Method for mining and analyzing abnormal security behaviors of industrial internet
By collecting and preprocessing multi-dimensional data in real time on industrial Internet nodes, establishing a baseline model using unsupervised learning algorithms to detect abnormal behaviors, solving the problems of inefficiency and limitations of feature extraction in the existing technology, and achieving more efficient and accurate security abnormal behavior detection.
Patent Information
- Application Number
- CN202411510513.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-28
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-10-28
AI Technical Summary
The existing industrial Internet security abnormal behavior detection methods are inefficient and cannot effectively detect unknown or mutated attack behaviors. There are limitations in feature extraction and data preprocessing, which cannot fully reflect the operating status of the equipment and the complexity of network behavior.
By installing data acquisition devices at various nodes of the industrial Internet, multi-dimensional data is collected in real time, data preprocessing and feature extraction are performed, and baseline models are established using unsupervised learning algorithms to detect abnormal behaviors that deviate from the baseline model. Specific steps include data cleaning, missing value filling, noise filtering, data format standardization, and timing and frequency feature extraction, combining mean clustering and discrete Fourier transform to generate key features.
Improve the accuracy and response speed of industrial Internet security abnormal behavior detection, can better reflect the operating status of equipment and networks, adapt to the complex and changeable industrial Internet environment, reduce false alarms and missed reports, and improve the security and reliability of the system. sex.
Smart Images

Figure CN119030799A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of industrial Internet security technology, and in particular to a method for mining and analyzing abnormal security behaviors of the industrial Internet. Background Art
[0002] In the existing technology, the detection methods of abnormal behavior in industrial Internet security mainly rely on traditional rule-based detection and signature-based detection. These methods identify abnormal behavior through predefined rules or known attack signatures. In the industrial Internet environment, data is usually collected through network traffic analysis, log analysis and equipment status monitoring, and analyzed through data mining and machine learning techniques to detect potential security threats and abnormal behaviors.
[0003] However, there are some major problems in the existing technologies. First, rule-based detection and signature-based detection methods require frequent updates of rules and signature libraries, and cannot effectively detect unknown or mutated attack behaviors. Second, traditional methods are often inefficient when processing massive multi-dimensional data in the industrial Internet, and are prone to false positives and false negatives. In addition, the existing technologies have limitations in feature extraction and data preprocessing, and cannot fully reflect the complexity of device operation status and network behavior.
[0004] Therefore, there is an urgent need for a method that can efficiently and accurately mine and analyze security abnormal behaviors in the industrial Internet to improve the security and reliability of the system. Summary of the invention
[0005] The present application provides a method for mining and analyzing abnormal security behaviors of the industrial Internet, so as to improve the detection accuracy and response speed of abnormal security behaviors of the industrial Internet.
[0006] The method for mining and analyzing abnormal industrial Internet security behaviors provided in this application includes: The multi-dimensional data in the industrial Internet is collected in real time through the data collection device installed on each node of the industrial Internet, wherein the multi-dimensional data includes equipment operation status data, network traffic data, user operation logs and sensor data; Preprocessing the multi-dimensional data to generate a preprocessed data set, wherein the preprocessing includes data cleaning, missing value filling, noise filtering and data format standardization; Performing feature extraction on the preprocessed data set to obtain key features reflecting the device operation status and network behavior, wherein the key features include timing features and frequency features; The extracted key features are analyzed through unsupervised learning algorithms to establish a baseline model for the normal operation of the Industrial Internet. During the actual operation of the Industrial Internet, new data is collected in real time and preprocessed and feature extracted. The extracted features are input into the baseline model for comparison to detect abnormal behaviors that deviate from the baseline model.
[0007] Furthermore, the extracted key features are analyzed by the unsupervised learning algorithm to establish a baseline model for the normal operation of the industrial Internet, including: Represent the extracted key features as a set of feature vectors , where each key feature vector Including timing characteristics and frequency characteristics, is the number of key features; use Mean clustering algorithm for feature vector set Perform cluster analysis and set the number of clusters , calculate the set of eigenvectors The distance from the feature vector to the cluster center, iteratively update the cluster center , until convergence; among them, the Cluster Centers Calculate according to the following formula (1): in, Indicates that it belongs to The set of feature vectors of clusters; is the feature vector set The number of eigenvectors in ; is the mean vector of all eigenvectors; To adjust the parameters, used to balance the offset and dispersion of cluster centers; For each feature vector , according to the following formula (2), calculate the distance to the nearest cluster center as the anomaly score: in, is the eigenvector The anomaly score of is the eigenvector With cluster center The Euclidean distance between To adjust the parameters, it is used to adjust the weights of Euclidean distance and absolute difference; Represents the feature vector In the The values in the dimensions; Represents the cluster center In the The values in the dimensions; is the dimension of the feature vector; For all anomaly scores Perform statistical analysis and fit its probability distribution; calculate the average value of the anomaly score based on the fitted probability distribution and standard deviation ; Calculate the threshold value according to the following formula (3): in, is the threshold for detecting abnormal behavior; is the number of anomaly scores; It is Anomaly score; The anomaly score is less than The set of feature vectors Serves as a baseline model for the normal operation of the Industrial Internet.
[0008] Furthermore, the step of inputting the extracted feature vector into the baseline model for comparison to detect abnormal behaviors that deviate from the baseline model includes: According to the following formula (4), calculate the new anomaly score , in, is the feature vector corresponding to the extracted feature; if , it is judged as abnormal behavior; otherwise, it is judged as normal behavior.
[0009] Furthermore, the feature extraction of the preprocessed data set to obtain key features reflecting the device operation status and network behavior includes: The preprocessed data set is divided into multiple subsets according to fixed time windows, and the mean is calculated in each time window according to the following formulas (5) and (6): and standard deviation : , in, The first data points; For the The weight of each data point; For the The timestamp of each data point, is the time attenuation coefficient; is the total number of data points in the time window; in, The first data points; For the The weight of each data point; For the The timestamp of each data point; is the time attenuation coefficient; is the total number of data points in the time window; The autocorrelation coefficient is calculated according to the following formula (7): : in, The first data points; is the number of lag steps; is the total number of data points in the time window; The calculated mean , Standard Deviation And the autocorrelation coefficient Determined as a time series feature; Apply discrete Fourier transform to the data in each time window to convert the time domain data into the frequency domain and obtain the frequency domain representation ; Calculate the main frequency according to the following formula (8): : The calculated Determined as frequency feature; The determined timing features and frequency features are spliced together to generate key features that reflect the device operating status and network behavior.
[0010] Furthermore, the data collection device installed on each node of the industrial Internet collects multi-dimensional data in the industrial Internet in real time, including: The sensors installed on industrial equipment collect the equipment operation status data in real time, and transmit the collected equipment operation status data to the data processing center through the industrial Internet for centralized storage and processing; Capture network traffic data in the industrial Internet through network traffic monitoring equipment; perform packet processing on network traffic data, extract source address, destination address, protocol type and data volume in the data packet; sort network traffic data by timestamp and transmit it to the data processing center for analysis; Deploy logging tools on user terminals and servers to record user operation behaviors; transmit user operation log data to the data processing center through a secure encrypted channel.
[0011] Furthermore, the sensor includes a temperature sensor, a pressure sensor and a vibration sensor.
[0012] Furthermore, user operation behaviors include user login, logout, command execution, and file access.
[0013] Furthermore, the preprocessing of the multi-dimensional data to generate a preprocessed data set includes: Conduct a preliminary screening of the collected data to eliminate obviously erroneous and invalid data; Check the integrity and consistency of data, merge and remove duplicate data; Identify and process abnormal values and outliers to ensure data accuracy.
[0014] Furthermore, the preprocessing of the multi-dimensional data to generate a preprocessed data set also includes: Analyze the missing values in the data set to determine the distribution and impact of the missing values; Use linear interpolation to fill missing values in time series data; For non-time series data, mean imputation or multiple imputation methods are used.
[0015] Furthermore, the preprocessing of the multi-dimensional data to generate a preprocessed data set also includes: Use a moving average filter to smooth the data and reduce the impact of random noise; The data is filtered in the frequency domain by combining high-pass filter and low-pass filter to remove high-frequency and low-frequency noise; The data format is standardized and converted into a unified format and dimension to facilitate subsequent analysis and processing.
[0016] The beneficial effects of the technical solution provided by this application include: (1) Through the data acquisition devices installed on each node of the industrial Internet, multi-dimensional data is collected in real time, including equipment operation status data, network traffic data, user operation logs and sensor data. In this way, various activities in the industrial Internet can be fully monitored to ensure the real-time and comprehensiveness of the data. (2) The multi-dimensional data is preprocessed to generate a preprocessed data set. The preprocessing steps include data cleaning, missing value filling, noise filtering and data format standardization. These steps greatly improve the quality of the data, reduce the errors caused by missing data and noise, and improve the accuracy of subsequent analysis. (3) Feature extraction is performed on the preprocessed data set to obtain key features reflecting the equipment operation status and network behavior, including time series features and frequency features. Accurate feature extraction can better reflect the operation status of the equipment and network, and provide high-quality input for subsequent anomaly detection. (4) The extracted key features are analyzed through an unsupervised learning algorithm to establish a baseline model for the normal operation of the industrial Internet. Unsupervised learning can automatically adapt to data changes, establish a dynamic baseline model, and effectively cope with the complexity and variability of the industrial Internet environment. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 This is a flowchart of a method for mining and analyzing abnormal security behaviors in the industrial Internet provided in the first embodiment of the present application. DETAILED DESCRIPTION
[0018] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.
[0019] The first embodiment of the present application provides a method for mining and analyzing abnormal security behaviors in the industrial Internet. Figure 1 , which is a schematic diagram of the first embodiment of the present application. Figure 1 The first embodiment of the present application provides a method for mining and analyzing abnormal security behaviors in the industrial Internet, which is described in detail.
[0020] Step S101: Collect multi-dimensional data in the industrial Internet in real time through the data acquisition device installed on each node of the industrial Internet, wherein the multi-dimensional data includes equipment operation status data, network traffic data, user operation logs and sensor data.
[0021] In step S101, multi-dimensional data in the Industrial Internet is collected in real time through the data collection devices installed on each node of the Industrial Internet. First, the collection of equipment operation status data requires the installation of corresponding sensors on each industrial equipment, including but not limited to temperature sensors, pressure sensors, vibration sensors, and humidity sensors. These sensors will monitor the operating status of the equipment in real time and transmit the data to the centralized data processing center through the Industrial Internet. Secondly, the collection of network traffic data requires the deployment of network traffic monitoring devices on network nodes. These devices can capture network traffic data packets, extract information such as source address, destination address, protocol type, data packet size, and transmission time, and store these data in the network traffic database.
[0022] At the same time, the collection of user operation logs requires the installation of log recording tools on user terminals and servers. These tools record every user operation, including login, logout, command execution, file access, and system change operation logs. These log data are transmitted to the data processing center through a secure encrypted channel to ensure the integrity and confidentiality of the data during transmission. In addition, the data acquisition devices of environmental sensors also include gas sensors, light sensors, etc. These sensors are used to monitor changes in the industrial environment to ensure that environmental conditions are within a safe range.
[0023] To ensure the comprehensiveness and accuracy of data collection, each data collection device needs to be calibrated and maintained regularly. Sensor data is initially processed by edge computing devices, such as data compression and format conversion, to reduce transmission delays and bandwidth usage. All collected data is regularly sent to the data center, where it is further stored and processed. Encryption technology, such as SSL / TLS, is used during data transmission to ensure the security and tamper-proof nature of data transmission.
[0024] In summary, step S101 ensures the real-time, comprehensive and accurate collection of multi-dimensional data in the industrial Internet by precisely arranging and using a variety of data acquisition devices, combined with edge computing and encrypted transmission technology, providing a solid data foundation for subsequent preprocessing, feature extraction and abnormal behavior detection.
[0025] Furthermore, the data collection device installed on each node of the industrial Internet collects multi-dimensional data in the industrial Internet in real time, including: The sensors installed on industrial equipment collect the equipment operation status data in real time, and transmit the collected equipment operation status data to the data processing center through the industrial Internet for centralized storage and processing; Capture network traffic data in the industrial Internet through network traffic monitoring equipment; perform packet processing on network traffic data, extract source address, destination address, protocol type and data volume in the data packet; sort network traffic data by timestamp and transmit it to the data processing center for analysis; Deploy logging tools on user terminals and servers to record user operation behaviors; transmit user operation log data to the data processing center through a secure encrypted channel.
[0026] Through the data collection devices installed on each node of the Industrial Internet, multi-dimensional data in the Industrial Internet is collected in real time, including the following aspects: First, various types of sensors are installed on industrial equipment to collect real-time data on the equipment's operating status. These sensors may include temperature sensors, pressure sensors, vibration sensors, and humidity sensors, etc. These sensors need to be properly connected and calibrated with industrial equipment to ensure the accuracy and reliability of the collected data. The collected data is transmitted to the data processing center via the industrial Internet for centralized storage and processing. During the data transmission process, advanced communication protocols and encryption technologies are used to ensure the security and integrity of data transmission.
[0027] Secondly, network traffic data in the industrial Internet is captured through network traffic monitoring devices. These devices are usually deployed at key nodes of the network and can capture and record detailed information of network data packets in real time. In order to process and analyze these data packets, they need to be sub-packetized. Specifically, sub-packetization includes decomposing the data packets into multiple fields and extracting key information such as source address, destination address, protocol type and data volume. These field information helps to understand the specific situation and traffic characteristics of network communication. Next, the extracted network traffic data is sorted according to timestamps to ensure the timing consistency of the data, and the sorted data is transmitted to the data processing center for further analysis. The data processing center uses specialized network analysis tools and algorithms to conduct in-depth analysis of these data to identify potential network anomalies and security threats.
[0028] Deploying logging tools on user terminals and servers to record user operations is another important step. These logging tools can record in detail all kinds of user operations in the system, including login, logout, file access, command execution, and system configuration changes. In order to protect the security of these sensitive data, log data uses secure encryption channels during transmission, such as SSL / TLS protocol, to ensure that the data will not be stolen or tampered with during transmission. After transmission to the data processing center, these log data will be stored in a dedicated log database and classified and indexed for subsequent query and analysis. By analyzing user operation logs, abnormal user behavior can be detected, such as frequent login failures, abnormal file access, and unauthorized system configuration changes.
[0029] The entire data collection and transmission process requires the comprehensive application of multiple technologies and equipment to ensure the comprehensiveness, accuracy and security of the data. Through efficient data processing and analysis methods, various security anomalies in the industrial Internet can be identified and responded to in a timely manner to ensure the stable operation of the system and information security.
[0030] Furthermore, the sensor includes a temperature sensor, a pressure sensor and a vibration sensor.
[0031] First of all, temperature sensors are used to monitor the temperature changes of industrial equipment. Temperature sensors are installed in key parts of the equipment, such as engines, heat exchangers or other heat-generating components, to ensure that the actual operating temperature of the equipment can be accurately measured. The temperature sensor needs to be connected to the control system of the equipment, usually by cable or wireless transmission to transmit temperature data to the data acquisition device. In order to ensure the accuracy of the data, the temperature sensor should be calibrated before installation to eliminate measurement errors.
[0032] Secondly, pressure sensors are used to monitor pressure changes during the operation of equipment. Pressure sensors are usually installed on the hydraulic system, pneumatic system or other key components of the equipment to obtain system pressure data in real time. The pressure sensor is connected to the pressure point of the equipment by using appropriate interfaces and pipelines. The collected pressure data is transmitted to the industrial Internet network through the data acquisition device. During the installation and use of the pressure sensor, attention should be paid to its range and response speed to ensure that it can accurately reflect the operating status of the equipment.
[0033] Vibration sensors are used to monitor the vibration of equipment and help identify whether the equipment is in normal operation. Vibration sensors are usually installed on the bearings, transmission devices or other parts of the equipment that are prone to vibration. By measuring the vibration frequency and amplitude of the equipment, vibration sensors can detect the smooth operation of the equipment and potential mechanical failures. The vibration sensor is installed on the equipment through appropriate fixtures and ensures its stability to obtain accurate vibration data. The vibration data is transmitted to the Industrial Internet through a data acquisition device for further analysis.
[0034] In order to achieve effective data transmission, all data collected by sensors will be transmitted to the data processing center through the industrial Internet. During the transmission process, advanced communication protocols and encryption technologies are used to ensure the integrity and security of the data. After receiving the data, the data processing center will centrally store and process it. Through the comprehensive analysis of temperature, pressure and vibration data, the operating status of the equipment can be fully understood, and potential abnormal behaviors and failures can be discovered and warned in a timely manner.
[0035] Furthermore, user operation behaviors include user login, logout, command execution, and file access.
[0036] First, for monitoring user login and logout behavior, the logging tool records relevant information every time a user attempts to log in to the system. The recorded content includes the user name, login time, login IP address, and login result (success or failure). When the user successfully logs in, the system will generate a log record containing the above information and a unique identifier for the user's session. Similarly, when the user logs out of the system, the logging tool records the time of the user's logout and the session identifier, ensuring that every login and logout behavior is recorded in detail.
[0037] Secondly, for command execution monitoring, the logging tool will capture every command executed by the user in the system. The recorded content includes the time of command execution, command content, executing user and command execution result. This can track the user's operation path and specific behavior in the system in detail, especially when an abnormality or security incident occurs, the specific operation steps and responsible persons can be traced through the command log. For example, when a user executes important system configuration commands or database operation commands on the server, the logging tool will record these operations in detail for subsequent audit and analysis.
[0038] Monitoring file access behavior is also critical, especially when sensitive data or business-critical files are involved. Logging tools record all user access behaviors to files, including file read, modification, deletion, and creation operations. The recorded content includes the path to the accessed file, access time, operation type, executing user, and operation results. For example, when a user reads a file containing sensitive information, the logging tool will record the behavior in detail, including the time the file was read, the user identity, and the reading result.
[0039] To ensure data security, all log data will be transmitted to the data processing center through a secure encrypted channel after being recorded. The data transmission uses encryption technology, such as SSL / TLS protocol, to ensure that the logs will not be intercepted or tampered with by unauthorized third parties during the transmission process. The log data transmitted to the data processing center will be stored in a dedicated log database, and will be classified and indexed for subsequent query and analysis.
[0040] In the data processing center, specialized analysis tools analyze the collected log data to identify potential security anomalies. For example, by analyzing login and logout logs, abnormally frequent login attempts or unauthorized login behaviors can be detected. By analyzing command execution logs, abnormal system configuration changes or database operations can be identified. By analyzing file access logs, unauthorized file reading or data leakage behaviors can be discovered.
[0041] Step S102: preprocessing the multi-dimensional data to generate a preprocessed data set, wherein the preprocessing includes data cleaning, missing value filling, noise filtering and data format standardization.
[0042] In step S102, the multi-dimensional data is preprocessed to generate a preprocessed data set. This process includes multiple key steps to ensure that the collected data can be accurately and effectively used for subsequent analysis and processing.
[0043] First, perform data cleaning. The purpose of data cleaning is to identify and correct or delete errors, duplications, and invalid data in the data. For example, for equipment operating status data, there may be abnormal readings due to sensor failure or communication problems. By setting a reasonable threshold range, obviously unreasonable temperature, pressure, or vibration readings can be identified and deleted. For network traffic data, the cleaning process can include deleting malformed data packets, identifying and deleting duplicate traffic records. User operation logs also need to be cleaned to remove invalid or duplicate log records and ensure that the format of each log record is consistent.
[0044] Next, deal with missing values in the data. In the industrial Internet environment, missing data is a common problem, which may be caused by sensor failure, network delays, etc. There are many methods that can be used to fill in missing values. For time series data, linear interpolation is a common method that infers missing values through the linear relationship of known data points. For more complex situations, Lagrange interpolation or multiple interpolation can be used. These methods take into account the overall trends and patterns of the data and provide more accurate filling results. For non-time series data, mean filling is a simple and effective method to replace missing values with the mean of the variable, thereby reducing the impact of missing data on the analysis results.
[0045] Then, noise filtering is performed. Noise is a random error or irrelevant information in the data, which will affect the accuracy of the analysis results. In order to filter out noise, a variety of filtering techniques can be used. For example, the moving average filter reduces random fluctuations by smoothing the data, which is suitable for stable time series data. For time series data with periodicity or trend, a combination of high-pass filter and low-pass filter can be used to remove low-frequency and high-frequency noise respectively, thereby retaining the main information in the data. In addition, Kalman filter is also an effective noise filtering method, especially suitable for data processing in dynamic systems.
[0046] Finally, the data format is standardized. The multi-dimensional data in the Industrial Internet comes from different devices and systems, with different formats and dimensions. In order to ensure the consistency and comparability of the data, the data needs to be standardized. Common standardization methods include normalizing the data by its mean and standard deviation to convert it into a standard normal distribution with a mean of zero and a standard deviation of one. Another method is to scale the data to the range of [0,1] by subtracting the minimum value and dividing by the range (maximum value minus minimum value). Standardized data is not only easy to compare and analyze, but also improves the efficiency and effectiveness of model training.
[0047] Through the above detailed and clear preprocessing steps, the quality and consistency of the data can be effectively improved, laying a solid foundation for subsequent feature extraction and abnormal behavior detection.
[0048] Furthermore, the preprocessing of the multi-dimensional data to generate a preprocessed data set includes: Conduct a preliminary screening of the collected data to eliminate obviously erroneous and invalid data; Check the integrity and consistency of data, merge and remove duplicate data; Identify and process abnormal values and outliers to ensure data accuracy.
[0049] First, the collected data is preliminarily screened to remove obviously erroneous and invalid data. The purpose of this step is to ensure the basic quality and credibility of the data set. Specific operations include checking whether the data format is correct, whether the numerical range is reasonable, and whether there are obvious logical errors. For example, for equipment operating status data, if the temperature value recorded by the sensor is far beyond the normal operating range of the equipment (such as -100°C or 1000°C), then these data are obviously unreasonable and need to be removed. Similarly, for network traffic data, if the source address or destination address format of the data packet is incorrect, these data should also be marked as invalid data and removed.
[0050] Secondly, check the integrity and consistency of the data, and merge and deduplicate duplicate data. Data integrity checks include confirming that all necessary data fields have been filled and that no key data is missing. For example, in a user operation log, each record should include information such as user name, operation type, and timestamp. Data records that lack this information will be considered incomplete. Data consistency checks ensure that data remains consistent across different sources or time periods. For example, data recorded by multiple sensors for the same event should be consistent. If inconsistencies are found, further investigation and processing are required. Duplicate data can be identified by comparing the content and timestamps of the records. If two records are exactly the same within a short period of time, they can be merged into one record to reduce data redundancy and improve data processing efficiency.
[0051] Finally, abnormal values and outliers are identified and processed to ensure data accuracy. Outliers are values that deviate significantly from other data points, which may be caused by sensor failure, data transmission errors, etc. Methods for handling outliers include using statistical analysis techniques, such as using standard deviation methods to identify data points that exceed a certain standard deviation range, and choose to delete, correct or mark them as abnormal depending on the specific situation. For example, for vibration sensor data, if the vibration amplitude of some data points is significantly higher than the vibration range under normal operating conditions, these data points can be marked as outliers and further analyzed. Outliers are data points that are significantly different from other data points and may reflect potential system failures or abnormal behavior. Methods for handling outliers include using cluster analysis techniques to divide data points into normal groups and outlier groups, and performing detailed analysis and processing on the outlier groups.
[0052] Through the above detailed steps, the collected multi-dimensional data is preprocessed to generate a high-quality and reliable preprocessed data set. This data set not only eliminates obvious errors and invalid data to ensure the integrity and consistency of the data, but also improves the accuracy and reliability of the data by identifying and processing abnormal values and outliers.
[0053] Furthermore, the preprocessing of the multi-dimensional data to generate a preprocessed data set also includes: Analyze the missing values in the data set to determine the distribution and impact of the missing values; Use linear interpolation to fill missing values in time series data; For non-time series data, mean imputation or multiple imputation methods are used.
[0054] First, the missing values in the data set are analyzed in detail. The purpose of this step is to determine the distribution and impact of the missing values in the data set. When analyzing missing values, it is necessary to count the number and location of missing values in each variable to understand the overall distribution of missing values. For example, some sensors may fail to collect data during a specific time period due to failure, resulting in missing values concentrated in that time period. Through such analysis, it can be clear which variables and time periods need to be processed in particular.
[0055] For missing values in time series data, linear interpolation is used to fill them. Linear interpolation is a commonly used filling method that uses the values of adjacent data points to calculate missing values according to a linear relationship. Specifically, for a missing value in time series data, the two nearest known values before and after the missing value can be found, and a linear interpolation can be calculated based on the time and value of these two known values. For example, if at time point and With known value and , and the time point If there are missing values, then the missing values It can be calculated by the following formula: , This method is simple and computationally efficient, and is suitable for situations where data changes are relatively stable.
[0056] For non-time series data, mean imputation or multiple imputation is used for processing. Mean imputation is the simplest imputation method, which replaces missing values with the mean of the variable. The advantage of this method is that it is simple to calculate, but the disadvantage is that it may underestimate the variability of the data. The specific operation is to calculate the mean of each variable, and then replace the missing values with the mean of the corresponding variable. Multiple imputation is a more complex and accurate method that estimates missing values by generating replacement values multiple times and takes into account the inherent variability of the data. Multiple imputation usually includes the following steps: first, generate multiple data sets containing replacement values; then, analyze each data set to obtain multiple analysis results; finally, combine these results to obtain the final imputation result. Multiple imputation can better maintain the statistical characteristics and variability of data, and is suitable for situations where the data is more complex and uneven.
[0057] Through the above steps, the missing values in the data set can be effectively processed, improving the completeness of the data set and the accuracy of the analysis.
[0058] Furthermore, the preprocessing of the multi-dimensional data to generate a preprocessed data set also includes: Use a moving average filter to smooth the data and reduce the impact of random noise; The data is filtered in the frequency domain by combining high-pass filter and low-pass filter to remove high-frequency and low-frequency noise; The data format is standardized and converted into a unified format and dimension to facilitate subsequent analysis and processing.
[0059] Preprocessing the multi-dimensional data to generate a preprocessed data set also includes several important steps, including smoothing using a moving average filter, frequency domain filtering using a combination of a high-pass filter and a low-pass filter, and formatting the data.
[0060] First, the data is smoothed using a moving average filter to reduce the impact of random noise. A moving average filter smoothes the data by calculating the average of a series of data points. This process involves sliding a fixed-size window over the data sequence, calculating the average of the data points in the window at each position, and then replacing the data point in the center of the window with this average. This can effectively reduce random fluctuations in the data, making the data smoother and more stable. For example, for time series data, a moving average filter can help eliminate short-term fluctuations in sensor data and highlight long-term trends.
[0061] Next, the data is filtered in the frequency domain using a combination of high-pass and low-pass filters to remove high- and low-frequency noise. Frequency domain filtering selectively retains or removes signals within a specific frequency range by analyzing the frequency components of the data. High-pass filters are used to remove low-frequency noise, such as slowly changing trends or DC components, while low-pass filters are used to remove high-frequency noise, such as rapidly changing interference signals. In specific operations, the data can be first subjected to a fast Fourier transform (FFT) to convert the time domain data to the frequency domain, and then high-pass and low-pass filters are applied to remove unwanted frequency components, and finally the data is converted back to the time domain through an inverse Fourier transform (IFFT). The processed data is not only smoother, but also more accurately reflects the actual operating status.
[0062] Finally, the data is formatted and converted into a unified format and dimension for subsequent analysis and processing. The purpose of format standardization is to eliminate the differences between different data sources and different types of data so that the data are numerically comparable. Standardization methods include normalizing the data by its mean and standard deviation to convert it into a standard normal distribution with a mean of zero and a standard deviation of one, or scaling the data to the range of [0,1]. For data of different dimensions, such as temperature (Celsius), pressure (Pascal), and vibration (Hertz), standardization can convert them to the same dimensional range through linear or logarithmic transformation. In this way, the standardized data is not only convenient for comparison and analysis, but also improves the efficiency and effectiveness of subsequent model training and algorithm processing.
[0063] Through the above detailed steps, the collected multi-dimensional data is smoothed, filtered in the frequency domain, and formatted in a standardized manner to generate a high-quality, uniformly formatted preprocessed data set. This data set not only reduces the impact of noise, ensures the smoothness and stability of the data, but also eliminates the differences between different data sources, improving the comparability of the data and processing efficiency.
[0064] Step S103: extracting features from the preprocessed data set to obtain key features reflecting the device operating status and network behavior, wherein the key features include timing features and frequency features.
[0065] In step S103, feature extraction is performed on the preprocessed data set to obtain key features reflecting the device operating status and network behavior, and these features include time series features and frequency features.
[0066] First, for time series feature extraction, the preprocessed data needs to be divided into fixed time windows. The data in each time window is processed as a unit. For example, if one minute is selected as the time window, the data of each minute is regarded as a separate unit for analysis. In each time window, multiple time series statistics are calculated, including mean, standard deviation, autocorrelation coefficient, etc. The mean can reflect the overall level of the data in the time window, and the standard deviation is used to measure the degree of fluctuation of the data. The autocorrelation coefficient is used to describe the correlation between time series data at different time points, so as to identify periodic and trend changes in the data.
[0067] Next, for the extraction of frequency features, it is necessary to perform frequency domain analysis on the data in each time window. A common method is to apply discrete Fourier transform (DFT) to convert the time domain data into the frequency domain. Through Fourier transform, the original time series data can be decomposed into the superposition of different frequency components to obtain a frequency domain representation. For each frequency component, its amplitude and phase are calculated, with special attention paid to the main frequency and its corresponding amplitude, because the main frequency reflects the most significant periodic component in the data. By analyzing the spectrum diagram, the main frequency components in the data and their energy distribution can be identified, thereby extracting the frequency characteristics that reflect the operating status of the system.
[0068] In addition, in order to further improve the effect of feature extraction, wavelet transform can be used in combination to perform multi-scale analysis on the data. Wavelet transform can analyze the local characteristics of the data in both the time domain and the frequency domain by decomposing the data into components of different scales. This method is particularly suitable for processing non-stationary time series data because it can capture the instantaneous changes and mutation characteristics in the data.
[0069] Finally, the extracted time series features and frequency features are fused to form a comprehensive feature vector. In order to ensure the uniformity and comparability of the feature vectors, it is usually necessary to standardize the feature vectors. Standardization methods include subtracting the mean of each feature and dividing it by its standard deviation, or scaling the feature to a specific range (such as [0,1]). The standardized feature vector can be better used for subsequent model training and anomaly detection.
[0070] Through the above steps, high-quality key features can be extracted from the preprocessed data set. These features can not only accurately reflect the equipment operation status and network behavior, but also provide a reliable data foundation for subsequent unsupervised learning algorithms, thereby realizing the effective mining and analysis of abnormal security behaviors of the industrial Internet.
[0071] Furthermore, the feature extraction of the preprocessed data set to obtain key features reflecting the device operation status and network behavior includes: The preprocessed data set is divided into multiple subsets according to fixed time windows, and the mean is calculated in each time window according to the following formulas (5) and (6): and standard deviation : , in, The first data points; For the The weight of each data point; For the The timestamp of each data point, is the time attenuation coefficient; is the total number of data points in the time window; in, The first data points; For the The weight of each data point; For the The timestamp of each data point; is the time attenuation coefficient; is the total number of data points in the time window; The autocorrelation coefficient is calculated according to the following formula (7): : in, The first data points; is the number of lag steps; is the total number of data points in the time window; The calculated mean , Standard Deviation And the autocorrelation coefficient Determined as a time series feature; Apply discrete Fourier transform to the data in each time window to convert the time domain data into the frequency domain and obtain the frequency domain representation ; Calculate the main frequency according to the following formula (8): : The calculated Determined as frequency feature; The determined timing features and frequency features are spliced together to generate key features that reflect the device operating status and network behavior.
[0072] First, the preprocessed data set is divided into multiple subsets according to fixed time windows. This division process is to perform detailed statistical analysis within each time window to ensure real-time monitoring and analysis of device operating status and network behavior. The size of the time window can be selected according to the specific application scenario, such as one minute or five minutes.
[0073] In each time window, the mean of the data is first calculated The mean is calculated as: Among them, among them, The first data points; For the The weight of each data point; For the The timestamp of each data point; is the time attenuation coefficient; is the total number of data points in the time window; the mean reflects the central trend of the data in the time window and is a basic description of the data distribution.
[0074] Next, calculate the standard deviation of the data , the standard deviation is calculated as: in, The first data points; For the The weight of each data point; For the The timestamp of each data point; is the time attenuation coefficient; is the total number of data points in the time window; The standard deviation is a measure of the dispersion of the data, describing how the data points are distributed around the mean. A high standard deviation means that the data points are more dispersed, while a low standard deviation means that the data points are concentrated around the mean.
[0075] In addition, the autocorrelation coefficient needs to be calculated , the calculation formula of the autocorrelation coefficient is: in, The first data points; The autocorrelation coefficient is used to measure the correlation between time series data at different time points, which can reveal the periodicity and trend changes of the data.
[0076] The calculated mean , Standard Deviation And the autocorrelation coefficient Determined as time series features; these time series features can reflect the operating status of the device within a specific time window.
[0077] Next, a discrete Fourier transform (DFT) is applied to the data in each time window to convert the time domain data into the frequency domain to obtain the frequency domain representation Through frequency domain analysis, the frequency characteristics of the data can be revealed and the periodic components in the data can be captured.
[0078] Apply discrete Fourier transform to the data in each time window to convert the time domain data into the frequency domain and obtain the frequency domain representation ; Calculate the main frequency according to the following formula (8): : The main frequency reflects the most significant periodic component in the data and is an important result of frequency domain analysis. Determined as frequency feature.
[0079] Finally, the determined timing features and frequency features are spliced to generate key features that reflect the operating status of the equipment and network behavior. The spliced feature vector contains information in both timing and frequency dimensions, which can fully describe the operating status and network behavior of the equipment. These key features will serve as the basis for subsequent unsupervised learning algorithm analysis and provide reliable data support for the detection of abnormal security behaviors in the industrial Internet.
[0080] Step S104: Analyze the extracted key features through an unsupervised learning algorithm to establish a baseline model for the normal operation of the Industrial Internet; during the actual operation of the Industrial Internet, collect new data in real time and perform preprocessing and feature extraction, input the extracted features into the baseline model for comparison, and detect abnormal behaviors that deviate from the baseline model.
[0081] In step S104, the extracted key features are analyzed by an unsupervised learning algorithm to establish a baseline model for the normal operation of the industrial Internet. In the actual implementation process, it is first necessary to prepare a training data set containing a large amount of normal operation data, which should have undergone the aforementioned preprocessing and feature extraction steps.
[0082] To build a baseline model, select a suitable unsupervised learning algorithm, such as K-means clustering. Before starting cluster analysis, set the number of clusters K, which can be determined based on the specific application scenario and data characteristics. Usually, the value of K is selected using empirical rules or data-based analysis methods to ensure that the clustering results can accurately represent the intrinsic structure of the data.
[0083] The feature vector set is input into the K-means algorithm, and K initial cluster centers are randomly selected first. Then, each feature vector is assigned to the cluster center closest to it, which is achieved by calculating the Euclidean distance from the feature vector to each cluster center. Each feature vector is assigned to the corresponding cluster according to the minimum distance principle.
[0084] Next, the new center position of each cluster is calculated, that is, the mean of all feature vectors belonging to the same cluster is calculated. After the cluster center is updated, the feature vectors are reallocated to the new cluster center, and the above process is repeated until the location of the cluster center no longer changes significantly or the preset number of iterations is reached. The final cluster center represents the baseline model under normal operating conditions.
[0085] In the actual operation of the Industrial Internet, new data collected in real time will be preprocessed and feature extracted to obtain new feature vectors. In order to detect abnormal behavior, these new feature vectors are input into the baseline model and their distances to each cluster center are calculated. Specifically, for each new feature vector, its Euclidean distance to all cluster centers is calculated, and the smallest distance is selected as the abnormal score of the feature vector.
[0086] Based on the normal behavior data in the training data set, calculate the anomaly scores of all feature vectors and obtain the distribution of anomaly scores. Through statistical analysis, calculate the mean and standard deviation of the anomaly scores and set the detection threshold. Usually, the threshold can be set to the mean plus twice the standard deviation, or the multiple can be adjusted according to specific needs to control the sensitivity and false alarm rate of anomaly detection.
[0087] In practical applications, when the anomaly score of a new feature vector exceeds a preset threshold, it is judged as abnormal behavior. This anomaly detection method based on unsupervised learning can adapt to the dynamic changes of data and improve the accuracy and robustness of detection.
[0088] Furthermore, the extracted key features are analyzed by the unsupervised learning algorithm to establish a baseline model for the normal operation of the industrial Internet, including: Represent the extracted key features as a set of feature vectors , where each key feature vector Including timing characteristics and frequency characteristics, is the number of key features; use Mean clustering algorithm for feature vector set Perform cluster analysis and set the number of clusters , calculate the set of eigenvectors The distance from the feature vector to the cluster center, iteratively update the cluster center , until convergence; among them, the Cluster Centers Calculate according to the following formula (1): in, Indicates that it belongs to The set of feature vectors of clusters; is the feature vector set The number of eigenvectors in ; is the mean vector of all eigenvectors; To adjust the parameters, used to balance the offset and dispersion of cluster centers; For each feature vector , according to the following formula (2), calculate the distance to the nearest cluster center as the anomaly score: in, is the eigenvector The anomaly score of is the eigenvector With cluster center The Euclidean distance between To adjust the parameters, it is used to adjust the weights of Euclidean distance and absolute difference; Represents the feature vector In the The values in the dimensions; Represents the cluster center In the The values in the dimensions; is the dimension of the feature vector; For all anomaly scores Perform statistical analysis and fit its probability distribution; calculate the average value of the anomaly score based on the fitted probability distribution and standard deviation ; Calculate the threshold value according to the following formula (3): in, is the threshold for detecting abnormal behavior; is the number of anomaly scores; It is Anomaly score; The anomaly score is less than The set of feature vectors Serves as a baseline model for the normal operation of the Industrial Internet.
[0089] First, the extracted key features are represented as a set of feature vectors , where each eigenvector Including time series characteristics and frequency characteristics; is the number of eigenvectors. This set contains all the eigenvectors extracted from the dataset and is ready for clustering analysis.
[0090] Next, use Mean clustering algorithm for feature vector set Perform cluster analysis. First, set the number of clusters , this value can be selected according to the characteristics and needs of the data. Then, the distance from each feature vector to the cluster center is calculated through an iterative process, and the cluster center is updated. Initialization After finding cluster centers, each feature vector is assigned to the nearest cluster center. Cluster Centers Calculated according to the following formula: in, Indicates that it belongs to The set of feature vectors of clusters; is the feature vector set The number of eigenvectors in ; is the mean vector of all eigenvectors; To adjust the parameters, it is used to balance the offset and dispersion of the cluster center; this formula takes into account the mean and dispersion of the eigenvectors, making the cluster center more representative.
[0091] Then, for each eigenvector , calculate its distance to the nearest cluster center as the anomaly score. The calculation formula is as follows: in, is the eigenvector The anomaly score of is the eigenvector With cluster center The Euclidean distance between To adjust the parameters, it is used to adjust the weights of Euclidean distance and absolute difference; Represents the feature vector In the The values in the dimensions; Represents the cluster center In the The values in the dimensions; is the dimension of the feature vector; For all anomaly scores Perform statistical analysis and fit its probability distribution. By calculating the average value of the anomaly score and standard deviation , the distribution characteristics of the anomaly score can be determined. The calculation formula is as follows: in, is the threshold for detecting abnormal behavior; is the number of anomaly scores; It is This formula not only takes into account the mean and standard deviation, but also introduces a skewness adjustment term, making the threshold calculation more accurate and sensitive.
[0092] Finally, the anomaly score is less than The set of feature vectors As a baseline model for the normal operation of the Industrial Internet, this method can effectively identify and detect abnormal behaviors and ensure the normal operation and security of the system.
[0093] Furthermore, the step of inputting the extracted feature vector into the baseline model for comparison to detect abnormal behaviors that deviate from the baseline model includes: According to the following formula (4), calculate the new anomaly score : in, is the feature vector corresponding to the extracted feature; if , it is judged as abnormal behavior; otherwise, it is judged as normal behavior.
[0094] First, the newly extracted feature vector Input into the established baseline model. The baseline model is obtained by analyzing and clustering the data under normal operation, and contains multiple cluster centers. .
[0095] To calculate the anomaly score for a new feature vector , we need to calculate its distance to all cluster centers. Specifically, for the new feature vector , calculate its Euclidean distance to each cluster center, and select the smallest distance as the anomaly score. The calculation formula is as follows: , In this formula, Represents the new feature vector With Cluster Centers By comparing the distance between the new feature vector and all cluster centers, the closest cluster center can be found, and this distance is used as the abnormal score of the feature vector.
[0096] Next, the anomaly score calculated by With the preset threshold Compare to determine whether the new feature vector is abnormal behavior. , then the behavior corresponding to the feature vector is determined to be abnormal behavior; otherwise, it is determined to be normal behavior.
[0097] Specifically, the threshold It is calculated based on statistical analysis of historical data, usually taking into account the mean, standard deviation and other statistical characteristics of the data to ensure that the threshold can effectively distinguish between normal and abnormal behavior. In actual applications, the setting of thresholds can be adjusted according to specific application scenarios and needs to balance the risks of false positives and false negatives.
[0098] A second embodiment of the present application provides an electronic device, the electronic device comprising: processor; The memory is used to store a program, and when the program is read and executed by the processor, the program executes the method for mining and analyzing abnormal security behaviors of the industrial Internet provided in the first embodiment of the present application.
[0099] The third embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method for mining and analyzing abnormal security behaviors in the industrial Internet provided in the first embodiment of the present application is executed.
[0100] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.
Claims
1. A method for mining and analyzing abnormal security behaviors in the industrial Internet, characterized in that: include: The multi-dimensional data in the industrial Internet is collected in real time through the data collection device installed on each node of the industrial Internet, wherein the multi-dimensional data includes equipment operation status data, network traffic data, user operation logs and sensor data; Preprocess multi-dimensional data to generate preprocessed data sets, where preprocessing includes data cleaning, missing value filling, noise filtering, and data format standardization; Perform feature extraction on the preprocessed data set to obtain key features that reflect the device operating status and network behavior, where the key features include timing features and frequency features; The extracted key features are analyzed through unsupervised learning algorithms to establish a baseline model for the normal operation of the Industrial Internet. During the actual operation of the Industrial Internet, new data is collected in real time and preprocessed and feature extracted. The extracted features are input into the baseline model for comparison to detect abnormal behaviors that deviate from the baseline model.
2. The method for mining and analyzing abnormal security behaviors of the industrial Internet according to claim 1 is characterized in that: The extracted key features are analyzed by unsupervised learning algorithms to establish a baseline model for the normal operation of the industrial Internet, including: Represent the extracted key features as a set of feature vectors , where each key feature vector Including timing characteristics and frequency characteristics, is the number of key features; use Mean clustering algorithm for feature vector set Perform cluster analysis and set the number of clusters , calculate the set of eigenvectors The distance from the feature vector to the cluster center, iteratively update the cluster center , until convergence; among them, the Cluster Centers Calculate according to the following formula (1): in, Indicates that it belongs to The set of feature vectors of clusters; is the feature vector set The number of eigenvectors in ; is the mean vector of all eigenvectors; To adjust the parameters, used to balance the offset and dispersion of cluster centers; For each feature vector , according to the following formula (2), calculate the distance to the nearest cluster center as the anomaly score: in, is the eigenvector The anomaly score of is the eigenvector With cluster center The Euclidean distance between To adjust the parameters, it is used to adjust the weights of Euclidean distance and absolute difference; Represents the feature vector In the The value of the dimension; Represents the cluster center In the The value of the dimension; is the dimension of the feature vector; For all anomaly scores Perform statistical analysis and fit its probability distribution; calculate the average value of the anomaly score based on the fitted probability distribution and standard deviation ; Calculate the threshold value according to the following formula (3): in, is the threshold for detecting abnormal behavior; is the number of anomaly scores; It is Anomaly score; The anomaly score is less than The set of feature vectors Serves as a baseline model for the normal operation of the Industrial Internet.
3. The method for mining and analyzing abnormal security behaviors of the industrial Internet according to claim 2 is characterized in that: The extracted feature vectors are input into the baseline model for comparison to detect abnormal behaviors that deviate from the baseline model, including: According to the following formula (4), calculate the new anomaly score , in, is the feature vector corresponding to the extracted feature; if , it is judged as abnormal behavior; otherwise, it is judged as normal behavior.
4. The method for mining and analyzing abnormal security behaviors of the industrial Internet according to claim 1 is characterized in that: Perform feature extraction on the preprocessed data set to obtain key features that reflect the device operating status and network behavior, including: The preprocessed data set is divided into multiple subsets according to fixed time windows, and the mean is calculated in each time window according to the following formulas (5) and (6): and standard deviation : , in, The first data points; For the The weight of each data point; For the The timestamp of each data point, is the time attenuation coefficient; is the total number of data points in the time window; in, The first data points; For the The weight of each data point; For the The timestamp of each data point; is the time attenuation coefficient; is the total number of data points in the time window; The autocorrelation coefficient is calculated according to the following formula (7): : in, The first data points; is the number of lag steps; is the total number of data points in the time window; The calculated mean , Standard Deviation And the autocorrelation coefficient Determined as a time series feature; Apply discrete Fourier transform to the data in each time window to convert the time domain data into the frequency domain and obtain the frequency domain representation ; Calculate the main frequency according to the following formula (8): : The calculated Determined as frequency feature; The determined timing features and frequency features are spliced together to generate key features that reflect the device operating status and network behavior.
5. The method for mining and analyzing abnormal security behaviors of the industrial Internet according to claim 1 is characterized in that: Through the data collection devices installed on each node of the Industrial Internet, multi-dimensional data in the Industrial Internet is collected in real time, including: The sensors installed on industrial equipment collect the equipment operation status data in real time, and transmit the collected equipment operation status data to the data processing center through the industrial Internet for centralized storage and processing; Capture network traffic data in the industrial Internet through network traffic monitoring equipment; perform packet processing on network traffic data, extract source address, destination address, protocol type and data volume in the data packet; sort network traffic data by timestamp and transmit it to the data processing center for analysis; Deploy logging tools on user terminals and servers to record user operation behaviors; transmit user operation log data to the data processing center through a secure encrypted channel.
6. The method for mining and analyzing abnormal security behaviors of the industrial Internet according to claim 5 is characterized in that: The sensors include temperature sensors, pressure sensors and vibration sensors.
7. The method for mining and analyzing abnormal security behaviors of the industrial Internet according to claim 5 is characterized in that: User operations include login, logout, command execution, and file access.
8. The method for mining and analyzing abnormal security behaviors of the industrial Internet according to claim 1 is characterized in that: Preprocess the multi-dimensional data to generate a preprocessed data set, including: Conduct a preliminary screening of the collected data to eliminate obviously erroneous and invalid data; Check the integrity and consistency of data, merge and remove duplicate data; Identify and process abnormal values and outliers to ensure data accuracy.
9. The method for mining and analyzing abnormal security behaviors of the industrial Internet according to claim 8 is characterized in that: Preprocessing multi-dimensional data to generate preprocessed data sets also includes: Analyze the missing values in the data set to determine the distribution and impact of the missing values; Use linear interpolation to fill missing values in time series data; For non-time series data, mean imputation or multiple imputation methods are used.
10. The method for mining and analyzing abnormal security behaviors of the industrial Internet according to claim 8, characterized in that: Preprocessing multi-dimensional data to generate preprocessed data sets also includes: Use a moving average filter to smooth the data and reduce the impact of random noise; The data is filtered in the frequency domain by combining high-pass filter and low-pass filter to remove high-frequency and low-frequency noise; The data format is standardized and converted into a unified format and dimension to facilitate subsequent analysis and processing.
Citation Information
Patent Citations
Automatic time series regression method and device
CN111967616A
Electricity stealing detection method and system based on time sequence automatic encoder and medium
CN113327008A
Fusion system security management method and system based on data analysis
CN117544420A
Production intelligent monitoring method, system and device based on industrial internet
CN118094449A
Network security communication control method and system based on Internet of Things
CN118337539A
Cited By
Electromechanical equipment state management method and system based on Internet of Things, and storage medium
CN119377586A
Flow anomaly identification method for numerical control machine tool protocol
CN119513671A
Industrial control protocol abnormal behavior analysis and detection method, equipment, medium and program product
CN119652591A
Time series data preprocessing system and method based on communication network
CN119906533A
Log data processing method based on cable phase measurement
CN120179518A
