CN117061322A - Internet of things flow pool management method and system - Google Patents
Internet of things flow pool management method and system Download PDFInfo
- Publication number
- CN117061322A CN117061322A CN202311262965.3A CN202311262965A CN117061322A CN 117061322 A CN117061322 A CN 117061322A CN 202311262965 A CN202311262965 A CN 202311262965A CN 117061322 A CN117061322 A CN 117061322A
- Authority
- CN
- China
- Prior art keywords
- flow
- time sequence
- pools
- sub
- pool
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007726 management method Methods 0.000 title claims description 31
- 238000000034 method Methods 0.000 claims abstract description 42
- 238000004458 analytical method Methods 0.000 claims abstract description 20
- 239000011159 matrix material Substances 0.000 claims description 125
- 239000013598 vector Substances 0.000 claims description 79
- 238000012549 training Methods 0.000 claims description 61
- 238000003062 neural network model Methods 0.000 claims description 17
- 238000012300 Sequence Analysis Methods 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 7
- 238000012546 transfer Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 abstract description 14
- 230000008859 change Effects 0.000 abstract description 12
- 238000012544 monitoring process Methods 0.000 abstract description 5
- 238000012098 association analyses Methods 0.000 abstract description 4
- 238000009826 distribution Methods 0.000 description 11
- 238000013528 artificial neural network Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 10
- 230000000694 effects Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 230000001960 triggered effect Effects 0.000 description 5
- 210000004027 cell Anatomy 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 239000003795 chemical substances by application Substances 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000000547 structure data Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000003012 network analysis Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
- H04L41/064—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis involving time analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/12—Discovery or management of network topologies
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The application discloses a method and a system for managing flow pools of the Internet of things, which are used for carrying out flow time sequence change trend analysis of each flow pool and topology association analysis among the flow pools by monitoring the flow value of each flow pool in real time and introducing a data processing and analyzing algorithm at the rear end, so as to carry out early warning of the flow pools.
Description
Technical Field
The application relates to the field of the Internet of things, and in particular relates to a method and a system for managing a flow pool of the Internet of things.
Background
The internet of things (Internet of Things, ioT) refers to connecting various physical devices together through the internet, enabling communication and data exchange between the devices. With the popularization of the internet of things equipment and the increase of application scenes, a large amount of data is generated and transmitted, wherein the data comprise flow data of the internet of things equipment. In order to effectively manage and monitor traffic usage of the internet of things device, a traffic pool management method becomes an important technology.
The flow pool refers to a collection of summarizing and managing flow resources of a plurality of internet of things devices, and can be used for limiting and distributing the flow resources. In flow pool management, a total threshold is typically set, and when the amount of flow in the flow pool exceeds the threshold, an alarm needs to be triggered. In addition, the flow pool can be further divided into a plurality of sub-flow pools, a sub-threshold value is set for each sub-flow pool, and when the flow usage of the sub-flow pool exceeds the corresponding sub-threshold value, an alarm is required to be triggered.
However, conventional schemes typically use thresholds to trigger early warning of the flow pool. However, since the threshold is set statically, it is not flexible to accommodate changes in flow usage. When the flow is rapidly increased, the threshold value may not be timely adjusted, resulting in early warning hysteresis, which makes it impossible for a manager to take measures to cope with overload or abnormal conditions of the flow in time, and affects timely monitoring and processing effects on the use condition of the flow pool. In addition, conventional internet of things traffic pool management schemes typically manage each traffic pool independently, lacking analysis of associations between traffic pools. However, in the internet of things, there may be an association relationship between different traffic pools, such as communication between devices, data exchange, and the like. Ignoring such associations may result in less comprehensive and accurate policies for management and optimization of the traffic pool.
Accordingly, an optimized internet of things traffic pool management scheme is desired.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides a method and a system for managing flow pools of the Internet of things, which are used for carrying out flow time sequence change trend analysis of each flow pool and topology association analysis among the flow pools by monitoring the flow value of each flow pool in real time and introducing a data processing and analyzing algorithm at the rear end, so as to carry out early warning of the flow pools.
According to one aspect of the present application, there is provided a method for managing an internet of things traffic pool, including:
acquiring flow values of a plurality of sub-flow pools in the flow pool at a plurality of preset time points in a preset time period;
carrying out time sequence analysis on the flow values of the plurality of preset time points to obtain flow time sequence feature vectors of a plurality of sub-flow pools;
performing inter-flow pool relevance topology analysis on the plurality of sub-flow pools to obtain a flow pool adjacent topology feature matrix;
Performing association coding based on a graph structure on the flow time sequence feature vectors of the plurality of sub-flow pools and the flow pool adjacent topology feature matrix to obtain adjacent topology flow global time sequence features; and
and determining whether to generate an early warning prompt or not based on the global time sequence characteristic of the adjacent topology traffic.
According to another aspect of the present application, there is provided an internet of things traffic pool management system, comprising:
the data acquisition module is used for acquiring flow values of a plurality of sub-flow pools in the flow pool at a plurality of preset time points in a preset time period;
the time sequence analysis module is used for performing time sequence analysis on the flow values of the plurality of preset time points to obtain flow time sequence feature vectors of a plurality of sub-flow pools;
the topology analysis module is used for carrying out association topology analysis among the flow pools on the plurality of sub-flow pools so as to obtain a flow pool adjacent topology feature matrix;
the association coding module is used for carrying out association coding based on a graph structure on the flow time sequence feature vectors of the plurality of sub-flow pools and the flow pool adjacent topology feature matrix so as to obtain adjacent topology flow global time sequence features; and
and the prompt result generation module is used for determining whether to generate an early warning prompt or not based on the global time sequence characteristics of the adjacent topology traffic.
Compared with the prior art, the flow pool management method and system of the Internet of things provided by the application have the advantages that the flow value of each flow pool is monitored in real time, and the flow time sequence change trend analysis of each flow pool and the topology association analysis among the flow pools are carried out by introducing the data processing and analysis algorithm at the rear end, so that the early warning of the flow pools is carried out, the hysteresis caused by the traditional threshold detection can be avoided, the flow time sequence change analysis among the flow pools can be carried out more accurately, the early warning of the flow pools can be carried out timely and effectively, and the efficiency and the effect of the flow pool management are improved.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
FIG. 1 is a flow chart of a method for managing an Internet of things traffic pool according to an embodiment of the application;
Fig. 2 is a system architecture diagram of a method for managing an internet of things traffic pool according to an embodiment of the present application;
FIG. 3 is a flow chart of a training phase of an Internet of things traffic pool management method according to an embodiment of the application;
FIG. 4 is a flowchart of sub-step S2 of the method for managing an Internet of things traffic pool according to an embodiment of the present application;
FIG. 5 is a flowchart of sub-step S3 of the method for managing an Internet of things traffic pool according to an embodiment of the present application;
fig. 6 is a block diagram of an internet of things traffic pool management system according to an embodiment of the present application.
Detailed Description
Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
As used in the specification and in the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Although the present application makes various references to certain modules in a system according to embodiments of the present application, any number of different modules may be used and run on a user terminal and/or server. The modules are merely illustrative, and different aspects of the systems and methods may use different modules.
A flowchart is used in the present application to describe the operations performed by a system according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.
Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Conventional schemes typically use thresholds to trigger pre-warning of the flow pool. However, since the threshold is set statically, it is not flexible to accommodate changes in flow usage. When the flow is rapidly increased, the threshold value may not be timely adjusted, resulting in early warning hysteresis, which makes it impossible for a manager to take measures to cope with overload or abnormal conditions of the flow in time, and affects timely monitoring and processing effects on the use condition of the flow pool. In addition, conventional internet of things traffic pool management schemes typically manage each traffic pool independently, lacking analysis of associations between traffic pools. However, in the internet of things, there may be an association relationship between different traffic pools, such as communication between devices, data exchange, and the like. Ignoring such associations may result in less comprehensive and accurate policies for management and optimization of the traffic pool. Accordingly, an optimized internet of things traffic pool management scheme is desired.
In the technical scheme of the application, the method for managing the flow pool of the Internet of things is provided. Fig. 1 is a flowchart of a method for managing an internet of things traffic pool according to an embodiment of the present application. Fig. 2 is a system architecture diagram of a method for managing an internet of things traffic pool according to an embodiment of the present application. As shown in fig. 1 and 2, the method for managing the flow pool of the internet of things according to the embodiment of the application comprises the following steps: s1, acquiring flow values of a plurality of sub-flow pools in a flow pool at a plurality of preset time points in a preset time period; s2, carrying out time sequence analysis on the flow values of the plurality of preset time points to obtain flow time sequence feature vectors of a plurality of sub-flow pools; s3, carrying out correlation topology analysis among the flow pools on the plurality of sub-flow pools to obtain a flow pool adjacent topology feature matrix; s4, carrying out association coding based on a graph structure on the flow time sequence feature vectors of the plurality of sub-flow pools and the adjacent topology feature matrix of the flow pool to obtain an adjacent topology flow global time sequence feature; and S5, determining whether an early warning prompt is generated or not based on the global time sequence characteristic of the adjacent topology traffic.
In particular, the S1 obtains flow values of a plurality of sub-flow cells in the flow cell at a plurality of predetermined time points within a predetermined period of time. The flow pool refers to a collection for summarizing and managing flow resources of a plurality of internet of things devices, and can be used for limiting and distributing the flow resources. In flow pool management, a total threshold is typically set, and when the amount of flow in the flow pool exceeds the threshold, an alarm needs to be triggered. In addition, the flow pool can be further divided into a plurality of sub-flow pools, a sub-threshold value is set for each sub-flow pool, and when the flow usage of the sub-flow pool exceeds the corresponding sub-threshold value, an alarm is required to be triggered. Thus, in one specific example of the present application, first, flow values of a plurality of sub-flow cells in a flow cell at a plurality of predetermined time points within a predetermined period of time are acquired.
Specifically, the step S2 is to perform a time sequence analysis on the flow values at the predetermined time points to obtain flow time sequence feature vectors of the sub-flow pools. In particular, in one specific example of the present application, as shown in fig. 4, the S2 includes: s21, arranging flow values of a plurality of preset time points of the plurality of sub-flow pools in a preset time period into flow time sequence input vectors of the plurality of sub-flow pools according to a time dimension; and S22, respectively carrying out feature extraction on flow time sequence input vectors of the plurality of sub-flow pools through a time sequence feature extractor based on a deep neural network model so as to obtain flow time sequence feature vectors of the plurality of sub-flow pools.
Specifically, the step S21 is to arrange the flow values of the plurality of sub-flow pools at a plurality of predetermined time points within a predetermined time period into flow timing input vectors of the plurality of sub-flow pools according to a time dimension. The flow values of the sub-flow pools not only have a time sequence dynamic change rule in the time dimension, but also have a time sequence association relation. Therefore, in the technical scheme of the application, the flow values of the plurality of sub-flow pools at a plurality of preset time points in a preset time period are arranged into flow time sequence input vectors of the plurality of sub-flow pools according to a time dimension, so that the distribution information of the flow values of the sub-flow pools on time sequence is integrated respectively.
Accordingly, in one possible implementation, the flow values of the plurality of sub-flow pools at a plurality of predetermined time points within a predetermined time period may be arranged into flow timing input vectors of the plurality of sub-flow pools according to a time dimension, for example: the method comprises the steps of obtaining flow values of a plurality of sub-flow pools at a plurality of predetermined time points within a predetermined time period, obtaining flow value data according to the steps, creating an empty flow time sequence input vector list, and for each predetermined time point: for each sub-traffic pool: the method comprises the steps of adding a flow value of a time point into a flow time sequence input vector, adding the flow time sequence input vector of each sub-flow pool into a flow time sequence input vector list, and returning to the flow time sequence input vector list of a plurality of sub-flow pools.
Specifically, the step S22 is to perform feature extraction on flow time sequence input vectors of the plurality of sub-flow pools through a time sequence feature extractor based on a deep neural network model, so as to obtain flow time sequence feature vectors of the plurality of sub-flow pools. In particular, in one specific example of the present application, the flow time sequence input vectors of the plurality of sub-flow pools are subjected to feature mining in a time sequence feature extractor based on a one-dimensional convolution layer, so as to extract time sequence dynamic change feature information of flow values of the respective sub-flow pools in a time dimension, thereby obtaining flow time sequence feature vectors of the plurality of sub-flow pools. More specifically, each layer using the one-dimensional convolution layer based timing feature extractor performs, in forward transfer of the layer, respectively, on input data: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature images based on a feature matrix to obtain pooled feature images; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the time sequence feature extractor based on the one-dimensional convolution layer is the flow time sequence feature vectors of the plurality of sub-flow pools, and the input of the first layer of the time sequence feature extractor based on the one-dimensional convolution layer is the flow time sequence input vectors of the plurality of sub-flow pools.
Notably, a one-dimensional convolutional layer is a neural network layer in deep learning for processing data having a temporal structure. The method is mainly applied to feature extraction and pattern recognition of one-dimensional data (such as time sequence data, signal data and text data). The core idea of the one-dimensional convolution layer is to carry out convolution operation on a local area of input data in a sliding window mode, and extract local features. This sliding window, called a convolution kernel (Filter) or Filter, derives a set of weights through learning for feature extraction of the input data. One-dimensional convolutional layers are typically used in combination with other types of layers (e.g., pooled layers, fully connected layers) to form convolutional neural networks. Convolutional neural networks have achieved significant success in many areas, such as natural language processing, speech recognition, image processing, and the like. By utilizing the locality of convolution operation and the characteristic of weight sharing, the method effectively processes and extracts relevant features in time sequence data.
It should be noted that, in other specific examples of the present application, the flow values at the plurality of predetermined time points may also be subjected to time sequence analysis in other manners to obtain flow time sequence feature vectors of a plurality of sub-flow pools, for example: acquiring flow values of a plurality of sub-flow pools at a plurality of preset time points in a preset time period, and acquiring flow value data according to the steps; for each sub-traffic pool: creating an empty flow timing feature vector; flow value for each predetermined point in time: adding the flow value to the flow timing feature vector; flow timing feature vector for each sub-flow pool: calculating statistical features such as mean, maximum, minimum, standard deviation, etc.; calculating timing characteristics, such as trending, periodicity, seasonal, etc.; and returning flow time sequence characteristic vectors of the plurality of sub-flow pools.
Specifically, S3, performing a topology analysis of correlation between flow pools on the plurality of sub-flow pools to obtain a topology feature matrix of adjacent flow pools. In particular, in one specific example of the present application, as shown in fig. 5, the S3 includes: s31, constructing a flow Chi Linjie matrix of the plurality of sub-flow pools, wherein the value of each position on the non-diagonal line in the flow pool adjacent matrix represents whether an association exists between the corresponding two sub-flow pools; and S32, the flow pool adjacent matrix passes through a flow pool associated feature extractor based on a two-dimensional convolution layer to obtain the flow pool adjacent topological feature matrix.
Specifically, the S31 builds a traffic Chi Linjie matrix of the plurality of sub-traffic pools, where a value of each position on a non-diagonal line in the traffic pool adjacency matrix indicates whether there is an association between the corresponding two sub-traffic pools. It should be appreciated that in the internet of things, there may be associated actions of communication, data exchange, etc. between different devices or sensors, which may result in the transmission and sharing of traffic between different sub-traffic pools. Therefore, in order to capture the association relationship between different sub-flow pools, in the technical scheme of the application, the flow pool adjacency matrix of the plurality of sub-flow pools is further constructed. In particular, by constructing the traffic Chi Linjie matrix, the association between the individual sub-traffic pools can be represented as off-diagonal elements of the matrix. In particular, the value of each position on the non-diagonal in the traffic pool adjacency matrix may indicate whether there is an association between the corresponding two sub-traffic pools, such as whether there is data transfer, communication, etc. If there is an association between the two sub-traffic pools, the value of the corresponding location may be set to 1; if there is no association between the two sub-traffic pools, the value of the corresponding location may be set to 0.
Accordingly, in one possible implementation, the traffic Chi Linjie matrix of the plurality of sub-traffic pools may be constructed by, for example: according to specific scenes and requirements, determining the relevance among sub-flow pools; an empty adjacency matrix is created of size (N, N), where N represents the number of sub-traffic pools. Initially, all elements of the adjacency matrix are 0, indicating no association; and setting the element of the corresponding position of the associated sub-flow pool in the adjacency matrix to be 1 according to the relevance among the sub-flow pools. If a concept of weight or association strength exists, the element of the corresponding position may be set to a corresponding value; through the above steps, a traffic Chi Linjie matrix is obtained, which reflects the correlation between sub-traffic pools. Each element of the adjacency matrix represents the strength of association or connection state between sub-traffic pools.
Specifically, the S32 passes the flow pool adjacency matrix through a two-dimensional convolution layer-based flow pool associated feature extractor to obtain the flow pool adjacency topology feature matrix. It is considered that the traffic Chi Linjie matrix is a matrix representing an association relationship between sub-traffic pools, wherein element values represent whether an association exists between two of the traffic pool adjacency matrices. However, directly using the adjacency matrix as input may not fully utilize the association information because the adjacency matrix is a sparse representation and does not take into account the strength and pattern of the association. Therefore, in order to extract features with more characterizations and usability from the flow pool adjacency matrix, in the technical scheme of the application, the flow pool adjacency matrix is further subjected to feature mining in a flow pool associated feature extractor based on a two-dimensional convolution layer so as to extract adjacency topology associated feature information among all the flow pools, thereby obtaining a flow pool adjacency topology feature matrix. The traffic pool adjacency topology feature matrix thus obtained may contain more information about the associations between sub-traffic pools, such as association strength, association pattern, etc. More specifically, each layer using the two-dimensional convolution layer-based flow pool associated feature extractor performs, in forward transfer of the layer, on input data: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature map along a channel dimension to obtain a pooled feature map; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the two-dimensional convolution layer-based flow pool associated feature extractor is the flow pool adjacent topological feature matrix, and the input of the first layer of the two-dimensional convolution layer-based flow pool associated feature extractor is the flow pool adjacent matrix.
It is noted that the two-dimensional convolution layer is a neural network layer commonly used in deep learning, and is mainly used for processing two-dimensional data, such as image data. It extracts local features of the input data by applying a set of learnable filters (also called convolution kernels). In a two-dimensional convolution layer, both the input data and the filter are two-dimensional. The filter is typically a small two-dimensional matrix that produces an output profile by sliding over the input data and performing element-by-element multiplication and summation operations. This sliding process is known as a convolution operation. By stacking multiple two-dimensional convolution layers, features of different layers of input data can be extracted layer by layer, thereby achieving high-level abstraction and representation of the input data. Two-dimensional convolution layers are widely used in computer vision tasks such as image classification, object detection, and image segmentation.
It should be noted that, in other specific examples of the present application, the inter-flow pool correlation topology analysis may be performed on the multiple sub-flow pools in other manners to obtain a flow pool adjacency topology feature matrix, for example: and determining the relevance among the sub-flow pools according to specific scenes and requirements. The associations may be defined based on physical location, network topology, business relationships, and the like. For example, if there is a significant interplay or dependency of traffic between sub-traffic pools, it may be considered that there is an association between them; and constructing a flow pool adjacency matrix according to the relevance among the sub-flow pools. The adjacency matrix is a two-dimensional matrix used for representing the association relationship between sub-flow pools. The size of the matrix is (N, N), where N represents the number of sub-traffic pools. The elements of the adjacency matrix represent the association strength or connection weight between sub-traffic pools; and filling the association strength or the connection weight into the adjacency matrix according to the specific association definition. The filling mode can be binary (with the association being 1 and the no association being 0) or real (representing the magnitude of the association strength or weight); to ensure comparability and consistency of the adjacency matrix, the adjacency matrix may be normalized. Common normalization methods include dividing the elements of each row by the sum of the rows, or dividing the elements of each column by the sum of the columns; through the above steps, a traffic Chi Linjie matrix is obtained, which reflects the correlation between sub-traffic pools. This adjacency matrix can be regarded as an adjacency topology feature matrix of the traffic pools for describing the topology between sub-traffic pools.
Specifically, the step S4 is to perform association coding based on a graph structure on the traffic timing characteristic vectors of the plurality of sub-traffic pools and the traffic pool adjacent topology characteristic matrix to obtain an adjacent topology traffic global timing characteristic. That is, the flow time sequence feature vector of each sub-flow pool is used as the feature representation of the node, the adjacent topological feature matrix of the flow pool is used as the feature representation of the edge between the nodes, and the flow time sequence feature matrix of the sub-flow pool and the adjacent topological feature matrix of the flow pool obtained by two-dimensional arrangement of the flow time sequence feature vectors of the sub-flow pools are used for obtaining the adjacent topological flow global time sequence feature matrix through a graph neural network model. Specifically, the graph neural network model performs graph structure data coding on the flow time sequence feature matrix of the sub-flow pool and the adjacent topology feature matrix of the flow pool through the learnable neural network parameters to obtain the adjacent topology flow global time sequence feature matrix containing irregular flow Chi Linjie topological association features and flow time sequence change feature information of each sub-flow pool.
Notably, the graph neural network (Graph Neural Network, GNN) is a type of neural network model for processing graph structure data. Unlike conventional neural networks that primarily handle data in vector or matrix form, GNN is dedicated to modeling and learning nodes and edges in the graph. The core idea of the graph neural network is to learn the representation of the nodes through local information transfer between the nodes. It enables each node to aggregate and integrate the information of its neighboring nodes by iteratively updating the representation vector of each node. Thus, each node's representation vector contains its own characteristics as well as the characteristics of its associated neighboring nodes. The graph neural network has an advantage in that graph data having a complex topology can be processed and association relations between nodes can be considered. The method is widely applied to the fields of social network analysis, recommendation systems, molecular chemistry, computer vision and the like.
In particular, the step S5 is to determine whether to generate an early warning prompt based on the global timing characteristic of the adjacent topology traffic. In a specific example of the application, the global time sequence feature matrix of the adjacent topology traffic is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether an early warning prompt is generated or not. In other words, the classification processing is performed by utilizing the association characteristic information based on the graph structure between the adjacent topology association characteristic of each sub-flow pool and the flow time sequence change characteristic of each sub-flow pool, so as to perform early warning judgment of the flow pool. Specifically, the adjacent topological traffic global time sequence feature matrix is unfolded to be a classification feature vector based on a row vector or a column vector; performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
A Classifier (Classifier) refers to a machine learning model or algorithm that is used to classify input data into different categories or labels. The classifier is part of supervised learning, which performs classification tasks by learning mappings from input data to output categories.
The fully connected layer (Fully Connected Layer) is one type of layer commonly found in neural networks. In the fully connected layer, each neuron is connected to all neurons of the upper layer, and each connection has a weight. This means that each neuron in the fully connected layer receives inputs from all neurons in the upper layer, and weights these inputs together, and then passes the result to the next layer.
The Softmax classification function is a commonly used activation function for multi-classification problems. It converts each element of the input vector into a probability value between 0 and 1, and the sum of these probability values equals 1. The Softmax function is commonly used at the output layer of a neural network, and is particularly suited for multi-classification problems, because it can map the network output into probability distributions for individual classes. During the training process, the output of the Softmax function may be used to calculate the loss function and update the network parameters through a back propagation algorithm. Notably, the output of the Softmax function does not change the relative magnitude relationship between elements, but rather normalizes them. Thus, the Softmax function does not change the characteristics of the input vector, but simply converts it into a probability distribution form.
It should be noted that, in other specific examples of the present application, it may also be determined whether to generate the early warning prompt based on the global timing characteristics of the adjacent topology traffic in other manners, for example: collecting data related to adjacency topology traffic, including traffic values for each sub-traffic pool at a plurality of predetermined points in time over a predetermined period of time; and preprocessing the collected data, including data cleaning, missing value processing, outlier detection, denoising and the like. The quality and the accuracy of the data are ensured; global timing characteristics are extracted based on the adjacency topology traffic data. These characteristics may include total flow, average flow, flow volatility, flow trend, etc. These features may be extracted using statistical methods, time series analysis methods, or deep learning methods; and defining an early warning rule according to actual requirements and service scenes. The pre-warning rules may be based on specific thresholds or models. For example, if the average flow of a certain sub-flow pool exceeds a preset threshold value or the flow fluctuation exceeds a certain range, triggering early warning; and judging the extracted global time sequence characteristics according to the early warning rules, and determining whether an early warning prompt is generated. If the characteristic value meets the condition of the early warning rule, namely exceeds a preset threshold value or the model is judged to be abnormal, triggering early warning; once the pre-warning is triggered, relevant personnel or systems need to be notified in time. The notification mode can be sending alarm information, mail, short message or displaying through a system interface; and (5) feeding back and adjusting the condition of triggering the early warning. The reasons of the early warning can be analyzed, the early warning rule is optimized or the threshold value is adjusted, so that the accuracy and the reliability of the early warning are improved.
It should be appreciated that the one-dimensional convolutional layer-based timing feature extractor, the two-dimensional convolutional layer-based flow pool correlation feature extractor, the graph neural network model, and the classifier need to be trained prior to inference using the neural network model described above. That is, the method for managing the flow pool of the internet of things further comprises a training stage, wherein the training stage is used for training the time sequence feature extractor based on the one-dimensional convolution layer, the flow pool associated feature extractor based on the two-dimensional convolution layer, the graph neural network model and the classifier.
Fig. 3 is a flowchart of a training phase of the method for managing an internet of things traffic pool according to an embodiment of the present application. As shown in fig. 3, the method for managing the flow pool of the internet of things according to the embodiment of the application comprises the following steps: a training phase comprising: s110, training data are acquired, wherein the training data comprise training flow values of a plurality of sub-flow pools in the flow pools at a plurality of preset time points in a preset time period, and whether a real value of an early warning prompt is generated or not; s120, training flow values of a plurality of preset time points of the plurality of sub-flow pools in a preset time period are arranged into flow time sequence input vectors of the plurality of training sub-flow pools according to a time dimension; s130, enabling flow time sequence input vectors of the training sub-flow pools to pass through the time sequence feature extractor based on the one-dimensional convolution layer to obtain flow time sequence feature vectors of the training sub-flow pools; s140, constructing a training flow Chi Linjie matrix of the plurality of sub-flow pools, wherein the values of each position on the non-diagonal line in the training flow pool adjacent matrix represent whether an association exists between the corresponding two sub-flow pools; s150, passing the training flow Chi Linjie matrix through the two-dimensional convolution layer-based flow pool correlation feature extractor to obtain a training flow pool adjacent topological feature matrix; s160, the flow time sequence feature vectors of the training sub-flow pools and the training flow pool adjacent topological feature matrix are processed through the graph neural network model to obtain a training adjacent topological flow global time sequence feature matrix; s170, passing the training adjacent topology flow global time sequence feature matrix through the classifier to obtain a classification loss function value; and S180, training the time sequence feature extractor based on the one-dimensional convolution layer, the flow pool associated feature extractor based on the two-dimensional convolution layer, the graph neural network model and the classifier based on the classification loss function value and through gradient descent direction propagation, wherein the training adjacent topological flow global time sequence feature vector obtained after the training adjacent topological flow global time sequence feature matrix is unfolded is subjected to weight space exploration constraint optimization based on regularization of the class matrix when the training weight matrix iterates each time.
In particular, in the technical scheme of the application, the flow time sequence feature vector of each sub-flow pool represents the time sequence association feature of the flow value of the corresponding sub-flow pool, so that after the flow time sequence feature vector and the adjacent topological feature matrix of the flow pool pass through a graph neural network model, the topological association based on the spatial distribution topology of the adjacent relation of the flow value can be further carried out, that is, the adjacent topological flow global time sequence feature matrix simultaneously comprises the feature representation of the diversified time-space association dimension corresponding to the flow value of each sub-flow pool under the time domain space and space association topology, and therefore, when the flow value association feature representation of the adjacent topological flow global time sequence feature matrix under the time-space dimension is promoted, the label distribution enrichment corresponding to the feature distribution diversity of the time-space association dimension in the probability distribution domain of the classification result is also caused when the adjacent topological flow global time sequence feature matrix carries out classification regression through the classifier, and the convergence effect of the weight matrix of the classifier is influenced in the classification process. Based on the above, when classifying the adjacent topology traffic global time sequence feature matrix by a classifier, the applicant performs weight space exploration constraint based on regularization of a class matrix on the adjacent topology traffic global time sequence feature vector obtained after the adjacent topology traffic global time sequence feature matrix is expanded at each iteration of the weight matrix, specifically expressed as:
，
Wherein the method comprises the steps ofIs the training adjacent topology flow global time sequence feature vector obtained after the training adjacent topology flow global time sequence feature matrix is unfolded,/I>Is the global time sequence feature vector of the optimized training adjacent topology traffic obtained after the global time sequence feature matrix of the optimized training adjacent topology traffic is developed, and +.>For column vector, +.>Is a row vector, +.>Is a domain transfer matrix which can be learned, +.>Weight matrix representing last iteration, +.>Representing the weight matrix after iteration, +.>Representing a matrix multiplication. Here, the global timing feature vector of the adjacent topology traffic is considered in terms of the weight spatial domain of the weight matrix +.>Domain differences (domain gap) between probability distribution domains of classification results of (2) by weight matrix +.>Global timing feature vector with respect to said adjacency topology traffic>The regularized representation of the class matrix of (2) is used as an inter-domain migration agent (inter-domain transferring agent) to transfer the probability distribution of valuable label constraint into a weight space, so that excessive exploration (over-explloit) of the weight distribution in the weight space by a rich labeled probability distribution domain in the classification process based on the weight space is avoided, the convergence effect of the weight matrix is improved, and the training effect of the adjacent topological traffic global time sequence feature matrix in classification regression through a classifier is also improved. Therefore, the early warning of the flow pool can be more accurately carried out based on the flow time sequence change trend of the flow pool and the association relation between the flow pools, and the efficiency and the effect of the management of the flow pool are improved 。
In summary, the method for managing the flow pools of the internet of things according to the embodiment of the application is explained, by monitoring the flow value of each flow pool in real time and introducing a data processing and analyzing algorithm at the rear end to perform flow time sequence change trend analysis of each flow pool and topology association analysis among each flow pool, thereby performing early warning of the flow pool.
Further, an Internet of things flow pool management system is also provided.
Fig. 6 is a block diagram of an internet of things traffic pool management system according to an embodiment of the present application. As shown in fig. 6, the flow pool management system 300 of the internet of things according to an embodiment of the present application includes: a data acquisition module 310, configured to acquire flow values of a plurality of sub-flow pools in the flow pool at a plurality of predetermined time points within a predetermined time period; the time sequence analysis module 320 is configured to perform time sequence analysis on the flow values at the plurality of predetermined time points to obtain flow time sequence feature vectors of the plurality of sub-flow pools; the topology analysis module 330 is configured to perform association topology analysis between the flow pools on the plurality of sub-flow pools to obtain a flow pool adjacent topology feature matrix; the association coding module 340 is configured to perform association coding based on a graph structure on the traffic timing feature vectors of the multiple sub-traffic pools and the traffic pool adjacent topology feature matrix to obtain an adjacent topology traffic global timing feature; and a prompt result generating module 350, configured to determine whether to generate an early warning prompt based on the global timing characteristic of the adjacent topology traffic.
As described above, the internet of things traffic pool management system 300 according to the embodiment of the present application may be implemented in various wireless terminals, for example, a server having an internet of things traffic pool management algorithm, etc. In one possible implementation, the internet of things traffic pool management system 300 according to an embodiment of the present application may be integrated into a wireless terminal as one software module and/or hardware module. For example, the internet of things traffic pool management system 300 may be a software module in the operating system of the wireless terminal, or may be an application developed for the wireless terminal; of course, the flow pool management system 300 of the internet of things can also be one of a plurality of hardware modules of the wireless terminal.
Alternatively, in another example, the internet of things traffic pool management system 300 and the wireless terminal may be separate devices, and the internet of things traffic pool management system 300 may connect to the wireless terminal through a wired and/or wireless network and transmit interaction information in a agreed data format.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (10)
1. The method for managing the flow pool of the Internet of things is characterized by comprising the following steps of:
acquiring flow values of a plurality of sub-flow pools in the flow pool at a plurality of preset time points in a preset time period;
carrying out time sequence analysis on the flow values of the plurality of preset time points to obtain flow time sequence feature vectors of a plurality of sub-flow pools;
performing inter-flow pool relevance topology analysis on the plurality of sub-flow pools to obtain a flow pool adjacent topology feature matrix;
performing association coding based on a graph structure on the flow time sequence feature vectors of the plurality of sub-flow pools and the flow pool adjacent topology feature matrix to obtain adjacent topology flow global time sequence features; and
and determining whether to generate an early warning prompt or not based on the global time sequence characteristic of the adjacent topology traffic.
2. The method for managing the traffic pool of the internet of things according to claim 1, wherein performing a time sequence analysis on the traffic values at the plurality of predetermined time points to obtain traffic time sequence feature vectors of the plurality of sub-traffic pools comprises:
arranging flow values of a plurality of preset time points of the plurality of sub-flow pools in a preset time period into flow time sequence input vectors of the plurality of sub-flow pools according to a time dimension; and
And respectively carrying out feature extraction on flow time sequence input vectors of the plurality of sub-flow pools by a time sequence feature extractor based on a deep neural network model so as to obtain flow time sequence feature vectors of the plurality of sub-flow pools.
3. The internet of things traffic pool management method according to claim 2, wherein the deep neural network model-based timing feature extractor is a one-dimensional convolutional layer-based timing feature extractor.
4. The method for managing flow pools of the internet of things according to claim 3, wherein performing a topology analysis of relevance among flow pools on the plurality of sub-flow pools to obtain a topology feature matrix of adjacency of the flow pools, comprises:
constructing a flow Chi Linjie matrix of the plurality of sub-flow pools, wherein values of respective positions on non-diagonals in the flow pool adjacency matrix represent whether an association exists between the corresponding two sub-flow pools;
and the flow pool adjacent matrix passes through a flow pool associated feature extractor based on a two-dimensional convolution layer to obtain the flow pool adjacent topological feature matrix.
5. The method for managing the traffic pool of the internet of things according to claim 4, wherein performing graph structure-based association coding on the traffic timing feature vectors of the plurality of sub-traffic pools and the traffic pool adjacency topology feature matrix to obtain adjacency topology traffic global timing features, comprises: and the flow time sequence feature vectors of the plurality of sub-flow pools and the adjacent topological feature matrix of the flow pool are processed through a graph neural network model to obtain an adjacent topological flow global time sequence feature matrix serving as the adjacent topological flow global time sequence feature.
6. The method of claim 5, wherein determining whether to generate an early warning hint based on the adjacency topology traffic global timing feature comprises: and the adjacent topological flow global time sequence feature matrix passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether an early warning prompt is generated or not.
7. The method for traffic pool management in the internet of things according to claim 6, further comprising the training step of: the method is used for training the time sequence feature extractor based on the one-dimensional convolution layer, the flow pool associated feature extractor based on the two-dimensional convolution layer, the graph neural network model and the classifier.
8. The method for traffic pool management in the internet of things according to claim 7, wherein the training step comprises:
acquiring training data, wherein the training data comprises training flow values of a plurality of sub-flow pools in a flow pool at a plurality of preset time points in a preset time period, and whether a real value of an early warning prompt is generated or not;
arranging training flow values of a plurality of preset time points of the plurality of sub-flow pools in a preset time period into flow time sequence input vectors of the plurality of training sub-flow pools according to a time dimension;
The flow time sequence input vectors of the training sub-flow pools pass through the time sequence feature extractor based on the one-dimensional convolution layer to obtain flow time sequence feature vectors of the training sub-flow pools;
constructing a training flow Chi Linjie matrix of the plurality of sub-flow pools, wherein values of positions on non-diagonals in the training flow pool adjacency matrix represent whether an association exists between the corresponding two sub-flow pools;
passing the training traffic Chi Linjie matrix through the two-dimensional convolution layer-based traffic pool correlation feature extractor to obtain a training traffic pool adjacent topological feature matrix;
the flow time sequence feature vectors of the training sub-flow pools and the adjacent topological feature matrix of the training flow pools pass through the graph neural network model to obtain a global time sequence feature matrix of the training adjacent topological flow;
the training adjacent topological flow global time sequence feature matrix passes through the classifier to obtain a classification loss function value;
training the time sequence feature extractor based on the one-dimensional convolution layer, the flow pool associated feature extractor based on the two-dimensional convolution layer, the graph neural network model and the classifier based on the classification loss function value and through gradient descent direction propagation, wherein the training adjacent topological flow global time sequence feature vector obtained after the training adjacent topological flow global time sequence feature matrix is unfolded is subjected to weight space exploration constraint optimization based on regularization of a class matrix when each weight matrix of the training is iterated.
9. The method for managing the flow pool of the internet of things according to claim 8, wherein, when the weight matrix of the training is iterated each time, the training adjacent topology flow global time sequence feature vector obtained after the training adjacent topology flow global time sequence feature matrix is expanded is subjected to weight space exploration constraint optimization based on class matrix regularization by using the following optimization formula to obtain an optimized training adjacent topology flow global time sequence feature matrix;
wherein, the optimization formula is:
，
wherein the method comprises the steps ofIs the training adjacent topology flow global time sequence feature vector obtained after the training adjacent topology flow global time sequence feature matrix is unfolded,/I>Is the global time sequence feature vector of the optimized training adjacent topology traffic obtained after the global time sequence feature matrix of the optimized training adjacent topology traffic is developed, and +.>For column vector, +.>Is a row vector, +.>Is a domain transfer matrix which can be learned, +.>Weight matrix representing last iteration, +.>Representing the weight matrix after iteration, +.>Representing a matrix multiplication.
10. An internet of things traffic pool management system, comprising:
the data acquisition module is used for acquiring flow values of a plurality of sub-flow pools in the flow pool at a plurality of preset time points in a preset time period;
The time sequence analysis module is used for performing time sequence analysis on the flow values of the plurality of preset time points to obtain flow time sequence feature vectors of a plurality of sub-flow pools;
the topology analysis module is used for carrying out association topology analysis among the flow pools on the plurality of sub-flow pools so as to obtain a flow pool adjacent topology feature matrix;
the association coding module is used for carrying out association coding based on a graph structure on the flow time sequence feature vectors of the plurality of sub-flow pools and the flow pool adjacent topology feature matrix so as to obtain adjacent topology flow global time sequence features; and
and the prompt result generation module is used for determining whether to generate an early warning prompt or not based on the global time sequence characteristics of the adjacent topology traffic.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311262965.3A CN117061322A (en) | 2023-09-27 | 2023-09-27 | Internet of things flow pool management method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311262965.3A CN117061322A (en) | 2023-09-27 | 2023-09-27 | Internet of things flow pool management method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117061322A true CN117061322A (en) | 2023-11-14 |
Family
ID=88657490
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311262965.3A Pending CN117061322A (en) | 2023-09-27 | 2023-09-27 | Internet of things flow pool management method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117061322A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117319090A (en) * | 2023-11-28 | 2023-12-29 | 江苏云网数智信息技术有限公司 | Intelligent network safety protection system |
| CN117454301A (en) * | 2023-12-22 | 2024-01-26 | 深圳大学 | Method and system for displaying change of PV value tracked by nuclear fusion test |
| CN117530684A (en) * | 2024-01-09 | 2024-02-09 | 深圳市双佳医疗科技有限公司 | Blood glucose abnormality detection and early warning system and method based on health big data |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108650218A (en) * | 2018-03-22 | 2018-10-12 | 平安科技（深圳）有限公司 | Network Traffic Monitoring method, apparatus, computer equipment and storage medium |
| US20190133480A1 (en) * | 2017-11-08 | 2019-05-09 | Koninklijke Philips N.V. | Discretized embeddings of physiological waveforms |
| CN109889999A (en) * | 2018-12-29 | 2019-06-14 | 深圳市中天网景科技有限公司 | A kind of flow cell management method, system and the terminal of Internet of Things network interface card |
| CN109905274A (en) * | 2019-01-17 | 2019-06-18 | 虫极科技（北京）有限公司 | A kind of Intelligent cargo cabinet and its Network Warning method |
| CN114077811A (en) * | 2022-01-19 | 2022-02-22 | 华东交通大学 | Electric power Internet of things equipment abnormality detection method based on graph neural network |
| CN114389966A (en) * | 2022-03-24 | 2022-04-22 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | Network traffic identification method and system based on graph neural network and stream space-time correlation |
| CN114647198A (en) * | 2022-03-09 | 2022-06-21 | 深圳市经纬纵横科技有限公司 | Intelligent home control method and system based on Internet of things and electronic equipment |
| CN114724386A (en) * | 2022-03-31 | 2022-07-08 | 温州大学 | Short-time traffic flow prediction method and system under intelligent traffic and electronic equipment |
| CN115834433A (en) * | 2023-02-17 | 2023-03-21 | 杭州沄涞科技有限公司 | Data processing method and system based on Internet of things technology |
| CN115994321A (en) * | 2021-10-15 | 2023-04-21 | 腾讯科技（深圳）有限公司 | Object classification method and related device |
| CN116032845A (en) * | 2023-02-13 | 2023-04-28 | 杭银消费金融股份有限公司 | Data center network overhead management method and system |
| CN116647411A (en) * | 2023-07-17 | 2023-08-25 | 厦门巴掌互动科技有限公司 | Game platform network security monitoring and early warning method |
| CN116639010A (en) * | 2023-07-24 | 2023-08-25 | 江西驴宝宝通卡科技有限公司 | Intelligent control system and method for charging pile |
| CN116699297A (en) * | 2023-08-08 | 2023-09-05 | 江西驴充充充电技术有限公司 | Charging pile detection system and method thereof |
| CN116760772A (en) * | 2023-08-21 | 2023-09-15 | 北京中科网芯科技有限公司 | Control system and method for converging flow divider |
-
2023
- 2023-09-27 CN CN202311262965.3A patent/CN117061322A/en active Pending
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190133480A1 (en) * | 2017-11-08 | 2019-05-09 | Koninklijke Philips N.V. | Discretized embeddings of physiological waveforms |
| CN108650218A (en) * | 2018-03-22 | 2018-10-12 | 平安科技（深圳）有限公司 | Network Traffic Monitoring method, apparatus, computer equipment and storage medium |
| CN109889999A (en) * | 2018-12-29 | 2019-06-14 | 深圳市中天网景科技有限公司 | A kind of flow cell management method, system and the terminal of Internet of Things network interface card |
| CN109905274A (en) * | 2019-01-17 | 2019-06-18 | 虫极科技（北京）有限公司 | A kind of Intelligent cargo cabinet and its Network Warning method |
| CN115994321A (en) * | 2021-10-15 | 2023-04-21 | 腾讯科技（深圳）有限公司 | Object classification method and related device |
| CN114077811A (en) * | 2022-01-19 | 2022-02-22 | 华东交通大学 | Electric power Internet of things equipment abnormality detection method based on graph neural network |
| CN114647198A (en) * | 2022-03-09 | 2022-06-21 | 深圳市经纬纵横科技有限公司 | Intelligent home control method and system based on Internet of things and electronic equipment |
| CN114389966A (en) * | 2022-03-24 | 2022-04-22 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | Network traffic identification method and system based on graph neural network and stream space-time correlation |
| CN114724386A (en) * | 2022-03-31 | 2022-07-08 | 温州大学 | Short-time traffic flow prediction method and system under intelligent traffic and electronic equipment |
| CN116032845A (en) * | 2023-02-13 | 2023-04-28 | 杭银消费金融股份有限公司 | Data center network overhead management method and system |
| CN115834433A (en) * | 2023-02-17 | 2023-03-21 | 杭州沄涞科技有限公司 | Data processing method and system based on Internet of things technology |
| CN116647411A (en) * | 2023-07-17 | 2023-08-25 | 厦门巴掌互动科技有限公司 | Game platform network security monitoring and early warning method |
| CN116639010A (en) * | 2023-07-24 | 2023-08-25 | 江西驴宝宝通卡科技有限公司 | Intelligent control system and method for charging pile |
| CN116699297A (en) * | 2023-08-08 | 2023-09-05 | 江西驴充充充电技术有限公司 | Charging pile detection system and method thereof |
| CN116760772A (en) * | 2023-08-21 | 2023-09-15 | 北京中科网芯科技有限公司 | Control system and method for converging flow divider |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117319090A (en) * | 2023-11-28 | 2023-12-29 | 江苏云网数智信息技术有限公司 | Intelligent network safety protection system |
| CN117454301A (en) * | 2023-12-22 | 2024-01-26 | 深圳大学 | Method and system for displaying change of PV value tracked by nuclear fusion test |
| CN117454301B (en) * | 2023-12-22 | 2024-04-16 | 深圳大学 | Method and system for displaying change of PV value tracked by nuclear fusion test |
| CN117530684A (en) * | 2024-01-09 | 2024-02-09 | 深圳市双佳医疗科技有限公司 | Blood glucose abnormality detection and early warning system and method based on health big data |
| CN117530684B (en) * | 2024-01-09 | 2024-04-16 | 深圳市双佳医疗科技有限公司 | Blood glucose abnormality detection and early warning system and method based on health big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11521221B2 (en) | Predictive modeling with entity representations computed from neural network models simultaneously trained on multiple tasks | |
| CN117061322A (en) | Internet of things flow pool management method and system | |
| CN112700056A (en) | Complex network link prediction method, complex network link prediction device, electronic equipment and medium | |
| CN114004328A (en) | AI model updating method, device, computing equipment and storage medium | |
| CN110321493B (en) | Abnormity detection and optimization method and system of social network and computer equipment | |
| CN117156442B (en) | Cloud data security protection method and system based on 5G network | |
| CN117041017B (en) | Intelligent operation and maintenance management method and system for data center | |
| JP6971514B1 (en) | Information processing equipment, information processing methods and programs | |
| CN112085281B (en) | Method and device for detecting safety of business prediction model | |
| CN116662817B (en) | Asset identification method and system of Internet of things equipment | |
| CN115099326A (en) | Behavior prediction method, behavior prediction device, behavior prediction equipment and storage medium based on artificial intelligence | |
| Qian et al. | Short-term demand forecasting for on-demand mobility service | |
| CN115861246A (en) | Product quality abnormity detection method and system applied to industrial Internet | |
| CN114880482A (en) | Graph embedding-based relation graph key personnel analysis method and system | |
| CN113343123A (en) | Training method and detection method for generating confrontation multiple relation graph network | |
| Shrivastava et al. | Selection of efficient and accurate prediction algorithm for employing real time 5G data load prediction | |
| CN115952438B (en) | Social platform user attribute prediction method, system, mobile device and storage medium | |
| CN111898249A (en) | Landslide displacement nonparametric probability density prediction method, equipment and storage medium | |
| CN114757391B (en) | Network data space design and application method oriented to service quality prediction | |
| CN114842247B (en) | Characteristic accumulation-based graph convolution network semi-supervised node classification method | |
| CN116029760A (en) | Message pushing method, device, computer equipment and storage medium | |
| CN113837481B (en) | Financial big data management system based on block chain | |
| CN115510948A (en) | Block chain fishing detection method based on robust graph classification | |
| CN115019342A (en) | Endangered animal target detection method based on class relation reasoning | |
| Shen et al. | Long-term multivariate time series forecasting in data centers based on multi-factor separation evolutionary spatial–temporal graph neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |