A device port security alarm system

By monitoring device port temperature in real time and using a deep neural network model for time-series analysis, the problem of traditional systems being unable to identify device port temperature patterns has been solved. This enables anomaly detection and early warning for device ports, ensuring the security of device ports and business continuity.

CN118175065BActive Publication Date: 2024-11-01YUNNAN KUHAO TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410600727.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-15
Publication Date
2024-11-01
Estimated Expiration
2044-05-15

AI Technical Summary

Technical Problem

Traditional device port security alarm systems cannot accurately identify the complexity and changing patterns of device port temperature modes, leading to false alarms or missed alarms. They cannot distinguish between normal and abnormal temperature modes and cannot adapt to temperature changes under different conditions.

Method used

By monitoring the device port temperature in real time using a temperature sensor and performing time-series analysis using a deep neural network model, the device port temperature time-series correlation pattern features are extracted. The receptive field focusing module and noise reduction technology are used for feature focusing and noise suppression to achieve abnormal temperature pattern detection and issue early warning.

Benefits of technology

It enables real-time monitoring and anomaly detection of device port temperature patterns, allowing for timely identification of potential problems, reducing service interruption time, and ensuring device port security.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118175065B_ABST
    Figure CN118175065B_ABST
Patent Text Reader

Abstract

The present invention discloses a device port safety alarm system, which uses a temperature sensor to monitor and collect the real-time temperature value of the device port in real time, and introduces a data processing and analysis algorithm at the back end to perform a time series analysis of the real-time temperature value of the device port, so as to detect the abnormal temperature mode of the device port, and issue an early warning prompt when the temperature mode is detected to be abnormal. In this way, the abnormality detection of the device port can be automatically performed by real-time monitoring of the temperature mode and change of the device port, so as to timely discover potential problems and issue early warnings. In this way, the problem can be quickly located and corresponding measures can be taken, thereby reducing the time and impact of service interruption and ensuring the safety of the device port.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of intelligent alarms, and more specifically, to a device port security alarm system. Background Art

[0002] The equipment ports in the communication room are connected to the interfaces of various network devices, including servers, switches, routers, etc. The security of these equipment ports is crucial to protecting the integrity, confidentiality and availability of data.

[0003] Once an abnormality occurs in a device port, it may cause abnormal interruption of system operation, affect the correctness of data in the system or damage the system database, causing partial or even all data loss. At the same time, port abnormality may also cause security problems of device ports, affect the business continuity of the communication room, and cause significant losses to the enterprise.

[0004] The temperature of the device port is an important indicator that can reflect the operating status of the device and the possible risk of failure. Traditional device port security alarm systems can usually only monitor the temperature value changes of the device port and use fixed thresholds to determine whether the temperature is abnormal. However, the temperature of the device port is affected by many factors, such as ambient temperature, device load, etc. Therefore, the traditional monitoring method cannot capture the complexity and change rules of the temperature pattern, resulting in the inability to accurately judge potential abnormal situations. At the same time, the fixed threshold alarm method cannot adapt to temperature changes in different situations, which is prone to false alarms or missed alarms.

[0005] In addition, traditional device port security alarm systems lack pattern recognition capabilities in detecting abnormal temperature patterns by monitoring changes in device port temperature values. That is, temperature patterns usually have certain periodicity and regularity, such as the temperature rise process when the device starts or periodic load changes. Traditional systems cannot accurately identify and analyze these patterns, resulting in the inability to distinguish between normal temperature changes and abnormal temperature patterns.

[0006] Therefore, an optimized equipment port security alarm system is desired. Summary of the invention

[0007] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a device port security alarm system, which uses a temperature sensor to monitor and collect the real-time temperature value of the device port in real time, and introduces a data processing and analysis algorithm at the back end to perform a timing analysis of the real-time temperature value of the device port, so as to detect the abnormal temperature mode of the device port, and issue an early warning prompt when the temperature mode is abnormal. In this way, the abnormality detection of the device port can be automatically performed by real-time monitoring of the temperature mode and changes of the device port, so as to timely discover potential problems and issue early warnings. In this way, the problem can be quickly located and corresponding measures can be taken, thereby reducing the time and impact of service interruptions and ensuring the safety of the device port.

[0008] According to one aspect of the present application, there is provided a device port security alarm system, comprising:

[0009] The device port real-time temperature acquisition module is used to obtain the time series of the real-time temperature values ​​of the monitored device port collected by the temperature sensor in a predetermined time period;

[0010] A temperature time series vector domain conversion module is used to arrange the time series of the real-time temperature values ​​into a real-time temperature time series input vector according to the time dimension and then obtain a real-time temperature time series conversion image of the device port through a vector-image converter;

[0011] A temperature timing correlation pattern feature extraction module is used to extract features from the device port real-time temperature timing conversion image through a device port temperature timing pattern feature extractor based on a deep neural network model to obtain a device port real-time temperature timing correlation pattern feature graph;

[0012] A temperature timing correlation pattern feature noise reduction module, used for passing the device port real-time temperature timing correlation pattern feature graph through a noise reducer based on a position mask module to obtain a device port real-time temperature timing correlation pattern feature graph after noise reduction;

[0013] A receptive field focusing module, used for passing the noise-reduced device port real-time temperature timing correlation mode characteristic diagram through the receptive field focusing module to obtain a local manifestation device port real-time temperature timing correlation mode characteristic;

[0014] The temperature mode abnormality detection and warning module is used to determine whether there is an abnormality in the temperature mode of the device port based on the real-time temperature time series correlation mode characteristics of the local manifestation device port, and determine whether to generate a warning information prompt.

[0015] Compared with the prior art, the device port security alarm system provided by the present application monitors and collects the real-time temperature value of the device port in real time through a temperature sensor, and introduces a data processing and analysis algorithm at the back end to perform a time series analysis of the real-time temperature value of the device port, so as to detect abnormal temperature patterns of the device port, and issue an early warning when an abnormal temperature pattern is detected. In this way, the abnormality detection of the device port can be automatically performed by real-time monitoring of the temperature pattern and changes of the device port, so as to timely discover potential problems and issue early warnings. In this way, the problem can be quickly located and corresponding measures can be taken, thereby reducing the time and impact of service interruptions and ensuring the safety of the device port. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0017] Figure 1 is a block diagram of a device port security alarm system according to an embodiment of the present application;

[0018] Figure 2 A system architecture diagram of a device port security alarm system according to an embodiment of the present application;

[0019] Figure 3 4 is a block diagram of a temperature mode abnormality detection and early warning module in a device port security alarm system according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0021] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0022] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0023] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0024] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0025] The temperature of the device port is an important indicator that can reflect the operating status of the device and the possible risk of failure. The traditional device port security alarm system can usually only monitor the temperature value changes of the device port and use a fixed threshold to determine whether the temperature is abnormal. However, the temperature of the device port is affected by many factors, such as ambient temperature, device load, etc. Therefore, the traditional monitoring method cannot capture the complexity and change rules of the temperature pattern, resulting in the inability to accurately judge potential abnormal conditions. At the same time, the alarm method with a fixed threshold cannot adapt to temperature changes in different situations, which is prone to false alarms or missed alarms. In addition, the traditional device port security alarm system lacks pattern recognition capabilities in detecting temperature pattern anomalies by monitoring the temperature value changes of the device port. That is, the temperature pattern usually has a certain periodicity and regularity, such as the temperature rise process when the device is started or the periodic load changes. The traditional system cannot accurately identify and analyze these patterns, resulting in the inability to distinguish between normal temperature changes and abnormal temperature patterns. Therefore, an optimized device port security alarm system is desired.

[0026] In the technical solution of the present application, a device port security alarm system is proposed. Figure 1 4 is a block diagram of a device port security alarm system according to an embodiment of the present application. Figure 2 FIG. 1 is a system architecture diagram of a device port security alarm system according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to the embodiment of the present application, the device port security alarm system 300 includes: a device port real-time temperature acquisition module 310, which is used to obtain the time series of the real-time temperature values ​​of the monitored device port collected by the temperature sensor in a predetermined time period; a temperature time series vector domain conversion module 320, which is used to arrange the time series of the real-time temperature values ​​into a real-time temperature time series input vector according to the time dimension and then obtain the device port real-time temperature time series conversion image through a vector-image converter; a temperature time series association pattern feature extraction module 330, which is used to extract features from the device port real-time temperature time series conversion image through a device port temperature time series pattern feature extractor based on a deep neural network model to obtain the device port real-time temperature time series conversion image. A port real-time temperature timing association pattern characteristic diagram; a temperature timing association pattern characteristic noise reduction module 340, used to pass the device port real-time temperature timing association pattern characteristic diagram through a denoiser based on a position mask module to obtain a denoised device port real-time temperature timing association pattern characteristic diagram; a receptive field focusing module 350, used to pass the denoised device port real-time temperature timing association pattern characteristic diagram through a receptive field focusing module to obtain a local manifestation device port real-time temperature timing association pattern characteristic; a temperature pattern abnormality detection and early warning module 360, used to determine whether there is an abnormality in the temperature pattern of the device port based on the local manifestation device port real-time temperature timing association pattern characteristic, and determine whether to generate an early warning information prompt.

[0027] In particular, the device port real-time temperature acquisition module 310 is used to obtain the time series of the real-time temperature value of the monitored device port collected by the temperature sensor in a predetermined time period. It should be understood that the temperature value of the device port plays a key role in determining whether there is an abnormality in the temperature mode of the device port. By monitoring and analyzing the temperature value of the device port, abnormal conditions in the temperature mode can be identified to help determine the operating status and safety of the device. Specifically, in a specific example of the present application, the real-time temperature value of the device port is collected by monitoring the temperature sensor in real time, and a data processing and analysis algorithm is introduced in the back end to perform a time series analysis of the real-time temperature value of the device port, so as to detect abnormal temperature modes of the device port, and issue an early warning prompt when an abnormal temperature mode is detected.

[0028] It is worth mentioning that a temperature sensor is a device used to measure the temperature of an environment or object. It can convert temperature into an electrical signal or other form of output for temperature monitoring, control or data recording.

[0029] In particular, the temperature time series vector domain conversion module 320 is used to arrange the time series of the real-time temperature value into a real-time temperature time series input vector according to the time dimension and then pass it through a vector-image converter to obtain a real-time temperature time series conversion image of the device port. Considering that compared with a simple time series vector representation, the time series image data can provide more information, including the time series relationship, fluctuation and trend change of the temperature data. Therefore, in order to better express and analyze the pattern and trend of the temperature change of the device port, in the technical solution of the present application, the time series of the real-time temperature value is further arranged into a real-time temperature time series input vector according to the time dimension and then passed through a vector-image converter to obtain a real-time temperature time series conversion image of the device port. Such a conversion can provide a more intuitive and visual way to understand the time series pattern and characteristics of the temperature data, and help detect abnormal conditions of the temperature pattern. In other words, the real-time temperature time series conversion image of the device port can more intuitively display the overall trend, periodic pattern, abnormal points and other information of the temperature change of the device port, which helps to more comprehensively analyze the characteristics and abnormal conditions of the temperature pattern.

[0030] In particular, the temperature timing association pattern feature extraction module 330 is used to perform feature extraction on the device port real-time temperature timing conversion image through a device port temperature timing pattern feature extractor based on a deep neural network model to obtain a device port real-time temperature timing association pattern feature map. Specifically, in a specific example of the present application, the deep neural network model is a convolutional neural network model. That is, a device port temperature timing pattern feature extractor based on a convolutional neural network model with excellent performance in extracting implicit features of an image is used to perform feature extraction on the device port real-time temperature timing conversion image to extract the timing association pattern feature information about the device port temperature in the device port real-time temperature timing conversion image, thereby obtaining a device port real-time temperature timing association pattern feature map. More specifically, each layer of the device port temperature timing pattern feature extractor based on the convolutional neural network model performs the following operations on the input data in the forward pass of the layer: convolution processing on the input data to obtain a convolution feature map; pooling based on the local feature matrix on the convolution feature map to obtain a pooling feature map; and nonlinear activation on the pooling feature map to obtain an activation feature map; wherein the output of the last layer of the device port temperature timing pattern feature extractor based on the convolutional neural network model is the real-time temperature timing association pattern feature map of the device port, and the input of the first layer of the device port temperature timing pattern feature extractor based on the convolutional neural network model is the real-time temperature timing conversion image of the device port.

[0031] It is worth noting that Convolutional Neural Network (CNN) is a deep learning model that is particularly suitable for processing tasks with grid-structured data (such as images, audio). CNN has achieved great success in the field of computer vision and is widely used in tasks such as image classification, object detection, and image generation. The core idea of ​​CNN is to extract and learn the feature representation of data through convolutional layers, pooling layers, and fully connected layers. The following are the main components and working principles of CNN: Convolutional layer: The convolutional layer is the core component of CNN. It contains a set of learnable convolution kernels (also called filters), each of which performs a convolution operation on the input data to extract local features. The convolution operation is to multiply the convolution kernel by a part of the input data element by element, and sum the results to obtain an element of the output feature map. The convolutional layer can automatically learn features of different sizes and gradually extract abstract feature representations from local to global; Activation function: A nonlinear activation function, such as ReLU, is applied to the output of the convolutional layer to introduce nonlinear features. The activation function performs an element-by-element nonlinear transformation on the output of the convolutional layer to enhance the expressive power of the model; Pooling layer: The pooling layer is used to reduce the size of the feature map and retain the most important features. Common pooling operations include maximum pooling and average pooling, which take the maximum value or average value in the local area as the output after pooling. The pooling layer can reduce the spatial dimension of the feature, improve the computational efficiency of the model, and has a certain degree of translation invariance; Fully connected layer: The fully connected layer flattens the feature map output by the pooling layer into a one-dimensional vector and connects it to one or more fully connected layers. The neurons in the fully connected layer are connected to all the neurons in the previous layer, and the features are combined and classified by learning weights and biases; Dropout layer: Dropout is a regularization technique used to reduce the overfitting of the model. During the training process, the Dropout layer randomly discards the output of some neurons with a certain probability, thereby forcing the model to learn a more robust feature representation. By stacking multiple convolutional layers, pooling layers, and fully connected layers, CNN can extract and combine features layer by layer, thereby achieving advanced representation and classification of complex data. During the training process, CNN optimizes network parameters through the back-propagation algorithm so that the model can better fit the training data.

[0032] In particular, the temperature timing association pattern feature noise reduction module 340 is used to pass the device port real-time temperature timing association pattern feature map through a noise reducer based on a position mask module to obtain a device port real-time temperature timing association pattern feature map after noise reduction. It should be understood that in order to highlight the semantic parts of the device port real-time temperature timing association pattern feature map with more important information about the temperature timing pattern and change characteristics, and reduce noise and irrelevant information interference, in the technical solution of the present application, the device port real-time temperature timing association pattern feature map is further passed through a noise reducer based on a position mask module to obtain a device port real-time temperature timing association pattern feature map after noise reduction. It should be understood that the position mask mechanism can mask the characteristic values ​​that meet the conditions at each position in the input feature map according to specific conditions to give the characteristic values ​​that meet the conditions a greater weight. In this way, positions with more important semantics can be retained and highlighted more in the noise reduction process, while unimportant positions are suppressed or filtered out, thereby improving the quality and accuracy of the output feature map, which is conducive to more accurate detection and early warning of temperature pattern anomalies at device ports. Specifically, the device port real-time temperature timing correlation mode characteristic graph is passed through a denoiser based on a position mask module to obtain the device port real-time temperature timing correlation mode characteristic graph after denoising, including: the device port real-time temperature timing correlation mode characteristic graph is passed through a denoiser based on a position mask module to process the device port real-time temperature timing correlation mode characteristic graph with the following denoising formula to obtain the device port real-time temperature timing correlation mode characteristic graph after denoising;

[0033] Wherein, the noise reduction formula is:

[0034]

[0035]

[0036] in, is a characteristic diagram of the real-time temperature timing correlation mode of the device port, is the threshold feature map, is a characteristic diagram of the real-time temperature timing correlation mode of the device port after noise reduction, is the characteristic value of each position in the characteristic diagram of the real-time temperature timing correlation mode of the device port, is the feature value of each position in the threshold feature map, and is a hyperparameter, Represents the condition set of the threshold function.

[0037] In particular, the receptive field focusing module 350 is used to pass the real-time temperature timing correlation mode characteristic diagram of the device port after noise reduction through the receptive field focusing module to obtain the local manifestation device port real-time temperature timing correlation mode characteristics. In particular, in a specific example of the present application, the receptive field focusing module 350 includes: a temperature timing correlation mode local manifestation unit, which is used to pass the real-time temperature timing correlation mode characteristic diagram of the device port after noise reduction through the receptive field focusing module to obtain the local manifestation device port real-time temperature timing correlation mode feature vector as the local manifestation device port real-time temperature timing correlation mode characteristics.

[0038] Specifically, the temperature time series correlation pattern local manifestation unit is used to pass the de-noised device port real-time temperature time series correlation pattern feature map through the receptive field focusing module to obtain the local manifestation device port real-time temperature time series correlation pattern feature vector as the local manifestation device port real-time temperature time series correlation pattern feature. Considering that there are local time series change patterns and associations about the device port temperature in the device port real-time temperature time series conversion image, these local features are particularly important for the abnormality of the temperature pattern of the device port. However, since the structure of the convolutional neural network model in the traditional sense is a convolution plus pooling layer, neurons with local receptive fields (such as 3×3 convolution kernels) can be used to extract features and reduce dimensions by pooling, and significant information on each channel can be obtained at the same time, but this method has a large receptive field, which will cause the extracted device port real-time temperature time series conversion image to have a blurred time series change pattern and trend characteristics about the temperature, and it is easy to ignore the quality detail feature information with resolution in the feature map. Therefore, in order to focus on a specific local area of ​​temperature time series distribution to further improve the expression and analysis capabilities of temperature time series features, in the technical solution of the present application, the denoised device port real-time temperature time series association pattern feature map is further passed through the receptive field focusing module to obtain a local explicit device port real-time temperature time series association pattern feature vector. It should be understood that the receptive field focusing module can add a 1×1 convolution kernel and a ReLU activation function after the traditional convolution. From the perspective of cross-channel pooling, this is equivalent to implementing cascaded cross-channel weighted pooling on a normal convolution layer, so that the model can learn the relationship between channels, and more efficiently model and characterize the local time series feature information about the device disconnection temperature in the device port real-time temperature time series conversion image, making the network more sensitive in a specific area and highlighting the local time series feature information of the port temperature in the area. In this way, the key time series features of the device port temperature in the local area can be extracted, and other feature information that is not related to the port temperature anomaly detection can be ignored, thereby obtaining the local explicit device port real-time temperature time series association pattern feature vector with more expressiveness and discrimination. More specifically, the real-time temperature timing correlation mode characteristic graph of the device port after noise reduction is passed through the receptive field focusing module to obtain the real-time temperature timing correlation mode characteristic vector of the local manifestation device port as the real-time temperature timing correlation mode characteristic of the local manifestation device port, including: passing the real-time temperature timing correlation mode characteristic graph of the device port after noise reduction through the receptive field focusing module to obtain the real-time temperature timing correlation mode characteristic vector of the local manifestation device port as the real-time temperature timing correlation mode characteristic of the local manifestation device port according to the following formula; wherein the formula is:

[0039]

[0040] in, It is global mean pooling; () is the activation function, represents one-dimensional convolution, represents a three-dimensional convolution, is a characteristic diagram of the real-time temperature timing correlation mode of the device port after noise reduction, It is the characteristic vector of the real-time temperature timing correlation pattern of the port of the local manifestation device.

[0041] It is worth mentioning that in other specific examples of the present application, the real-time temperature timing correlation pattern feature graph of the device port after denoising can also be passed through the receptive field focusing module in other ways to obtain the local manifestation device port real-time temperature timing correlation pattern feature, for example: input the real-time temperature timing correlation pattern feature graph of the device port after denoising; the receptive field focusing module is used to focus on a specific receptive field area to extract the local manifestation device port real-time temperature timing correlation pattern feature. The following are the steps of the receptive field focusing module: first, apply a one-dimensional convolution operation to the real-time temperature timing correlation pattern feature graph of the device port after denoising; after the convolution operation, the size of the receptive field can be adjusted by changing the size or stride of the convolution kernel to control the specific receptive field area focused by the module; apply a nonlinear activation function, such as ReLU, to introduce nonlinear features; after processing by the receptive field focusing module, the local manifestation device port real-time temperature timing correlation pattern feature vector is obtained as the local manifestation device port real-time temperature timing correlation pattern feature.

[0042] In particular, the temperature mode abnormality detection and warning module 360 ​​is used to determine whether the temperature mode of the device port is abnormal based on the local manifestation device port real-time temperature time series correlation mode characteristics, and determine whether to generate a warning information prompt. In particular, in a specific example of the present application, if Figure 3 As shown, the temperature mode abnormality detection and early warning module 360 ​​includes: an equipment port temperature mode abnormality detection unit 361, which is used to pass the local manifestation equipment port real-time temperature timing association pattern feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether there is an abnormality in the temperature mode of the equipment port; an early warning prompt unit 362, which is used to generate an early warning information prompt in response to the classification result that the temperature mode of the equipment port is abnormal, such as high temperature abnormality, low temperature abnormality, temperature fluctuation abnormality, etc.

[0043] Specifically, the device port temperature mode abnormality detection unit 361 is used to pass the local manifestation device port real-time temperature timing association mode feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether the temperature mode of the device port is abnormal. That is, the real-time temperature timing association mode feature information of the device port after local feature manifestation is used for classification processing to perform temperature mode abnormality detection of the device port. In particular, in a specific example of the present application, in response to the classification result that the temperature mode of the device port is abnormal, an early warning information prompt is generated. In this way, the abnormality detection of the device port can be automatically performed by real-time monitoring of the temperature mode and changes of the device port, so as to timely discover potential problems and issue early warnings, so that the problem can be quickly located and corresponding measures can be taken to ensure the safety of the device port. More specifically, the local manifestation device port real-time temperature timing association pattern feature vector is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether there is an abnormality in the temperature pattern of the device port, including: using multiple fully connected layers of the classifier to fully connect encode the local manifestation device port real-time temperature timing association pattern feature vector to obtain an encoded classification feature vector; and passing the encoded classification feature vector through the Softmax classification function of the classifier to obtain the classification result.

[0044] In the technical solution of the present application, each feature matrix of the device port real-time temperature timing association pattern feature map expresses the timing association characteristics of the device port real-time temperature in the local time domain based on the convolution kernel scale, and the channel distribution of the convolutional neural network model is followed between the feature matrices. The device port real-time temperature timing association pattern feature map is passed through a denoiser based on a position mask module, and the noise part of the device port real-time temperature timing association pattern feature map can be filtered out based on the comparison between the position feature value and the predetermined threshold to obtain the device port real-time temperature timing association pattern feature map after denoising. When the device port real-time temperature timing association pattern feature map after denoising is passed through the receptive field focusing module to obtain the local manifestation device port real-time temperature timing association pattern feature vector, considering that the channel dimension between the feature matrices and the spatial dimension within the feature matrix of the device port real-time temperature timing association pattern feature map after denoising is poor in overall distribution, the obtained local manifestation device port real-time temperature timing association pattern feature vector is poorly constrained relative to the overall distribution regression of the classifier, affecting the accuracy of the classification result obtained by the classifier.

[0045] Specifically, in a preferred embodiment of the present application, the device port temperature pattern abnormality detection unit includes: calculating the sum of the square root of the length of the local manifestation device port real-time temperature timing association pattern feature vector and the reciprocal of the square root of the second norm of the local manifestation device port real-time temperature timing association pattern feature vector to obtain the local manifestation device port real-time temperature timing association pattern offset feature vector; calculating the exponential function of the local manifestation device port real-time temperature timing association pattern offset feature vector with a natural constant as the base to obtain the local manifestation device port real-time temperature timing association pattern offset prediction feature. vector; calculate the dot product of the local manifestation device port real-time temperature timing association pattern feature vector and the norm of the local manifestation device port real-time temperature timing association pattern feature vector and the weight hyperparameter to obtain the local manifestation device port real-time temperature timing association pattern constraint feature vector; calculate the dot sum of the local manifestation device port real-time temperature timing association pattern offset prediction feature vector and the local manifestation device port real-time temperature timing association pattern constraint feature vector to obtain the optimized local manifestation device port real-time temperature timing association pattern feature vector; pass the optimized local manifestation device port real-time temperature timing association pattern feature vector through the classifier to obtain the classification result.

[0046] Specifically, considering that the temporal propagation aggregation distribution of the local manifested device port real-time temperature timing association pattern feature vector associated with the data embedding encoding timing context of the transaction data is based on the overall distribution regression constraint of the classifier, in the above preferred example, the structured norm representation of the local manifested device port real-time temperature timing association pattern feature vector is used as the local canonical coordinate for each eigenvalue of the local manifested device port real-time temperature timing association pattern feature vector to determine the class rotation offset of the vector overall distribution representation of the local manifested device port real-time temperature timing association pattern feature vector relative to the eigenvalue for the class offset prediction direction of each eigenvalue of the local manifested device port real-time temperature timing association pattern feature vector as the center, and the bounding box of the vector eigenvalue distribution of the local manifested device port real-time temperature timing association pattern feature vector is used to perform eigenvalue constraints to improve the constraint of the local manifested device port real-time temperature timing association pattern feature vector under the overall regression distribution, thereby improving the training speed of the model and the accuracy of the classification result obtained by the local manifested device port real-time temperature timing association pattern feature vector through the classifier. In this way, the temperature patterns and changes of device ports can be monitored in real time to automatically detect abnormalities of device ports, so as to promptly discover potential problems and issue early warnings. In this way, problems can be quickly located and corresponding measures can be taken, thereby reducing the time and impact of service interruptions and ensuring the security of device ports.

[0047] Specifically, the early warning prompt unit 362 is used to generate an early warning information prompt in response to the classification result that the temperature mode of the device port is abnormal. In a specific example of the present application, when the classification result is that the temperature mode of the device port is abnormal, a early warning information prompt is generated in response to the abnormal result.

[0048] It is worth mentioning that in other specific examples of the present application, it is also possible to determine whether the temperature pattern of the device port is abnormal based on the local manifestation of the real-time temperature time series correlation pattern characteristics of the device port, and determine whether to generate an early warning information prompt, for example: extract features from the real-time temperature time series data of the device port; perform local manifestation processing on the extracted feature vector. Local manifestation can enhance the correlation and importance between features through some methods, such as self-attention mechanism, convolution operation, etc., so as to capture local information in the temperature pattern; represent the features after local manifestation processing as feature vectors as input; use a trained anomaly detection model or threshold method to perform anomaly detection on the feature vector. The anomaly detection model can be an unsupervised learning model, such as a method based on clustering or outlier detection; based on the results of the anomaly detection, determine whether the temperature pattern of the device port is abnormal. If the anomaly detection result indicates that an anomaly exists, a corresponding early warning message is generated.

[0049] As described above, the device port security alarm system 300 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a device port security alarm algorithm. In a possible implementation, the device port security alarm system 300 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the device port security alarm system 300 can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the device port security alarm system 300 can also be one of the many hardware modules of the wireless terminal.

[0050] Alternatively, in another example, the device port security alarm system 300 and the wireless terminal may also be separate devices, and the device port security alarm system 300 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0051] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A device port security alarm system, characterized in that: include: The device port real-time temperature acquisition module is used to obtain the time series of the real-time temperature values ​​of the monitored device port collected by the temperature sensor in a predetermined time period; A temperature time series vector domain conversion module is used to arrange the time series of the real-time temperature values ​​into a real-time temperature time series input vector according to the time dimension and then obtain a real-time temperature time series conversion image of the device port through a vector-to-image converter. Such conversion provides a visual way to understand the time series pattern and characteristics of the temperature data and helps detect abnormal conditions of the temperature pattern. The real-time temperature time series conversion image of the device port can show the overall trend, periodic pattern, and abnormal point information of the temperature change of the device port; A temperature timing association pattern feature extraction module is used to extract features from the real-time temperature timing conversion image of the device port through a device port temperature timing pattern feature extractor based on a deep neural network model to obtain a device port real-time temperature timing association pattern feature map, wherein the deep neural network model is a convolutional neural network model, and each layer of the device port temperature timing pattern feature extractor based on the convolutional neural network model is used to perform the following operations on the input data in the forward transfer of the layer: convolution processing on the input data to obtain a convolution feature map; pooling on the convolution feature map based on a local feature matrix to obtain a pooling feature map; and nonlinear activation on the pooling feature map to obtain an activation feature map; wherein the output of the last layer of the device port temperature timing pattern feature extractor based on the convolutional neural network model is the device port real-time temperature timing association pattern feature map, and the input of the first layer of the device port temperature timing pattern feature extractor based on the convolutional neural network model is the device port real-time temperature timing conversion image; A temperature timing correlation pattern feature noise reduction module, used for passing the device port real-time temperature timing correlation pattern feature graph through a noise reducer based on a position mask module to obtain a device port real-time temperature timing correlation pattern feature graph after noise reduction; A receptive field focusing module, used for passing the noise-reduced device port real-time temperature timing correlation mode characteristic diagram through the receptive field focusing module to obtain a local manifestation device port real-time temperature timing correlation mode characteristic; A temperature mode abnormality detection and warning module is used to determine whether there is an abnormality in the temperature mode of the device port based on the real-time temperature time series correlation mode characteristics of the local manifestation device port, and determine whether to generate a warning information prompt; Wherein, the receptive field focusing module includes: A temperature timing correlation pattern local manifestation unit, used for passing the noise-reduced device port real-time temperature timing correlation pattern feature graph through the receptive field focusing module to obtain a local manifestation device port real-time temperature timing correlation pattern feature vector as the local manifestation device port real-time temperature timing correlation pattern feature; The temperature timing correlation mode local manifestation unit is used to: pass the noise-reduced device port real-time temperature timing correlation mode feature map through the receptive field focusing module to obtain a local manifestation device port real-time temperature timing correlation mode feature vector as the local manifestation device port real-time temperature timing correlation mode feature according to the following formula; Wherein, the formula is: ; in, It is global mean pooling; () is the activation function, represents one-dimensional convolution, represents a three-dimensional convolution, is a characteristic diagram of the real-time temperature timing correlation mode of the device port after noise reduction, is the temporal correlation pattern feature vector of the real-time temperature of the local manifestation device port; Wherein, the temperature mode abnormality detection and early warning module includes: The device port temperature mode abnormality detection unit is used to pass the local manifested device port real-time temperature time series correlation mode feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether the temperature mode of the device port is abnormal; An early warning prompt unit, configured to generate an early warning message prompt in response to the classification result indicating that the temperature mode of the device port is abnormal; Wherein, the device port temperature mode abnormality detection unit includes: A fully connected encoding subunit, used for performing fully connected encoding on the local manifestation device port real-time temperature timing association pattern feature vector using multiple fully connected layers of the classifier to obtain an encoded classification feature vector; and A classification result generating subunit, used for passing the encoded classification feature vector through the Softmax classification function of the classifier to obtain the classification result; The device port temperature mode abnormality detection unit includes: calculating the sum of the square root of the length of the local manifestation device port real-time temperature timing association mode feature vector and the reciprocal of the square root of the second norm of the local manifestation device port real-time temperature timing association mode feature vector to obtain the local manifestation device port real-time temperature timing association mode offset feature vector; calculating the exponential function of the local manifestation device port real-time temperature timing association mode offset feature vector with a natural constant as the base to obtain the local manifestation device port real-time temperature timing association mode offset prediction feature vector; calculating the local manifestation device port real-time temperature timing association mode offset feature vector; calculating the local manifestation device port real-time temperature timing association mode offset prediction ... The local manifestation device port real-time temperature timing association mode feature vector is obtained by multiplying the dot product of the norm and weight hyperparameter of the local manifestation device port real-time temperature timing association mode feature vector with the local manifestation device port real-time temperature timing association mode feature vector; the local manifestation device port real-time temperature timing association mode offset prediction feature vector and the local manifestation device port real-time temperature timing association mode constraint feature vector are calculated to obtain the optimized local manifestation device port real-time temperature timing association mode feature vector; the optimized local manifestation device port real-time temperature timing association mode feature vector is passed through the classifier to obtain the classification result; The structured norm representation of the local manifested device port real-time temperature timing association pattern feature vector is used as the local canonical coordinate for each eigenvalue of the local manifested device port real-time temperature timing association pattern feature vector, the vector overall distribution representation of the local manifested device port real-time temperature timing association pattern feature vector is determined relative to the eigenvalue class rotation offset for the class offset prediction direction of each eigenvalue of the local manifested device port real-time temperature timing association pattern feature vector as the center, and the bounding box of the vector eigenvalue distribution of the local manifested device port real-time temperature timing association pattern feature vector is used to constrain the eigenvalue, so as to improve the constraint of the local manifested device port real-time temperature timing association pattern feature vector under the overall regression distribution.

2. The device port security alarm system according to claim 1, characterized in that: The temperature timing correlation pattern feature denoising module is used to: process the device port real-time temperature timing correlation pattern feature graph through a denoiser based on a position mask module using the following denoising formula to obtain the device port real-time temperature timing correlation pattern feature graph after denoising; Wherein, the noise reduction formula is: ; ; in, is a characteristic diagram of the real-time temperature timing correlation mode of the device port, is the threshold feature map, is a characteristic diagram of the real-time temperature timing correlation mode of the device port after noise reduction, is the characteristic value of each position in the characteristic diagram of the real-time temperature timing correlation mode of the device port, is the feature value of each position in the threshold feature map, and is a hyperparameter, Represents the condition set of the threshold function.