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flow chart of a method for suppressing pulse noise in a power line carrier communication system according to the present invention; Figure 2 It is a schematic diagram of the structure of an exemplary electronic device of the present invention.

[0016] Reference numerals

[0017] 300: bus; 301: receiver; 302: processor; 303: transmitter; 304: memory; 305: bus interface. DETAILED DESCRIPTION

[0018] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0019] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0020] Embodiment 1

[0021] For example, in a smart meter network in a residential area, the noise suppression system ensures accurate transmission of energy metering data by monitoring communication interference on the line, avoiding billing errors or data loss caused by pulse noise interference. Or, in an industrial plant, the noise suppression system monitors the high-frequency pulse interference generated when large motors start or stop, ensuring the stability of communication between production equipment.

[0022] See also Figure 1 This embodiment provides a method for suppressing pulse noise in a power line carrier communication system. The method is applied to a noise suppression system. The method includes: Establishing digital communication with a power line carrier communication system, and acquiring a communication data set within a preset time period based on the digital communication; In this embodiment, the noise suppression system exchanges data with the power line carrier communication system through a digital communication interface, and collects communication data at a preset period (such as hours or days). This enables the noise suppression system to systematically collect and analyze communication data within a time period, ensure a comprehensive understanding of the communication process, ensure that the communication data of the power line carrier communication system can be continuously monitored and collected, and provide sufficient data support for subsequent noise suppression.

[0023] The communication data set includes data points collected at a preset collection frequency within a preset time period; each data point includes multiple features; for example: signal strength, noise level, signal-to-noise ratio, and communication error rate.

[0024] Based on the communication data set, a communication feature set is established; through the extraction of the feature set, the dimension of the data is compressed, and the main information in the communication process can be better reflected, thereby enhancing the accuracy of the noise suppression process.

[0025] Optimizing a preset deep learning network model through the communication feature set to obtain an optimized deep learning network model; In which, the deep learning network model includes a deep learning network and an additional attention layer embedded in the deep learning network for optimizing the deep learning network during the training process of the deep learning network model; in this embodiment, the additional attention layer enhances the network's responsiveness to important communication features and can better identify important information related to noise suppression.

[0026] Read the current data points and input them into the optimized deep learning network model to obtain the predicted configuration parameters; respond in real time and predict the appropriate configuration parameters so that noise suppression can quickly adapt to the current communication status and improve the system's dynamic adjustment capabilities. Configuration parameters determine key settings in communication, including precoding, spectrum selection, and subcarrier mapping, which directly affect the stability and anti-interference capabilities of communication.

[0027] The configuration parameters include precoding parameters, adaptive spectrum and subcarrier mapping; The power line carrier signal in the current data point is subjected to noise suppression processing according to the predicted configuration parameters. This embodiment adjusts signal transmission according to the predicted configuration parameters and performs noise suppression to reduce interference and optimize the power line carrier signal quality.

[0028] In this embodiment, in the smart grid, the smart meters of each household and enterprise send energy consumption data to the power company through power line carrier communication. Due to the complex power line environment, it may be interfered by pulse noise (such as high-power equipment startup, power outage recovery current inflow, etc.), resulting in unstable data transmission. The noise suppression system predicts and adjusts the communication configuration parameters in real time to ensure the accurate transmission of meter data. Reduce electricity metering errors and user disputes caused by communication errors. Alternatively, in industrial parks or factories, power line carrier communication is widely used for signal transmission between sensors, actuators and controllers. However, in industrial environments, equipment such as motors, welding machines or frequency converters will generate strong pulse noise, which interferes with communication signals. The deep learning network model optimized by the noise suppression system can quickly adapt to complex industrial environments, achieve high-reliability communication, and ensure synchronization and real-time monitoring between production line equipment. And, in rural power grid transformation projects, remote control equipment (such as voltage adjustment devices or distribution switches) are remotely operated through power line carrier communication. However, since rural power grid lines are long and old, they are easily affected by power equipment switching or external interference. Through the dynamic adjustment capability of the noise suppression system, communication parameters can be optimized in real time, the anti-interference performance of long-distance power line communications can be improved, and the accuracy and timeliness of remote control operations can be guaranteed.

[0029] Specifically, based on the communication data set, a communication feature set is established, which specifically includes: Preprocessing the data points in the communication data set to obtain a preprocessed communication data set; Based on the features in the preprocessed communication data set, obtain the corresponding covariance matrix; Performing eigenvalue decomposition on the covariance matrix to obtain a plurality of data groups, wherein each data group includes an eigenvalue and its corresponding eigenvector; Selecting eigenvectors corresponding to N eigenvalues ​​in a plurality of data groups according to a preset strategy, and projecting each data point in the preprocessed communication data set onto each of the screened eigenvectors based on the screened eigenvectors to obtain a reduced-dimensional data representation corresponding to the data point; The dimensionality-reduced data representations corresponding to all data points in the preprocessed communication data set are used to form a communication feature set.

[0030] For example, assume that a communication data set of a power line carrier communication system includes the following features: signal strength, noise level, communication error rate, and a data set containing 100 data points is collected. The data points in the communication data set are preprocessed to obtain a preprocessed communication data set; assume that there are 80 data points in the preprocessed communication data set.

[0031] Then, the corresponding covariance matrix is ​​obtained for the features in the preprocessed communication data set. Assume that the covariance matrix Cov of the three features in the preprocessed communication data set is: ; Then, the covariance matrix is ​​decomposed into eigenvalues ​​to obtain the following three eigenvalues ​​and their corresponding eigenvectors: Eigenvalues: λ1=2.0; λ2=1.5; λ3=0.6; Feature vectors: v1=[0.8,0.4,0.4]; v2=[0.3,0.9,0.3]; v3=[0.1,0.1,0.9]; According to the preset strategy (such as selecting two eigenvalues), the corresponding eigenvector v1 and eigenvector v2 are selected. Then, each data point is projected onto the eigenvector v1 and eigenvector v2, and the reduced-dimensional data representation is calculated. Finally, the reduced-dimensional data representation corresponding to all data points in the preprocessed communication data set is combined into a communication feature set.

[0032] In this embodiment, high-dimensional data is compressed into a low-dimensional feature set by decomposing and projecting eigenvalues, thereby reducing computational complexity. Feature vector selection focuses on the main direction of change, which helps to more efficiently identify noise characteristics and improve communication quality. By extracting and reducing the dimension of key features, the feature set can more effectively cope with complex noise environments and provide more accurate data support for subsequent noise suppression strategies.

[0033] Specifically, the eigenvectors corresponding to the N eigenvalues ​​are selected from the multiple data groups according to a preset strategy, including: For each eigenvalue and the corresponding eigenvector in each data group, the target value of the data group is obtained using formula (1); Wherein, formula (1) is: ; Q is the target value of the data set; is the characteristic value in the data set; is the eigenvalue in the data set The variance of the corresponding eigenvector; M is the sum of the eigenvalues ​​in all data sets; is the preset weight parameter; Based on the target value of each data group, the eigenvectors corresponding to the eigenvalues ​​in the first N data groups with the largest target values ​​are selected.

[0034] Formula (1) in this embodiment is used to optimize the target value Q according to a pre-set strategy by analyzing the eigenvalues ​​and eigenvectors, thereby providing more accurate input for tasks such as noise suppression. It helps to identify noise fluctuations and reduce interference to the signal by optimizing the processing method. In this embodiment, the data dimension is greatly reduced through dimensionality reduction, thereby reducing the computational complexity. Assuming that the original data point has 10 features, only 3 features may be needed for processing after dimensionality reduction. In this way, the subsequent calculation speed will be greatly accelerated. Selecting the eigenvector with the largest eigenvalue can retain the core information in the data point. This helps to remove irrelevant data noise, such as unimportant background noise or signal fluctuations in communication, thereby improving the accuracy of subsequent algorithms (such as deep learning models). The data set after dimensionality reduction is more concise, reduces noise interference, and helps to more accurately predict and suppress noise, thereby improving communication quality.

[0035] Preprocessing the communication data set to obtain a preprocessed communication data set; specifically comprising: Screening out data points whose time series attenuation coefficients meet a preset threshold value from the communication data set to form an initial communication data set; Screening the initial communication data set according to a preset screening strategy to obtain a preprocessed communication data set; Among them, the time series attenuation coefficient of the data point is calculated using formula (2); The formula (2) is: ; in, is a preset first decay rate; is a preset first weight coefficient; is a preset second decay rate; is a preset second weight coefficient; is a third pre-set weight coefficient; t is the time from the acquisition moment of the data point corresponding to the time series attenuation coefficient to the current moment; It is the time series attenuation coefficient corresponding to the data point when the time between the collection moment and the current moment is t.

[0036] Specifically, the calculation formula of the time series decay coefficient is mainly used for time series screening of data points. Formula (2) adjusts the time series characteristics of the data points through the decay rate and weight coefficient, and performs weighted decay based on time (t) and the moment of data collection, thereby affecting the weight of the data point and its importance. Specifically, the time series decay coefficient of the data point gradually decays over time, which can effectively screen out data points that are "closer" to the current moment and have higher reference value, thereby improving the quality of the data set during the preprocessing process, reducing unnecessary noise effects, and optimizing the efficiency and accuracy of data processing.

[0037] The initial communication data set is screened according to a preset screening strategy to obtain a preprocessed communication data set, specifically including: Screening out data points that meet similar conditions in the initial communication data set, and marking the screened data points that meet similar conditions with screening identifiers; By marking and filtering the identifiers, the scope of the target data points in the initial communication data set is clarified, the interference of irrelevant data is reduced, and the analysis efficiency is improved.

[0038] A stepped window is used to perform trend adaptation analysis on the data points with filtering marks in the initial communication data set, so that each data point with filtering marks obtains a corresponding fitness score; a stepped window method is introduced, and the trend changes of data points are evaluated in combination with the fitness scores, so that the filtered data are more in line with actual needs and reflect the dynamic characteristics of the data.

[0039] Filtering the data points with the filter identifier in the initial communication data set to obtain a first data point set; The first data point set consists of data points with screening identifiers whose fitness scores are greater than a preset fitness value; According to the fitness score and the preset credibility score corresponding to each data point in the first data point set, the comprehensive score corresponding to the data point is obtained; the comprehensive score is calculated by the fitness score and the credibility score, and the data points are further filtered to ensure that the remaining data is of higher quality and more representative.

[0040] The data points in the first data point set whose comprehensive scores are greater than a preset threshold and the preset configuration parameters corresponding to the data points are used to form a preprocessed communication data set. The corresponding configuration parameters are retained on the basis of the data points to facilitate subsequent operations and analysis, forming a more complete preprocessed data set.

[0041] This embodiment combines time series analysis and trend adaptability assessment to filter out data points that meet specific similarity conditions and have obvious trend characteristics. The dual screening mechanism of fitness and comprehensive score further improves the representativeness and accuracy of data points, while reducing the interference of noise data, and is suitable for multiple scenario requirements in communication data set processing, such as anomaly detection, trend prediction or pattern recognition.

[0042] In the practical application of this embodiment, a step window is used to perform trend adaptation analysis on the data points with screening identifiers in the initial communication data set, so that each data point with screening identifiers obtains a corresponding fitness score, specifically including: Using stepped windows, the initial communication data set is divided into multiple consecutive or overlapping time windows, each time window contains a preset number of data points; In this embodiment, the step window partitioning decomposes the initial communication data set into multiple small time windows, so that the local trend can be analyzed separately in each window. This segmented analysis can better capture short-term fluctuations or local changes, which are not easily obscured by the overall trend.

[0043] In each step window, the specified features of the data points with screening marks are respectively fitted with a predetermined fitting algorithm to generate a fitting trend line corresponding to the specified features, and the predicted value of the specified features of each data point with screening marks is calculated based on the fitting trend line, and the fitness score of the data point is further calculated based on the predicted value of the specified features; specifically, by screening the identified data points, the analysis is focused on data points with specific features or importance, rather than a comprehensive analysis of the entire data set. This selective analysis method can significantly improve efficiency and analysis accuracy, and quantify the importance or degree of deviation of each data point through the fitness score.

[0044] The specified characteristics include: signal strength, noise level, signal-to-noise ratio; The predetermined fitting algorithm is a linear fitting algorithm. The predetermined fitting algorithm in this embodiment may also be a nonlinear fitting algorithm, which is not specifically limited.

[0045] Preferably, the step of calculating the fitness score of the data point based on the predicted value of the specified feature specifically includes: Based on the predicted value of the specified feature, the fitness score of the data point is calculated using formula (3); The formula (3) is: ; ; For the first i The fitness score of each data point; For the first i The true value of the signal strength in the data points; For the first i The predicted value of the signal strength in the data points; For the first i The true value of the noise level in the data points; For the first i The predicted value of the noise level in the data points; For the first i The true value of the signal-to-noise ratio in the data points; For the first i The predicted value of the signal-to-noise ratio in the data points; is the preset maximum deviation value of signal strength; is the preset maximum deviation value of the noise level; is the preset maximum deviation value of the signal-to-noise ratio; is the time position weight, which indicates the importance of the i-th data point with the screening mark in the step window; is the collection time corresponding to the i-th data point with the screening mark; is the center time point of the step window; is a pre-set parameter value. In this embodiment, formula (3) introduces multiple features (signal strength, noise level, signal quality) to comprehensively calculate the fitness score of the data point. This method is more comprehensive than a single feature and can more accurately reflect the overall quality or importance of the data point. At the same time, in formula (3) The weighted method introduces a time decay mechanism, which takes into account the time attribute of the data point. The farther away, the smaller its weight, highlighting the temporal correlation. This helps to give more weight to the latest data in trend analysis, thereby improving timeliness and accuracy. Formula (3) achieves quantitative evaluation of data point quality by calculating the deviation between the feature value and the predicted value, and combining the normalized range of the deviation (such as R1, R2, R3). This evaluation method is both specific and detailed.

[0046] The step of obtaining a comprehensive score corresponding to each data point in the first data point set according to the fitness score corresponding to the data point and a preset credibility score specifically includes: According to the fitness score and the preset credibility score corresponding to each data point in the first data point set, the comprehensive score corresponding to the data point is obtained using formula (4); The formula (4) is: ; For the first i The comprehensive score corresponding to each data point; is the average of the fitness scores corresponding to all data points with the screening mark; is the standard deviation of the fitness scores corresponding to all data points with the screening mark; For the first i The pre-set credibility score corresponding to each data point; is the average of the credibility scores corresponding to all data points with the screening mark; is the standard deviation of the credibility scores corresponding to all data points with the screening mark; is the pre-set fitness weight coefficient; is a pre-set credibility weight coefficient. In this embodiment, the fitness score and the credibility score are integrated by weighted combination, which not only considers the trend and deviation of the data point, but also introduces the credibility evaluation. This multi-factor fusion method surpasses the traditional single evaluation mode.

[0047] Among them, the deep learning network model is obtained by embedding an additional attention layer between the feature extraction layer and the fully connected layer in the convolutional neural network; The attention addition layer adopts any one of the self-attention mechanism, the soft attention mechanism and the hard attention mechanism.

[0048] In this embodiment, the first few layers of the convolutional neural network (CNN) extract low-level to high-level features, such as edges, textures, patterns, etc., from the input data through convolution and pooling operations. These features are passed to the fully connected layer for further processing. The fully connected layer maps high-dimensional features to the output space of the classification or regression task, and is a key link in completing the task. However, the fully connected layer of the traditional CNN is usually difficult to handle the long-distance dependencies of features or global context information. In this embodiment, an additional attention layer is embedded between the feature extraction layer and the fully connected layer to optimize feature selection and information flow transmission, thereby improving the performance of the model. Embedding the additional attention layer between the feature extraction layer and the fully connected layer of the CNN not only optimizes the selection and transmission of features, but also effectively solves the problem of fixed receptive field of the traditional CNN and insufficient global information capture of the fully connected layer.

[0049] In addition, this embodiment also provides an electronic device, comprising: at least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the power line carrier communication system pulse noise suppression method described in embodiment 1.

[0050] Attached Figure 2 This is a schematic diagram of the structure of an exemplary electronic device of this application. Figure 2 In the embodiment, the bus architecture is represented by bus 300, which may include any number of interconnected buses and bridges, and bus 300 connects various circuits including one or more processors represented by processor 302 and memory represented by memory 304. Bus 300 may also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. Bus interface 305 provides an interface between bus 300 and receiver 301 and transmitter 303. Receiver 301 and transmitter 303 may be the same element, namely a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 302 is responsible for managing bus 300 and general processing, while memory 304 may be used to store data used by processor 302 when performing operations.

[0051] Embodiment 2

[0052] This embodiment solves the technical problem in the prior art that the high peak-to-average ratio characteristic of OFDM (Orthogonal Frequency Division Multiplexing) modulation introduces nonlinear noise into the power line carrier communication system, resulting in a decrease in the quality of the communication signal, by providing a method for suppressing pulse noise in a power line carrier communication system. By configuring an additional attention layer to optimize the deep learning network, reading real-time communication data, configuring precoding parameters, adaptive spectrum and subcarrier mapping, and performing noise suppression processing, the technical effect of improving the signal quality of the power line carrier communication system is achieved. This embodiment provides a method for suppressing pulse noise in a power line carrier communication system, including: Step 1: Establish digital communication with the power line carrier communication system, and acquire a communication data set within a preset time period based on the digital communication.

[0053] Specifically, based on specific communication protocols and standards, ensure that data can be effectively transmitted between the two systems. Establish a digital communication connection between the power line carrier communication system and the noise suppression system. Set a time period, such as a few hours, a day, a week, or longer, depending on the frequency requirements of data analysis and application. On the basis of establishing digital communication, collect data points of communication data within the preset time period, including information such as signal strength, noise level, communication error rate, etc., which are essential for analyzing and optimizing the communication process. Perform pre-processing operations such as denoising, verification and formatting on the collected data points, and store the processed data points as communication data sets.

[0054] Step 2: Extract communication features based on the communication data set and establish a communication feature set.

[0055] Specifically, the communication data set is analyzed and the covariance matrix is ​​calculated to reflect the correlation between the various features in the communication data set. The covariance matrix is ​​subjected to eigenvalue decomposition to obtain a set of eigenvalues ​​and corresponding eigenvectors. The eigenvalues ​​represent the variance of the data in the direction of these eigenvectors, while the eigenvectors represent the main change directions of the data. Based on the size of the eigenvalues, the eigenvectors are sequentially screened, and the eigenvectors corresponding to the largest eigenvalues ​​are usually selected because they represent the main change directions in the data. Based on the screened eigenvectors, the features of the communication data set are extracted, and the original data is projected onto these eigenvectors to obtain a reduced-dimensional data representation. The extracted features are organized into a feature set, namely the communication feature set. Extracting key feature information from the original communication data and establishing a communication feature set can help the system better understand the communication environment and improve the stability and reliability of communication.

[0056] Step three: construct a deep learning network through big data, configure an additional attention layer through the communication feature set, and optimize the deep learning network through the additional attention layer.

[0057] Specifically, big data is used for data collection to establish a basic data set. The collected data is screened for data similarity, a similar screening identifier is established, and data points that may be inaccurate or insignificant are removed. The results of data preprocessing are randomly clustered, and the data set is divided into a training data set, a validation data set, and a test data set. A deep learning network is trained, which contains multiple layers, such as convolutional layers, fully connected layers, etc., to identify and process complex patterns in the data. An attention mechanism is selected as an additional layer, such as a self-attention mechanism, a soft attention mechanism, or a hard attention mechanism. The self-attention mechanism allows the model to focus on different parts of the input sequence, while the soft attention mechanism allows the model to learn a weight distribution, and the hard attention mechanism directly selects a specific part of the input sequence. The structure of the attention additional layer involved in the selected attention mechanism usually contains a sequence of query, key, and value, and an attention mechanism module. The attention additional layer is integrated into the deep learning network, usually by adding the attention additional layer after a specific layer of the network. The deep learning network is optimized by the attention additional layer, that is, model pruning optimization, using various techniques to identify and remove unimportant parts and reduce the complexity of the model. Model pruning is a technique that reduces the complexity of the model by removing unimportant connections or neurons in the network, which can reduce the number of model parameters, thereby reducing the computational cost and increasing the running speed of the model. During the pruning process, various techniques are used to identify and remove unimportant parts. For example, the importance of neurons is evaluated based on the model's weights, activation values, gradients, etc. After removing neurons or connections with low importance, the model needs to be retrained to adapt to the new structure, including adjusting the parameters of the network, improving the training algorithm, etc. Through model pruning optimization, the amount of model calculation can be reduced, thereby improving the local accuracy and response speed of the model. When the pruned model processes new data, the calculation speed will be faster due to the reduction in parameters, while maintaining a high prediction accuracy. The attention mechanism is a technique that can improve the model's ability to pay attention to and respond to important features. It can automatically learn the important parts of the input data and adjust the network's output accordingly. The attention additional layer can be applied to sequence data, image data, or text data to enhance the model's ability to identify key parts of the data. By configuring the attention additional layer, the network can process data more efficiently, especially when dealing with long-distance dependencies. By configuring additional attention layers in the deep learning network, the model's ability to identify key parts of the data can be improved, helping the model to automatically learn the important parts of the data, and significantly improving the local accuracy and response speed of the power line carrier communication system.

[0058] Step 4: Read real-time communication data, and configure precoding parameters, adaptive spectrum, and subcarrier mapping based on the real-time communication data according to the optimized deep learning network.

[0059] Specifically, sensors, network monitoring equipment, etc. are used to collect communication data in real time, including signal strength, noise level, communication error rate, etc. Models that have been trained and optimized through big data and deep learning networks are used to process real-time communication data. The optimized deep learning network can identify and understand the key features in the communication data, and configure precoding parameters, adaptive spectrum, and subcarrier mapping accordingly. Precoding parameters refer to specific settings used in the precoding process, such as coding algorithms, modulation methods, signal power allocation, etc., which can improve the signal's anti-interference ability and reduce transmission losses. Adaptive spectrum refers to the technology of dynamically selecting and adjusting the communication frequency according to the current spectrum conditions during the communication process. It can automatically identify and avoid interference frequency bands and select the best communication frequency, which helps to adapt to complex power line environments. Subcarrier mapping is a key step in orthogonal frequency division multiplexing technology. It determines how to allocate data to different subcarriers and improve the spectrum efficiency and anti-interference ability of the communication system. In the power line carrier communication system, subcarrier mapping maps the data stream to a series of subcarriers that are orthogonal to each other in the frequency domain. The performance of power line carrier communication system can be significantly improved through reasonable configuration and optimization of precoding parameters, adaptive spectrum and subcarrier mapping.

[0060] Step 5: Perform noise suppression processing on the power line carrier signal in the real-time communication data according to the configured precoding parameters, adaptive spectrum and subcarrier mapping.

[0061] Specifically, according to the output of the previously optimized deep learning network, the precoding parameters are configured, including the coding algorithm, modulation method, signal power allocation, etc., which helps to improve the signal's anti-interference ability and reduce transmission loss, thereby improving the communication quality. According to the configured parameters, the communication frequency is selected and adjusted, and adaptive spectrum and subcarrier mapping are performed. Adaptive spectrum technology can automatically identify and avoid interference frequency bands and select the best communication frequency. Subcarrier mapping maps the data stream to a series of subcarriers, which are mutually orthogonal in the frequency domain, which helps to improve spectrum efficiency and anti-interference ability. According to the configured precoding parameters, adaptive spectrum and subcarrier mapping, the power line carrier signal in the real-time communication data is subjected to noise suppression processing, and various signal processing techniques such as filter design, noise estimation and suppression, signal reconstruction, etc. are used to reduce the impact of noise and interference on the signal. By suppressing the noise of the power line carrier signal, the system can adapt to changes in the communication environment in real time and improve the stability of communication.

[0062] Furthermore, step one in this embodiment includes: obtaining a data iteration cycle of the power line carrier communication system; establishing a timing attenuation coefficient based on the data iteration cycle; performing data identification of a communication data set within a preset time period through the timing attenuation coefficient, and completing acquisition of the communication data set based on the data identification result.

[0063] Specifically, the appropriate data iteration cycle is determined according to the characteristics of the power line carrier communication system, such as signal stability, noise frequency, etc. The data iteration cycle refers to the time interval for the system to collect and update data, which determines the frequency of data update and the application scope of the timing decay coefficient. During this period, the system collects relevant data of the power line carrier communication, such as signal strength, noise level, communication error rate, etc. Based on the determined data iteration cycle, a timing decay coefficient is established to indicate the degree to which the importance or influence of the data decreases over time. As time goes by, the influence of old data will gradually weaken, while the influence of new data will increase. The communication data set within the preset time period is identified using the timing decay coefficient, and each data point is marked or classified according to the timestamp of the data and the timing decay coefficient to reflect its relative importance and influence. According to the data identification results, the required data is filtered and extracted from the data set, including selecting data within a specific time period, excluding outdated or unimportant data, etc., to complete the acquisition of the communication data set. The timing decay coefficient is a coefficient used to indicate the change of data importance over time. By establishing the timing decay coefficient, the system can more accurately evaluate the impact of historical data on the current communication status. By determining the data iteration cycle, establishing the time series attenuation coefficient, and performing data identification and acquisition, the system can more effectively utilize historical data and improve the accuracy of data processing.

[0064] In the specific application of this embodiment, step two includes: calculating and obtaining the covariance matrix of the communication data set; performing eigenvalue decomposition on the covariance matrix to establish eigenvalues ​​and mapping eigenvectors; sequentially screening the mapping eigenvectors based on the eigenvalues, and completing communication feature extraction based on the sequential screening results.

[0065] Specifically, the acquired communication data set is cleaned to remove invalid data and erroneous data. The data is normalized or standardized to ensure the accuracy of the calculation results. For the processed data, the covariance matrix is ​​calculated. The covariance matrix is ​​a symmetric matrix used to represent the correlation between the features in the data set, where each element represents the covariance between two features in the data set. The covariance matrix is ​​subjected to eigenvalue decomposition, which is a mathematical method used to decompose a matrix into a set of eigenvalues ​​and corresponding eigenvectors. The eigenvalue represents the variance of the data in the direction of these eigenvectors, while the eigenvector represents the main direction of change of the data. Mapping the eigenvector refers to the new representation obtained by projecting the original data onto the eigenvector, which can be regarded as converting the data from the original feature space to a new feature space, which is defined by the eigenvector.

[0066] The eigenvalues ​​are usually sorted from large to small. The size of the eigenvalue reflects the size of the variance of the data in the direction of the corresponding eigenvector, so a larger eigenvalue indicates that the data has a larger change in that direction. Based on the sorting results of the eigenvalues, the eigenvectors corresponding to the first few largest eigenvalues ​​are selected, representing the main direction of change in the data, so they are the most representative features in the data. Based on the screened eigenvectors, the features of the communication data set are extracted, and the original data is projected onto these eigenvectors to obtain a reduced-dimensional data representation. By calculating the covariance matrix, eigenvalue decomposition, eigenvector screening, and communication feature extraction, the complexity of the data is reduced and the performance of the communication system is improved.

[0067] Furthermore, step three of the present application includes: collecting data through big data to establish a basic data set; performing data preprocessing on the basic data set to establish data preprocessing results; performing random data clustering on the data preprocessing results, and establishing a training data set, a verification data set, and a test data set based on the data random clustering results; and completing the construction of a deep learning network based on the training data set, the verification data set, and the test data set.

[0068] Specifically, a large amount of data is collected from various sources. These data can be structured or unstructured, depending on the application scenario. For power line carrier communication systems, this includes communication signal data, noise data, channel state information, etc. Sensors, web crawlers, database access and other technical means are used to collect data from different sources and formats, such as real-time monitoring, historical records, user data, etc. from communication systems. The collected data is sorted and cleaned to remove irrelevant or erroneous data points. These data are integrated into a basic data set that can be used for machine learning. The basic data set should contain enough information to facilitate subsequent analysis and modeling. Data similarity screening is performed on the basic data set to identify data points with similar features or patterns and establish similar screening identifiers. The data set is divided into multiple fixed-size windows, each containing a certain number of data points, and the data with similar screening identifiers are subjected to in-window trend adaptation analysis through a stepped window. Based on the results of the in-window trend adaptation analysis, a fitness ranking is generated for the data in each window. Data with similar screening identifiers are eliminated according to the fitness ranking, and those data points with low scores in the fitness ranking are eliminated. According to the results of data elimination, data preprocessing is completed.

[0069] Random clustering is performed on the data preprocessing results to group the data points into different clusters, so that the data points in the same cluster are similar, while the data points in different clusters are not similar. Random data clustering is an unsupervised learning technique that initializes the clustering process by randomly selecting the initial cluster center. It can automatically discover natural groupings in the data without knowing the category labels of the data in advance. It is usually necessary to select the initial cluster center, assign the data points to the nearest cluster center, and update the cluster center. This step is repeated until a certain termination condition is reached, such as the change in the cluster center is less than a certain threshold. In power line carrier communication systems, random data clustering can be used to group communication data into different clusters according to its characteristics (such as signal strength, noise level, communication error rate, etc.). Each cluster represents a different communication mode or feature combination, which helps to understand and analyze complex patterns in communication data.

[0070] According to the clustering results, the dataset is divided into three different subsets: training dataset, validation dataset and test dataset. The training dataset is used to train the model, that is, to use the model to learn the characteristics and patterns of the data; the validation dataset is used to evaluate the performance of the model during the training process and help adjust the parameters of the model; the test dataset is used to finally evaluate the generalization ability of the model, that is, the performance of the model on unseen data.

[0071] According to the characteristics and analysis requirements of the power line carrier communication system, a suitable deep learning model architecture is selected, such as convolutional neural network, recurrent neural network, generative adversarial network, etc. First, the training data set is used to train the network, constantly trying to learn the characteristics and patterns of the data and predict or classify new data points. During the training process, the parameters are constantly adjusted to minimize the prediction error. Next, the validation data set is used to evaluate the performance of the model to prevent the model from overfitting, that is, the model performs well only on the training data but performs poorly on new or unseen data. By comparing the predicted results of the validation data set with the actual results, the parameters of the network are adjusted to optimize the performance of the model. Finally, the test data set is used to evaluate the generalization ability of the model. The test data set contains unseen data points, so the performance of the model on new data can be better evaluated. A trained and optimized deep learning network is constructed using big data for signal processing and analysis of the power line carrier communication system. This network can effectively identify and process complex patterns in communication data based on characteristics such as signal strength, noise level, and communication error rate, and improve the stability of communication.

[0072] Furthermore, the present application also includes the following steps: performing data similarity screening on the basic data set and establishing a similar screening identifier; configuring a step window, performing in-window trend adaptation analysis on the data with similar screening identifiers in the basic data set through the step window, and generating a fitness ranking; eliminating data with similar screening identifiers according to the fitness ranking, and completing data preprocessing according to the data elimination results.

[0073] Specifically, similarity screening is performed on the basic data set to identify similar patterns or features in the data set, such as data points with similar signal strength, noise level, or communication error rate. Similarity screening usually involves the use of various data processing and analysis techniques, such as cluster analysis, feature matching, etc. Based on the results of similarity screening, an identifier is established for the screened similar data points to indicate whether they are similar data. For example, a basic data set containing a large amount of communication data can identify data points with similar signal strength and noise level through data similarity screening, and establish identifiers for these data points, such as "high noise level group" or "low signal strength group".

[0074] A step window is a sliding window that can be slid across a data set to analyze data trends within the window. The size and shape of the window can be configured based on the characteristics of the data and the needs of the analysis, splitting the data set into multiple fixed-size windows, each containing a certain number of data points. Trend adaptation analysis is performed on the data within each window to identify patterns and trends in the data set that change over time, including fluctuations in signal strength, periodic changes in noise levels, trends in communication error rates, etc. By analyzing how the data changes over time, long-term and short-term trends in the data can be revealed, as well as periodic or random changes. Trend adaptation analysis typically involves the use of various statistical and machine learning techniques, such as time series analysis, autoregressive models, etc.

[0075] Based on the results of the trend adaptation analysis within the window, a fitness ranking is generated for each data within the window, reflecting the changing trend and pattern of the data within the window, helping the system to identify and understand the key features in the data. For example, data points are sorted according to the changing trend of signal strength to identify important communication patterns. Data points with similar filtering identifiers are eliminated based on similar filtering identifiers. Elimination is based on a variety of reasons, such as data redundancy, data unreliability, or low value of data. For example, data points with little change in signal strength or small fluctuations in noise levels are eliminated.

[0076] The purpose of eliminating these similar data points is to optimize the data set and remove data points that may be inaccurate or insignificant, which helps to reduce the size of the data set, reduce the complexity of the data, and improve the efficiency of subsequent analysis. According to the results of data elimination, data preprocessing is completed to obtain an optimized and filtered data set that contains more critical and useful data points. Ensure that the preprocessed data set contains diverse and representative data while reducing redundancy and noise. By establishing a similar screening representation, the basic data set is analyzed for trend adaptation, and data points that do not match the trend within the window or have high similarity are eliminated, effectively preprocessing the basic data set, improving data quality, and preparing data for the training of deep learning models.

[0077] Furthermore, the present application also includes the following steps: obtaining the data source of the basic data set, performing a credible evaluation of the data based on the data source, and establishing a credible evaluation ranking; performing a weighted analysis based on the credible evaluation ranking and the fitness ranking, and eliminating data identified by similar screening based on the weighted analysis results.

[0078] Specifically, when collecting basic data, the source of each data point is usually recorded, including sensor data, network monitoring data, user behavior data, etc. Based on the source of the data, the data collection method, data consistency, data update frequency and other factors are analyzed to evaluate the credibility of the data, assess the quality and reliability of the data, and identify possible errors or deviations. If a data point comes from a data source that is considered more reliable, then its credibility will be higher. For example, sensor data is considered more reliable because it comes directly from the communication system, followed by network monitoring data, and user behavior data has lower reliability due to its subjectivity. Sort the data points according to their credibility, and put the data points with higher credibility in front. For example, sensor data points are given higher credibility scores, while user behavior data points are given lower scores.

[0079] Combine the trust evaluation ranking and fitness ranking for weighted analysis, and assign a weighted value to the data point according to the reliability of the data and its trend adaptability within the window. Weighted analysis usually involves the use of various statistical and machine learning techniques, such as weighted averaging, weighted sorting, etc., to analyze as needed to comprehensively consider the multi-faceted characteristics of the data. For example, give sensor data points higher weights because they are more reliable, while considering the importance of the trend of signal strength changes. Perform similar screening on the data set to identify data points with similar characteristics or patterns in the data set, which have similar characteristics such as signal strength, noise level or communication error rate. According to the results of weighted analysis, eliminate the data with similar screening identifiers, and give priority to eliminating those data points with low scores in weighted analysis. For example, eliminate those data points that have low scores in weighted analysis and are similar to other data points to optimize the data set. The purpose of data elimination is to optimize the data set and remove data points that may be inaccurate or irrelevant, thereby improving the efficiency and accuracy of subsequent analysis and model building. Through weighted analysis, inaccurate or irrelevant data points are identified and eliminated to improve the quality of the data set.

[0080] Furthermore, the present application also includes the following steps: performing real-time processing monitoring on the power line carrier signal noise suppression processing to establish a monitoring and evaluation data set; performing performance monitoring and evaluation using the monitoring and evaluation data set to generate compensation data; and performing optimized deep learning network compensation processing using the compensation data.

[0081] Specifically, a real-time processing monitoring system is used to continuously monitor the noise suppression processing effect of the power line carrier signal through various sensors, network monitoring equipment or software tools. Data such as signal quality, noise level, and communication error rate are collected in real time to promptly discover problems in the signal processing process, such as poor noise suppression effect and increased communication error rate. According to the results of real-time monitoring, a monitoring and evaluation data set is established, which includes processed signal data, original signal data, noise data, and related performance indicators such as signal quality and communication error rate. The monitoring and evaluation data set is used to perform performance detection and evaluation, evaluate the effect of the current noise suppression processing, and compare it with the preset performance standards, including calculating signal quality index (such as signal-to-noise ratio), evaluating noise suppression efficiency, analyzing communication error rate, etc. According to the detection results, compensation data is generated to adjust the noise suppression processing parameters, such as adjusting precoding parameters, adaptive spectrum and subcarrier mapping, etc. The purpose of the compensation data is to correct the deficiencies in the noise suppression processing, improve signal quality, and reduce the communication error rate. Next, the generated compensation data is applied to the optimized deep learning network, including adjusting the weights of the network, changing the algorithm parameters, optimizing the model structure, etc. The deep learning network is adjusted and optimized according to the compensation data, and the network is retrained to adapt to the new compensation parameters. Real-time monitoring of the effectiveness of noise suppression processing, generation of compensation data based on the monitoring results, and application of this data to adjust the deep learning network will help the system adapt to the ever-changing communication environment.

[0082] In summary, the method for suppressing pulse noise in a power line carrier communication system provided by the present application establishes digital communication with the power line carrier communication system, obtains a communication data set within a preset time period based on the digital communication; extracts communication features based on the communication data set and establishes a communication feature set; constructs a deep learning network through big data, configures an additional attention layer through the communication feature set, and optimizes the deep learning network through the additional attention layer; reads real-time communication data, configures precoding parameters, adaptive spectrum and subcarrier mapping based on the real-time communication data according to the optimized deep learning network; performs power line carrier signal noise suppression processing in the real-time communication data according to the configured precoding parameters, adaptive spectrum and subcarrier mapping. That is to say, by configuring the additional attention layer to optimize the deep learning network, reading the real-time communication data, configuring the precoding parameters, adaptive spectrum and subcarrier mapping, and performing noise suppression processing, the technical effect of improving the signal quality of the power line carrier communication system is achieved.

[0083] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0084] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0085] In the present invention, unless otherwise clearly specified and limited, when a first feature is “on” or “below” a second feature, it may be that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. Moreover, when a first feature is “above”, “above” or “above” a second feature, it may be that the first feature is directly above or obliquely above the second feature, or it may simply mean that the first feature is higher in level than the second feature. When a first feature is “below”, “below” or “below” a second feature, it may be that the first feature is directly below or obliquely below the second feature, or it may simply mean that the first feature is lower in level than the second feature.

[0086] In the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0087] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may alter, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for suppressing impulse noise in a power line carrier communication system, characterized in that: The method is applied to a noise suppression system, and the method comprises: Establishing digital communication with a power line carrier communication system, and acquiring a communication data set within a preset time period based on the digital communication; The communication data set includes data points collected at a preset collection frequency within a preset time period; each data point includes multiple features; Based on the communication data set, establishing a communication feature set; Optimizing a preset deep learning network model through the communication feature set to obtain an optimized deep learning network model; The deep learning network model includes a deep learning network and an additional attention layer embedded in the deep learning network for optimizing the deep learning network during the training process of the deep learning network model; Read the current data points and input them into the optimized deep learning network model to obtain the predicted configuration parameters; The configuration parameters include precoding parameters, adaptive spectrum and subcarrier mapping; Noise suppression processing is performed on the power line carrier signal in the current data point according to the predicted configuration parameters.

2. The method for suppressing impulse noise in a power line carrier communication system according to claim 1, characterized in that: Based on the communication data set, a communication feature set is established, specifically including: Preprocessing the data points in the communication data set to obtain a preprocessed communication data set; Based on the features in the preprocessed communication data set, obtain the corresponding covariance matrix; Performing eigenvalue decomposition on the covariance matrix to obtain a plurality of data groups, wherein each data group includes an eigenvalue and its corresponding eigenvector; Selecting eigenvectors corresponding to N eigenvalues ​​in a plurality of data groups according to a preset strategy, and projecting each data point in the preprocessed communication data set onto each of the screened eigenvectors based on the screened eigenvectors to obtain a reduced-dimensional data representation corresponding to the data point; The dimensionality-reduced data representations corresponding to all data points in the preprocessed communication data set are used to form a communication feature set.

3. The method for suppressing impulse noise in a power line carrier communication system according to claim 2, characterized in that: Selecting eigenvectors corresponding to N eigenvalues ​​in a plurality of data groups according to a preset strategy, specifically includes: For each eigenvalue and the corresponding eigenvector in each data group, the target value of the data group is obtained using formula (1); Wherein, formula (1) is: ； Q is the target value of the data set; is the characteristic value in the data set; is the eigenvalue in the data set The variance of the corresponding eigenvector; M is the sum of the eigenvalues ​​in all data sets; is the preset weight parameter; Based on the target value of each data group, the eigenvectors corresponding to the eigenvalues ​​in the first N data groups with the largest target values ​​are selected.

4. The method for suppressing impulse noise in a power line carrier communication system according to claim 3, characterized in that: Preprocessing the communication data set to obtain a preprocessed communication data set; Specifically include: Screening out data points whose time series attenuation coefficients meet a preset threshold value from the communication data set to form an initial communication data set; Screening the initial communication data set according to a preset screening strategy to obtain a preprocessed communication data set; Among them, the time series attenuation coefficient of the data point is calculated using formula (2); The formula (2) is: ； in, is a preset first decay rate; is a preset first weight coefficient; is a preset second decay rate; is a preset second weight coefficient; is a preset third weight coefficient; t is the time from the collection moment of the data point corresponding to the time series attenuation coefficient to the current moment; It is the time series attenuation coefficient corresponding to the data point when the time between the collection moment and the current moment is t.

5. The method for suppressing impulse noise in a power line carrier communication system according to claim 4, characterized in that: The screening of the initial communication data set according to a preset screening strategy to obtain a pre-processed communication data set specifically includes: Screening out data points that meet similar conditions in the initial communication data set, and marking the screened data points that meet similar conditions with screening identifiers; Using a step window, the trend adaptation analysis is performed on the data points with the filter mark in the initial communication data set, so that each data point with the filter mark gets a corresponding fitness score; Filtering the data points with the filter identifier in the initial communication data set to obtain a first data point set; The first data point set consists of data points with screening identifiers whose fitness scores are greater than a preset fitness value; Obtaining a comprehensive score corresponding to each data point in the first data point set according to the fitness score corresponding to the data point and a preset credibility score; The data points in the first data point set whose comprehensive scores are greater than a preset threshold and the preset configuration parameters corresponding to the data points one by one form a preprocessed communication data set.

6. The method for suppressing impulse noise in a power line carrier communication system according to claim 5, characterized in that: Using a step window, the trend adaptation analysis is performed on the data points with the filter mark in the initial communication data set, so that each data point with the filter mark obtains a corresponding fitness score, including: Using stepped windows, the initial communication data set is divided into multiple consecutive or overlapping time windows, each time window contains a preset number of data points; In each step window, a predetermined fitting algorithm is used to perform trend fitting on the designated features of the data points with the screening mark, and a fitting trend line corresponding to the designated features is generated. The predicted value of the designated feature of each data point with the screening mark is calculated according to the fitting trend line, and the fitness score of the data point is further calculated based on the predicted value of the designated feature. The specified characteristics include: signal strength, noise level, signal-to-noise ratio; Wherein, the predetermined fitting algorithm is a linear fitting algorithm.

7. The method for suppressing impulse noise in a power line carrier communication system according to claim 6, characterized in that: The step of calculating the fitness score of the data point based on the predicted value of the specified feature specifically includes: Based on the predicted value of the specified feature, the fitness score of the data point is calculated using formula (3); The formula (3) is: ； ； For the first i The fitness score of each data point; For the first i The true value of the signal strength in the data points; For the first i The predicted value of the signal strength in the data points; For the first i The true value of the noise level in the data points; For the first i The predicted value of the noise level in the data points; For the first i The true value of the signal-to-noise ratio in the data points; For the first i The predicted value of the signal-to-noise ratio in the data points; is the preset maximum deviation value of signal strength; is the preset maximum deviation value of the noise level; is the preset maximum deviation value of the signal-to-noise ratio; is the time position weight, which indicates the importance of the i-th data point with the screening mark in the step window; is the collection time corresponding to the i-th data point with the screening mark; is the center time point of the step window; is a preset parameter value.

8. The method for suppressing impulse noise in a power line carrier communication system according to claim 7, characterized in that: The step of obtaining a comprehensive score corresponding to each data point in the first data point set according to the fitness score corresponding to the data point and a preset credibility score specifically includes: According to the fitness score and the preset credibility score corresponding to each data point in the first data point set, the comprehensive score corresponding to the data point is obtained using formula (4); The formula (4) is: ； is the comprehensive score corresponding to the i-th data point with the screening mark; is the average of the fitness scores corresponding to all data points with the screening mark; is the standard deviation of the fitness scores corresponding to all data points with the screening mark; is the preset credibility score corresponding to the i-th data point with the screening mark; is the average of the credibility scores corresponding to all data points with the screening mark; is the standard deviation of the credibility scores corresponding to all data points with the screening mark; is the pre-set fitness weight coefficient; is a pre-set credibility weight coefficient.

9. The method for suppressing impulse noise in a power line carrier communication system according to claim 8, characterized in that: in, The deep learning network model is obtained by embedding an additional attention layer between the feature extraction layer and the fully connected layer in the convolutional neural network; The attention addition layer adopts any one of the self-attention mechanism, the soft attention mechanism and the hard attention mechanism.

Images