A method for detecting access security of carrier communication system in distribution network

Abstract

The present application relates to a method for detecting the access security of a distribution network carrier communication system. The method includes: collecting environmental parameter data through real-time monitoring, including power supply voltage value, temperature value and humidity value, and introducing data processing and signal analysis algorithms based on artificial intelligence and deep learning at the back end to analyze the original signal and the environmental parameter data, so as to learn and capture the waveform semantic features in the original signal and the multimodal time series correlation features of the environmental parameter data, and then use the cross-modal joint semantics between the two features to correct the original signal waveform semantics based on environmental fluctuations, so as to obtain a more robust device fingerprint. In this way, the impact of environmental interference on device fingerprint recognition and extraction can be reduced in a more intelligent way, thereby enhancing the stability and accuracy of fingerprint extraction and matching, so as to ensure the access security of the distribution network carrier communication system and ensure the stability and reliability of the power system.

Description

Technical Field

[0001] The present application relates to the technical field of power system network security, and in particular to a method for detecting access security of a distribution network carrier communication system. Background Art

[0002] With the development of smart grid, power line communication (PLC) technology has been widely used because it can use the existing power line infrastructure for data transmission. However, PLC systems face many security challenges, especially the risk of illegal access and eavesdropping. In order to ensure the safe operation of the distribution network, it is necessary to develop an effective access security detection method to prevent unauthorized devices from accessing the network and ensure the security and reliability of data transmission.

[0003] Chinese patent CN113612751A discloses a method for detecting access security of a power line carrier communication system in a distribution network, which can collect communication signals from each subcarrier machine through a main carrier machine, and extract device fingerprints from them to identify and authenticate power grid terminal devices. During the authentication process, the distance between the extracted device fingerprint and the fingerprint in the preset fingerprint library is calculated. If it exceeds the set threshold, it is considered that there is an illegal access device. In addition, the access security detection method also locates potential eavesdropping devices by calculating the fingerprint offset vector of all network devices and evaluating the similarity of fingerprint offset vectors between adjacent devices.

[0004] Although the above-mentioned power distribution network power line carrier communication system access security detection method provides an effective way to detect illegal access devices, there are still some challenges and potential problems in practical applications: for example, the extraction and matching of device fingerprints requires a high degree of stability and reliability. Environmental factors such as changes in temperature, humidity and power supply voltage, as well as electromagnetic interference and signal attenuation, may cause device fingerprints to change, thereby affecting the accuracy and stability of fingerprint matching. In addition, the power grid environment itself is relatively complex, with a variety of interference sources, such as electromagnetic interference and signal attenuation. These interferences will directly affect the quality of the signal, further affecting the accurate extraction and matching of device fingerprints.

[0005] Therefore, an optimized distribution network carrier communication system access security detection solution is desired. Summary of the invention

[0006] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0007] In a first aspect, the present application provides a method for detecting access security of a distribution network carrier communication system, the method comprising:

[0008] The main carrier machine collects the original signal sent by the subcarrier machine, and extracts the device fingerprint based on the original signal;

[0009] Calculate the distance between the device fingerprint and the reference device fingerprint extracted from the device fingerprint library to determine the identity authentication result of the subcarrier machine, where the identity authentication result is used to indicate whether the identity authentication of the subcarrier machine is passed;

[0010] In response to the identity authentication result of the subcarrier machine being that the identity authentication of the subcarrier machine fails, fingerprint collection is performed on all carrier devices in the power line carrier communication system of the distribution network, and fingerprint offset vectors between device fingerprints of all carrier devices and the device fingerprint of the subcarrier machine are calculated to obtain a sequence of fingerprint offset vectors;

[0011] Calculating the similarity between every two adjacent fingerprint offset vectors in the sequence of fingerprint offset vectors to obtain a sequence of fingerprint offset similarities;

[0012] Comparing each fingerprint offset similarity in the sequence of fingerprint offset similarities with a predetermined threshold range, respectively, and in response to the fingerprint offset similarity being within the predetermined threshold range, determining that there is an eavesdropping device between two carrier devices corresponding to the fingerprint offset similarities within the predetermined threshold range;

[0013] The main carrier machine collects the original signal sent by the subcarrier machine, and extracts the device fingerprint based on the original signal, including:

[0014] A time queue for collecting environmental parameters collected by the sensor group, wherein the environmental parameters include power supply voltage value, temperature value and humidity value;

[0015] Performing multi-modal temporal correlation pattern feature extraction on the time queue of the environmental parameters to obtain a multi-modal temporal correlation feature vector of the environmental parameters;

[0016] Extracting signal waveform semantic features from the original signal to obtain an original signal waveform semantic feature graph;

[0017] The original signal waveform semantic feature map and the environmental parameter multimodal temporal correlation feature vector are input into a gated cross-modal joint encoder based on an imitation converter structure to obtain an environmental fluctuation correction signal waveform semantic feature map as the device fingerprint.

[0018] Optionally, multimodal temporal association pattern feature extraction is performed on the time queue of the environmental parameters to obtain a multimodal temporal association feature vector of the environmental parameters, including: arranging the time queue of the environmental parameters into an environmental parameter temporal multimodal joint matrix according to the time dimension and the parameter sample dimension; inputting the environmental parameter temporal multimodal joint matrix into an environmental parameter multimodal temporal association pattern feature extractor based on a first deep neural network model to obtain the environmental parameter multimodal temporal association feature vector.

[0019] Optionally, the first deep neural network model is a first atrous convolutional neural network model.

[0020] Optionally, performing signal waveform semantic feature extraction on the original signal to obtain an original signal waveform semantic feature map includes: inputting the original signal into a signal waveform semantic feature extractor based on a second deep neural network model to obtain the original signal waveform semantic feature map.

[0021] Optionally, the second deep neural network model is a second atrous convolutional neural network model.

[0022] Optionally, the original signal waveform semantic feature map and the multimodal temporal correlation feature vector of the environmental parameters are input into a gated cross-modal joint encoder based on an analog converter structure to obtain an environmental fluctuation correction signal waveform semantic feature map as the device fingerprint, including: performing layer normalization and convolution coding on the original signal waveform semantic feature map to obtain an original signal waveform semantic space deepening feature map; copying the original signal waveform semantic space deepening feature map to obtain a backup original signal waveform semantic space deepening feature map; and performing layer normalization and convolution coding on the original signal waveform semantic feature map to obtain a backup original signal waveform semantic space deepening feature map. Perform attention-gated autocorrelation encoding to obtain the weight matrix of the semantically gated local autocorrelation library of the original signal waveform; use the environmental parameter multimodal temporal association feature vector as the value feature vector, calculate the product between the original signal waveform semantically gated local autocorrelation library weight matrix and the environmental parameter multimodal temporal association feature vector to obtain the environmental parameter multimodal temporal cross-modal interaction fusion conditional weight vector; use the environmental parameter multimodal temporal cross-modal interaction fusion conditional weight vector as the weight vector, perform channel-by-channel weighting on the original signal waveform semantic feature map to obtain the environmental fluctuation correction signal waveform semantic feature map.

[0023] Optionally, the original signal waveform semantic feature map is subjected to layer normalization and convolution coding processing to obtain the original signal waveform semantic space deepening feature map, including: performing layer normalization processing on the original signal waveform semantic feature map to obtain a layer normalized original signal waveform semantic feature map; performing point convolution coding on the layer normalized original signal waveform semantic feature map and then performing spatial convolution coding to obtain the original signal waveform semantic space deepening feature map.

[0024] Optionally, the original signal waveform semantic space deepening feature map and the backup original signal waveform semantic space deepening feature map are subjected to attention gating-based autocorrelation encoding to obtain the original signal waveform semantic gated local autocorrelation library weight matrix, including: respectively reshaping the feature shapes of the backup original signal waveform semantic space deepening feature map and the original signal waveform semantic space deepening feature map to obtain the backup original signal waveform semantic space deepening feature matrix as the imitation query feature matrix and the original signal waveform semantic space deepening feature matrix as the imitation key feature matrix; calculating the product between the imitation query feature matrix and the transposed matrix of the imitation key feature matrix to obtain the original signal waveform semantic local autocorrelation library matrix; inputting the original signal waveform semantic local autocorrelation library matrix into the attention gating unit based on the GELU function to obtain the original signal waveform semantic gated local autocorrelation library weight matrix.

[0025] The above technical solution is adopted to collect environmental parameter data through real-time monitoring, including power supply voltage value, temperature value and humidity value, and introduce data processing and signal analysis algorithms based on artificial intelligence and deep learning at the back end to analyze the original signal and the environmental parameter data, so as to learn and capture the waveform semantic features in the original signal and the multimodal time series correlation features of the environmental parameter data, and then use the cross-modal joint semantics between the two features to correct the original signal waveform semantics based on environmental fluctuations, so as to obtain a more robust device fingerprint. In this way, the impact of environmental interference on device fingerprint recognition and extraction can be reduced in a more intelligent way, thereby enhancing the stability and accuracy of fingerprint extraction and matching, so as to ensure the access security of the distribution network carrier communication system and ensure the stability and reliability of the power system.

[0026] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings:

[0028] Figure 1 The present invention is a flowchart of a method for detecting access security of a power distribution network carrier communication system according to an exemplary embodiment.

[0029] Figure 2 The invention is a block diagram of a distribution network carrier communication system access security detection system according to an exemplary embodiment.

[0030] Figure 3 It is a block diagram of an electronic device according to an exemplary embodiment.

[0031] Figure 4 It is an application scenario diagram of a method for detecting access security of a distribution network carrier communication system according to an exemplary embodiment. DETAILED DESCRIPTION

[0032] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0033] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0034] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0035] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0036] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0037] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0038] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0039] Figure 1 is a flow chart of a method for detecting access security of a distribution network carrier communication system according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0040] Step S101: collecting the original signal sent by the subcarrier machine through the main carrier machine, and extracting the device fingerprint based on the original signal;

[0041] Step S102: Calculate the distance between the device fingerprint and the reference device fingerprint extracted from the device fingerprint library to determine the identity authentication result of the subcarrier machine, where the identity authentication result is used to indicate whether the identity authentication of the subcarrier machine is passed;

[0042] Step S103: in response to the result of the identity authentication of the subcarrier machine being that the identity authentication of the subcarrier machine fails, fingerprint collection is performed on all carrier devices in the power line carrier communication system of the distribution network, and a fingerprint offset vector between the device fingerprints of all carrier devices and the device fingerprint of the subcarrier machine is calculated to obtain a sequence of fingerprint offset vectors;

[0043] Step S104, calculating the similarity between every two adjacent fingerprint offset vectors in the sequence of fingerprint offset vectors to obtain a sequence of fingerprint offset similarities;

[0044] Step S105: Compare each fingerprint offset similarity in the sequence of fingerprint offset similarities with a predetermined threshold range, and in response to the fingerprint offset similarity being within the predetermined threshold range, determine that there is an eavesdropping device between two carrier devices corresponding to the fingerprint offset similarities within the predetermined threshold range;

[0045] Wherein, step S101, collecting the original signal sent by the subcarrier machine through the main carrier machine, and extracting the device fingerprint based on the original signal, includes:

[0046] Step S1011, collecting a time queue of environmental parameters collected by the sensor group, wherein the environmental parameters include power supply voltage value, temperature value and humidity value;

[0047] Step S1012, performing multi-modal temporal correlation pattern feature extraction on the time queue of the environmental parameters to obtain a multi-modal temporal correlation feature vector of the environmental parameters;

[0048] Step S1013, extracting the signal waveform semantic features of the original signal to obtain a semantic feature graph of the original signal waveform;

[0049] Step S1014: input the original signal waveform semantic feature map and the environmental parameter multimodal temporal correlation feature vector into a gated cross-modal joint encoder based on an analog converter structure to obtain an environmental fluctuation correction signal waveform semantic feature map as the device fingerprint.

[0050] In particular, in the above-mentioned distribution network carrier communication system access security detection method, in order to be able to perform device fingerprint extraction and matching more stably and reliably, so as to maintain good fingerprint extraction and security detection performance even in a complex power grid environment, thereby improving the security of the entire PLC system, it is necessary to consider the impact of environmental changes (such as temperature, humidity, power supply fluctuations, etc.) during the device fingerprint extraction and matching process.

[0051] Based on this, the technical concept of this application is to collect environmental parameter data, including power supply voltage value, temperature value and humidity value, through real-time monitoring in the process of extracting device fingerprints, and introduce data processing and signal analysis algorithms based on artificial intelligence and deep learning in the back end to analyze the original signal and the environmental parameter data, so as to learn and capture the waveform semantic features in the original signal and the multimodal time series correlation features of the environmental parameter data, and then use the cross-modal joint semantics between the two features to correct the original signal waveform semantics based on environmental fluctuations, so as to obtain a more robust device fingerprint. In this way, by using artificial intelligence and deep learning technology to extract device fingerprints based on the original signal, the impact of environmental interference on device fingerprint recognition and extraction can be reduced in a more intelligent way, thereby enhancing the stability and accuracy of fingerprint extraction and matching, and at the same time, it can improve the generalization ability of the distribution network carrier communication system access security detection method, so as to ensure the access security of the distribution network carrier communication system and ensure the stability and reliability of the power system.

[0052] Specifically, in the technical solution of the present application, in the process of collecting the original signal sent by the subcarrier machine through the main carrier machine and extracting the device fingerprint based on the original signal, first, the time queue of the environmental parameters collected by the sensor group is collected, wherein the environmental parameters include power supply voltage value, temperature value and humidity value. Among them, the stability of the power supply voltage is crucial to the normal operation of the electronic device. In the fingerprint extraction device, the ripple and noise of the power supply voltage may affect the signal processing capability of the device. For example, the ripple voltage may cause the instability of the signal, thereby affecting the quality of the fingerprint image. The generation of power supply ripple is related to the design of the power supply, the selection and matching of components such as inductance and capacitance. When designing the power supply, by optimizing the inductance value, switching frequency and output capacitance, the ripple can be reduced, the stability of the power supply can be improved, and the impact on the quality of fingerprint extraction can be reduced; temperature changes may affect the performance of the mechanical parts and electronic components of the fingerprint extraction device. For example, too high a temperature may cause the device to overheat, affecting the stability and life of the electronic components, and thus affecting the acquisition and processing of the fingerprint image. On the other hand, changes in temperature may also affect the fingerprint itself, because components such as sweat and oil in fingerprints will change with temperature, which may affect the clarity and recognizability of fingerprints; humidity also has a significant impact on fingerprint extraction. A high humidity environment may cause blurred fingerprint images because the moisture in the fingerprint may diffuse during the extraction process, resulting in a decrease in image quality. In addition, humidity may also affect the electronic components of the fingerprint extraction device, such as capacitors and resistors, thereby affecting the response speed and accuracy of the device.

[0053] Next, considering that environmental parameters include power supply voltage value, temperature value and humidity value, these different types of environmental parameters carry different information, which will affect fingerprint extraction and subsequent matching. In addition, considering that these environmental parameter data not only have dynamic characteristics and change information of time series in the time dimension, but also have mutual influence and synergy between these different types of environmental parameters. For example, the stability of power supply voltage may be affected by temperature. In a high temperature environment, the power supply component may overheat, resulting in voltage fluctuations, thereby affecting the stability of the device and the quality of fingerprint extraction; the increase in temperature may accelerate the aging of the power supply component, thereby affecting the stability of the power supply voltage. In a high humidity environment, the power supply component may corrode or short-circuit due to moisture, which will affect the stability of the power supply voltage and thus affect the performance of the fingerprint extraction device. The fluctuation of the power supply voltage may cause the electronic components of the fingerprint extraction device to be more prone to failure in a high humidity environment. Temperature and humidity jointly affect the physical and chemical properties of fingerprints. For example, high humidity may increase the moisture in the fingerprint, and temperature changes may affect the volatilization rate of fingerprint residues. In some cases, the combination of temperature and humidity may accelerate the degradation of fingerprint residues, thereby affecting the durability and extractability of fingerprints. Therefore, these environmental parameters may influence each other and jointly determine the difficulty and quality of fingerprint extraction.

[0054] Based on this, in the technical solution of the present application, the time queue of the environmental parameters is further arranged into an environmental parameter time series multimodal joint matrix according to the time dimension and the parameter sample dimension, and the environmental parameter time series multimodal joint matrix is ​​input into an environmental parameter multimodal time series association pattern feature extractor based on the first void convolutional neural network model for feature mining, so as to extract the multimodal time series association feature information between the voltage value, temperature value and humidity value in the environmental parameters, thereby obtaining the environmental parameter multimodal time series association feature vector.

[0055] In one embodiment of the present application, multimodal temporal association pattern feature extraction is performed on the time queue of the environmental parameters to obtain a multimodal temporal association feature vector of the environmental parameters, including: arranging the time queue of the environmental parameters into an environmental parameter temporal multimodal joint matrix according to the time dimension and the parameter sample dimension; inputting the environmental parameter temporal multimodal joint matrix into an environmental parameter multimodal temporal association pattern feature extractor based on a first deep neural network model to obtain the environmental parameter multimodal temporal association feature vector.

[0056] Among them, the first deep neural network model is a first void convolutional neural network model.

[0057] Then, the original signal is input into a signal waveform semantic feature extractor based on a second atrous convolutional neural network model for feature mining to extract the waveform semantic feature information of the original signal, thereby obtaining a semantic feature map of the original signal waveform. In particular, since the atrous convolutional neural network model can expand the receptive field without increasing the network depth or significantly increasing the number of parameters, it can capture a wider range of signal waveform semantic feature information. This method is particularly suitable for signal processing in power line carrier communication systems, because power line signals usually have the characteristics of periodicity and frequency components, and the atrous convolutional neural network model can effectively capture these characteristics and extract semantically meaningful feature maps from the signal. The purpose of this process is to enhance the expressiveness of device fingerprints, making them more recognizable, while reducing the impact of environmental changes and other interference factors.

[0058] In one embodiment of the present application, signal waveform semantic feature extraction is performed on the original signal to obtain an original signal waveform semantic feature map, including: inputting the original signal into a signal waveform semantic feature extractor based on a second deep neural network model to obtain the original signal waveform semantic feature map.

[0059] Among them, the second deep neural network model is a second void convolutional neural network model.

[0060] It should be understood that the original signal waveform semantic feature map and the environmental parameter multimodal temporal association feature vector respectively contain the waveform semantic features of the original signal and the multimodal temporal association fusion features of the environmental parameters, and the process of extracting the device fingerprint from the original signal will be affected by environmental factors (such as temperature, humidity, power supply fluctuations). Therefore, in order to be able to use the multimodal parameter temporal collaborative fluctuation characteristics of the environment to correct and optimize the expression of the original signal waveform semantics, in the technical solution of the present application, the original signal waveform semantic feature map and the environmental parameter multimodal temporal association feature vector are further input into the gated cross-modal joint encoder based on the simulated transformer structure to obtain the environmental fluctuation correction signal waveform semantic feature map as the device fingerprint. The gated cross-modal joint encoder based on the simulated transformer structure can achieve effective fusion between different modal data features (i.e., signal waveform semantic features and environmental parameter multimodal temporal association features) by imitating the self-attention mechanism of the Transformer structure, and adjust the importance of different features in the joint encoding process through the gating mechanism, thereby obtaining a more robust device fingerprint.

[0061] Specifically, the gated cross-modal joint encoder based on the simulated transformer structure first performs point convolution encoding and spatial convolution encoding on the semantic feature map of the original signal waveform to extract the deep spatial features in the original signal waveform to generate the original signal waveform semantic space deepening feature map, thereby enhancing the spatial expression ability of the feature map. Subsequently, these feature maps are copied and reshaped to obtain the query and key feature matrices. Next, by calculating the product of the query feature matrix and the transposed key feature matrix, the original signal waveform semantic local autocorrelation library matrix is ​​constructed, and the modal internal correlation of the semantic features of the original signal waveform is realized. Furthermore, the attention gating unit based on the GELU function is used to dynamically adjust the autocorrelation weight to achieve adaptive selection and enhancement of features. This attention gating mechanism not only enhances the expressive power of the features, but also provides adaptive weight adjustment for subsequent cross-modal fusion between the multimodal temporal correlation features of the environmental parameters. It is worth mentioning that the attention gating unit dynamically adjusts the autocorrelation weight through a mechanism based on the GELU function, ensuring that the selection of features is not only based on static rules, but is dynamically adjusted according to the actual performance of the data. This makes the model more flexible and able to cope with changes in the original signal waveform semantics and environmental data in different scenarios. Then, by multiplying the weight matrix of the original signal waveform semantic gated local autocorrelation library with the environmental parameter multimodal time series associated feature vector, the environmental parameter multimodal time series cross-modal interactive fusion condition weight vector is obtained, which realizes the conditional fusion between the two different modalities and enhances the feature cross-modal association between the original signal waveform semantics and the environmental parameter multimodal time series associated semantics. This means that when the model generates the final device fingerprint, it not only considers the characteristics of the signal itself, but also fully considers the influence of environmental conditions on the original signal, thereby improving the stability and accuracy of device fingerprint extraction. Finally, the original signal waveform semantic feature map is weighted channel by channel using the environmental parameter multimodal time series cross-modal interactive fusion condition weight vector to generate the environmental fluctuation correction signal waveform semantic feature map, thereby integrating the feature information of different modalities and improving the model's understanding and processing capabilities of multimodal data. The environmental fluctuation correction signal waveform semantic feature map generated by the above steps can better resist the influence of environmental fluctuations on signal characteristics, thereby maintaining a high device fingerprint recognition accuracy under various conditions. In this way, artificial intelligence and deep learning technologies can be used to reduce the impact of environmental interference on device fingerprint recognition and extraction in a more intelligent way during the process of extracting device fingerprints based on original signals, thereby enhancing the stability and accuracy of fingerprint extraction and matching. At the same time, the generalization ability of the distribution network carrier communication system access security detection method can be improved, thereby ensuring the access security of the distribution network carrier communication system and ensuring the stability and reliability of the power system.

[0062] In one embodiment of the present application, the original signal waveform semantic feature map and the environmental parameter multimodal temporal correlation feature vector are input into a gated cross-modal joint encoder based on an imitation converter structure to obtain an environmental fluctuation correction signal waveform semantic feature map as the device fingerprint, including: performing layer normalization and convolution coding on the original signal waveform semantic feature map to obtain an original signal waveform semantic space deepening feature map; copying the original signal waveform semantic space deepening feature map to obtain a backup original signal waveform semantic space deepening feature map; and performing layer normalization and convolution coding on the original signal waveform semantic feature map to obtain a backup original signal waveform semantic space deepening feature map. The feature map is subjected to attention-gated autocorrelation encoding to obtain the weight matrix of the semantically gated local autocorrelation library of the original signal waveform; the environmental parameter multimodal temporal association feature vector is used as the value feature vector, and the product between the semantically gated local autocorrelation library weight matrix of the original signal waveform and the environmental parameter multimodal temporal association feature vector is calculated to obtain the environmental parameter multimodal temporal cross-modal interaction fusion conditional weight vector; the environmental parameter multimodal temporal cross-modal interaction fusion conditional weight vector is used as the weight vector, and the original signal waveform semantic feature map is weighted channel by channel to obtain the environmental fluctuation correction signal waveform semantic feature map.

[0063] Furthermore, in one embodiment of the present application, the original signal waveform semantic feature map is layer-normalized and convolutionally encoded to obtain an original signal waveform semantic space deepening feature map, including: performing layer-normalization processing on the original signal waveform semantic feature map to obtain a layer-normalized original signal waveform semantic feature map; performing point convolution encoding on the layer-normalized original signal waveform semantic feature map and then performing spatial convolution encoding on the layer-normalized original signal waveform semantic feature map to obtain the original signal waveform semantic space deepening feature map.

[0064] Furthermore, in one embodiment of the present application, the original signal waveform semantic space deepening feature map and the backup original signal waveform semantic space deepening feature map are subjected to attention gating-based autocorrelation encoding to obtain the original signal waveform semantic gated local autocorrelation library weight matrix, including: respectively reshaping the feature shape of the backup original signal waveform semantic space deepening feature map and the original signal waveform semantic space deepening feature map to obtain the backup original signal waveform semantic space deepening feature matrix as the imitation query feature matrix and the original signal waveform semantic space deepening feature matrix as the imitation key feature matrix; calculating the product between the imitation query feature matrix and the transposed matrix of the imitation key feature matrix to obtain the original signal waveform semantic local autocorrelation library matrix; inputting the original signal waveform semantic local autocorrelation library matrix into the attention gating unit based on the GELU function to obtain the original signal waveform semantic gated local autocorrelation library weight matrix.

[0065] Specifically, the original signal waveform semantic feature map and the environmental parameter multimodal temporal correlation feature vector are input into the gated cross-modal joint encoder based on the imitation converter structure and processed with the following cross-modal joint formula to obtain the environmental fluctuation correction signal waveform semantic feature map;

[0066] Among them, the cross-modal joint formula is:

[0067] ,

[0068] in, is the semantic feature map of the original signal waveform, is the layer normalization operation, is point convolutional coding, The convolution kernel is The convolution operation, Deepen the feature map of the original signal waveform semantic space, For copy operation, To back up the original signal waveform semantic space to deepen the feature map, For the reshape operation, is the transpose of the feature matrix, is matrix multiplication, is the GELU function, is the weight matrix of the semantically gated local autocorrelation library of the original signal waveform, is the multi-modal temporal correlation feature vector of the environmental parameters, is the weight vector of the multi-modal temporal cross-modal interaction fusion condition of the environmental parameters, It is the semantic feature map of the environmental fluctuation correction signal waveform.

[0069] In a preferred example, considering that the original signal waveform semantic feature map and the environmental parameter multimodal temporal association feature vector respectively represent the waveform image semantic features of the original signal and the environmental parameter temporal association features, when the gated cross-modal joint encoding based on the converter-like structure is performed, the differences in feature modalities and feature manifold structures will lead to a diversified set expression distribution of the semantic features of the environmental fluctuation correction signal waveform semantic feature map. Therefore, it is expected to improve the balance between the accuracy and completeness of the regression mapping of the environmental fluctuation correction signal waveform semantic feature map to the common semantic space during fingerprint recognition, thereby improving the accuracy of device fingerprint comparison and recognition.

[0070] Based on this, in the preferred example, the waveform semantic feature map of the environmental fluctuation correction signal is subjected to feature distribution optimization modulation, including:

[0071] Determine the maximum eigenvalue and the minimum eigenvalue of the environment fluctuation correction signal waveform semantics of the environment fluctuation correction signal waveform semantics feature graph, and calculate the environment fluctuation correction signal waveform semantics mean and the environment fluctuation correction signal waveform semantics standard deviation of the feature set of the environment fluctuation correction signal waveform semantics feature graph;

[0072] Calculating the quotient of the semantic mean of the environmental fluctuation correction signal waveform and the semantic standard deviation of the environmental fluctuation correction signal waveform to obtain a semantic statistical normalization value of the environmental fluctuation correction signal waveform;

[0073] Calculate the inverse of each eigenvalue of the environmental fluctuation correction signal waveform semantic feature map, multiply it by the difference between the maximum eigenvalue of the environmental fluctuation correction signal waveform semantics and the minimum eigenvalue of the environmental fluctuation correction signal waveform semantics, and then subtract it from the environmental fluctuation correction signal waveform semantic statistical normalization value to obtain an environmental fluctuation correction signal waveform semantic distribution approximation map;

[0074] Calculate an exponential function with a natural constant as the base, with each eigenvalue of the semantic distribution approximation graph of the environmental fluctuation correction signal waveform as an exponent, and perform point addition with the semantic statistical normalization value of the environmental fluctuation correction signal waveform to obtain a semantic probability approximation graph of the environmental fluctuation correction signal waveform;

[0075] The base-two logarithm of the absolute value of each eigenvalue of the environmental fluctuation correction signal waveform semantic probability approximation graph is calculated to obtain an optimized environmental fluctuation correction signal waveform semantic feature graph.

[0076] The optimized representation of the semantic feature map of the environmental fluctuation correction signal waveform, denoted as F, is:

[0077] ,

[0078] are the feature sets of the semantic feature graph of the environmental fluctuation correction signal waveform The mean, standard deviation, minimum and maximum values ​​of is the logarithm to base 2, is the reciprocal of each eigenvalue of the semantic feature graph of the environmental fluctuation correction signal waveform, It is an exponential function that calculates the exponential of each eigenvalue of the characteristic graph with a natural constant as the base. is the optimized semantic feature map of the environmental fluctuation correction signal waveform, It is added by location point. It is the point multiplication by position. It is subtracted by position point.

[0079] That is, by performing random statistical normalization on the diversified feature set of the environmental fluctuation correction signal waveform semantic feature map, the confidence space constructed based on the overall eigenvalue of the environmental fluctuation correction signal waveform semantic feature map is approximated by a standardized continuous probability density distribution relative to the response hypothesis test of each eigenvalue of the environmental fluctuation correction signal waveform semantic feature map, thereby establishing the target reachability of the discrete feature distribution of the environmental fluctuation correction signal waveform semantic feature map to a unified regression target, thereby improving the balanced executableness between the accuracy and completeness of the regression mapping when the environmental fluctuation correction signal waveform semantic feature map is mapped to a common semantic space for fingerprint recognition, thereby improving the accuracy of device fingerprint comparison and recognition.

[0080] In summary, the above scheme is adopted to collect environmental parameter data through real-time monitoring, including power supply voltage value, temperature value and humidity value, and introduce data processing and signal analysis algorithms based on artificial intelligence and deep learning at the back end to analyze the original signal and the environmental parameter data, so as to learn and capture the waveform semantic features in the original signal and the multimodal time series correlation features of the environmental parameter data, and then use the cross-modal joint semantics between the two features to correct the original signal waveform semantics based on environmental fluctuations, so as to obtain a more robust device fingerprint. In this way, by using artificial intelligence and deep learning technology to extract device fingerprints based on the original signal, the impact of environmental interference on device fingerprint recognition and extraction is reduced in a more intelligent way, thereby enhancing the stability and accuracy of fingerprint extraction and matching, and at the same time, it can improve the generalization ability of the distribution network carrier communication system access security detection method, so as to ensure the access security of the distribution network carrier communication system and ensure the stability and reliability of the power system.

[0081] Figure 2 FIG. 1 is a block diagram of a distribution network carrier communication system access security detection system according to an exemplary embodiment. Figure 2 As shown, the system 200 includes:

[0082] The original signal acquisition module 201 is used to acquire the original signal sent by the subcarrier machine through the main carrier machine, and extract the device fingerprint based on the original signal;

[0083] An identity authentication result determination module 202, configured to calculate the distance between the device fingerprint and a reference device fingerprint extracted from a device fingerprint library to determine an identity authentication result of the subcarrier machine, wherein the identity authentication result is used to indicate whether the identity authentication of the subcarrier machine is passed;

[0084] The fingerprint collection module 203 is used for collecting fingerprints of all carrier devices in the power line carrier communication system of the power distribution network in response to the identity authentication result of the subcarrier machine being that the identity authentication of the subcarrier machine fails, and calculating the fingerprint offset vectors between the device fingerprints of all carrier devices and the device fingerprint of the subcarrier machine to obtain a sequence of fingerprint offset vectors;

[0085] A similarity calculation module 204, configured to calculate the similarity between every two adjacent fingerprint offset vectors in the sequence of fingerprint offset vectors to obtain a sequence of fingerprint offset similarities;

[0086] A module 205 for determining whether there is an eavesdropping device between carrier devices is used to compare each fingerprint offset similarity in the sequence of fingerprint offset similarities with a predetermined threshold range, and in response to the fingerprint offset similarity being within the predetermined threshold range, determine that there is an eavesdropping device between two carrier devices corresponding to the fingerprint offset similarities within the predetermined threshold range;

[0087] Wherein, the original signal acquisition module 201 includes:

[0088] The environmental parameter acquisition unit 2011 is used to acquire a time queue of environmental parameters acquired by the sensor group, wherein the environmental parameters include power supply voltage value, temperature value and humidity value;

[0089] A multi-modal temporal correlation pattern feature extraction unit 2012, configured to extract multi-modal temporal correlation pattern features from the time queue of the environmental parameters to obtain a multi-modal temporal correlation feature vector of the environmental parameters;

[0090] A signal waveform semantic feature extraction unit 2013, used for extracting signal waveform semantic features from the original signal to obtain an original signal waveform semantic feature graph;

[0091] The gated cross-modal joint encoding unit 2014 is used to input the original signal waveform semantic feature map and the environmental parameter multi-modal temporal correlation feature vector into a gated cross-modal joint encoder based on an analog converter structure to obtain the environmental fluctuation correction signal waveform semantic feature map as the device fingerprint.

[0092] Reference below Figure 3 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0093] like Figure 3 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0094] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0095] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

[0096] It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0097] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0098] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0099] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0100] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0101] The modules involved in the embodiments described in this application can be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the test parameter acquisition module can also be described as a "module for acquiring device test parameters corresponding to the target device".

[0102] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0103] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0104] Figure 4 FIG. 1 is an application scenario diagram of a method for detecting access security of a distribution network carrier communication system according to an exemplary embodiment. Figure 4 As shown, in this application scenario, first, a time queue of environmental parameters collected by the sensor group is collected, wherein the environmental parameters include a power supply voltage value (for example, Figure 4 C1 as indicated in the figure), temperature values ​​(e.g. Figure 4 C2) and humidity values ​​(e.g. Figure 4 Then, the obtained power supply voltage value, temperature value and humidity value are input to a server (for example, Figure 4 In S) shown in , the server is capable of processing the power supply voltage value, the temperature value and the humidity value based on the distribution network carrier communication system access security detection algorithm to determine that there is an eavesdropping device between two carrier devices corresponding to the fingerprint offset similarity within the predetermined threshold range.

[0105] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

[0106] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the application. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0107] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms of implementation. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims

1. A method for detecting access security of a distribution network carrier communication system, characterized in that: include: The main carrier machine collects the original signal sent by the subcarrier machine, and extracts the device fingerprint based on the original signal; Calculate the distance between the device fingerprint and the reference device fingerprint extracted from the device fingerprint library to determine the identity authentication result of the subcarrier machine, where the identity authentication result is used to indicate whether the identity authentication of the subcarrier machine is passed; In response to the identity authentication result of the subcarrier machine being that the identity authentication of the subcarrier machine fails, fingerprint collection is performed on all carrier devices in the power line carrier communication system of the distribution network, and fingerprint offset vectors between device fingerprints of all carrier devices and the device fingerprint of the subcarrier machine are calculated to obtain a sequence of fingerprint offset vectors; Calculating the similarity between every two adjacent fingerprint offset vectors in the sequence of fingerprint offset vectors to obtain a sequence of fingerprint offset similarities; Comparing each fingerprint offset similarity in the sequence of fingerprint offset similarities with a predetermined threshold range, respectively, and in response to the fingerprint offset similarity being within the predetermined threshold range, determining that there is an eavesdropping device between two carrier devices corresponding to the fingerprint offset similarities within the predetermined threshold range; The main carrier machine collects the original signal sent by the subcarrier machine, and extracts the device fingerprint based on the original signal, including: A time queue for collecting environmental parameters collected by the sensor group, wherein the environmental parameters include power supply voltage value, temperature value and humidity value; Performing multi-modal temporal correlation pattern feature extraction on the time queue of the environmental parameters to obtain a multi-modal temporal correlation feature vector of the environmental parameters; Extracting signal waveform semantic features from the original signal to obtain an original signal waveform semantic feature graph; The original signal waveform semantic feature map and the environmental parameter multimodal temporal correlation feature vector are input into a gated cross-modal joint encoder based on an imitation converter structure to obtain an environmental fluctuation correction signal waveform semantic feature map as the device fingerprint.

2. The method for detecting access security of a distribution network carrier communication system according to claim 1, characterized in that: Performing multi-modal temporal correlation pattern feature extraction on the time queue of the environmental parameters to obtain a multi-modal temporal correlation feature vector of the environmental parameters includes: Arranging the time queue of the environmental parameters into an environmental parameter time series multimodal joint matrix according to the time dimension and the parameter sample dimension; The environmental parameter time series multimodal joint matrix is ​​input into the environmental parameter multimodal time series association pattern feature extractor based on the first deep neural network model to obtain the environmental parameter multimodal time series association feature vector.

3. The method for detecting access security of a distribution network carrier communication system according to claim 2, characterized in that: The first deep neural network model is a first hole convolutional neural network model.

4. The method for detecting access security of a distribution network carrier communication system according to claim 3, characterized in that: Performing signal waveform semantic feature extraction on the original signal to obtain an original signal waveform semantic feature map, comprising: inputting the original signal into a signal waveform semantic feature extractor based on a second deep neural network model to obtain the original signal waveform semantic feature map.

5. The method for detecting access security of a distribution network carrier communication system according to claim 4, characterized in that: The second deep neural network model is a second atrous convolutional neural network model.

6. The method for detecting access security of a distribution network carrier communication system according to claim 5, characterized in that: Inputting the original signal waveform semantic feature map and the multi-modal temporal correlation feature vector of the environmental parameters into a gated cross-modal joint encoder based on an imitation converter structure to obtain an environmental fluctuation correction signal waveform semantic feature map as the device fingerprint, including: Performing layer normalization and convolution coding processing on the original signal waveform semantic feature map to obtain the original signal waveform semantic space deepening feature map; Copying the original signal waveform semantic space deepening feature map to obtain a backup original signal waveform semantic space deepening feature map; Performing attention-gated autocorrelation coding on the original signal waveform semantic space deepening feature map and the backup original signal waveform semantic space deepening feature map to obtain an original signal waveform semantic gated local autocorrelation library weight matrix; Taking the environmental parameter multimodal time series associated feature vector as the value feature vector, calculating the product between the original signal waveform semantic gated local autocorrelation library weight matrix and the environmental parameter multimodal time series associated feature vector to obtain the environmental parameter multimodal time series cross-modal interactive fusion condition weight vector; The environmental parameter multimodal temporal cross-modal interactive fusion condition weight vector is used as the weight vector, and the original signal waveform semantic feature map is weighted channel by channel to obtain the environmental fluctuation correction signal waveform semantic feature map.

7. The method for detecting access security of a distribution network carrier communication system according to claim 6, characterized in that: The original signal waveform semantic feature map is subjected to layer normalization and convolution coding processing to obtain the original signal waveform semantic space deepening feature map, including: Performing layer normalization processing on the original signal waveform semantic feature map to obtain a layer normalized original signal waveform semantic feature map; The layer-normalized original signal waveform semantic feature map is point-convolutionally encoded and then spatially convolutionally encoded to obtain the original signal waveform semantic space deepening feature map.

8. The method for detecting access security of a distribution network carrier communication system according to claim 7, characterized in that: The original signal waveform semantic space deepening feature map and the backup original signal waveform semantic space deepening feature map are subjected to attention gating-based autocorrelation encoding to obtain an original signal waveform semantic gating local autocorrelation library weight matrix, including: Reshape the backup original signal waveform semantic space deepening feature map and the original signal waveform semantic space deepening feature map respectively to obtain a backup original signal waveform semantic space deepening feature matrix as a simulated query feature matrix and an original signal waveform semantic space deepening feature matrix as a simulated key feature matrix; Calculating the product of the query-like feature matrix and the transposed matrix of the key-like feature matrix to obtain the original signal waveform semantic local autocorrelation library matrix; The original signal waveform semantic local autocorrelation library matrix is ​​input into the attention gating unit based on the GELU function to obtain the original signal waveform semantic gated local autocorrelation library weight matrix.

Images