CN117955746A - Data circulation method, device, equipment and storage medium based on Internet of things platform - Google Patents
Data circulation method, device, equipment and storage medium based on Internet of things platform Download PDFInfo
- Publication number
- CN117955746A CN117955746A CN202410348418.5A CN202410348418A CN117955746A CN 117955746 A CN117955746 A CN 117955746A CN 202410348418 A CN202410348418 A CN 202410348418A CN 117955746 A CN117955746 A CN 117955746A
- Authority
- CN
- China
- Prior art keywords
- data
- target
- circulation
- data flow
- internet
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 74
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 230000003993 interaction Effects 0.000 claims abstract description 86
- 238000010586 diagram Methods 0.000 claims abstract description 69
- 238000004458 analytical method Methods 0.000 claims abstract description 64
- 238000012545 processing Methods 0.000 claims abstract description 51
- 238000012546 transfer Methods 0.000 claims abstract description 50
- 238000012544 monitoring process Methods 0.000 claims abstract description 18
- 230000005856 abnormality Effects 0.000 claims description 66
- 238000005457 optimization Methods 0.000 claims description 29
- 238000009826 distribution Methods 0.000 claims description 25
- 230000007246 mechanism Effects 0.000 claims description 19
- 230000002159 abnormal effect Effects 0.000 claims description 18
- 239000013598 vector Substances 0.000 claims description 18
- 238000000605 extraction Methods 0.000 claims description 17
- 230000011218 segmentation Effects 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000005096 rolling process Methods 0.000 claims description 13
- 238000013528 artificial neural network Methods 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 238000013496 data integrity verification Methods 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 4
- 230000010354 integration Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000011156 evaluation Methods 0.000 claims description 2
- 230000006855 networking Effects 0.000 claims 4
- 238000004891 communication Methods 0.000 claims 1
- 230000005540 biological transmission Effects 0.000 description 32
- 230000008569 process Effects 0.000 description 24
- 230000035945 sensitivity Effects 0.000 description 8
- 238000001514 detection method Methods 0.000 description 5
- 230000006399 behavior Effects 0.000 description 3
- 125000004122 cyclic group Chemical group 0.000 description 3
- 230000009545 invasion Effects 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000004451 qualitative analysis Methods 0.000 description 2
- 238000004445 quantitative analysis Methods 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The application relates to the technical field of data processing of the Internet of things, and discloses a data circulation method, a device, equipment and a storage medium based on an Internet of things platform. The method comprises the following steps: acquiring a plurality of distributed data centers of an Internet of things platform and creating a data circulation target network; receiving an initial data stream set and classifying to obtain a target data stream of each target Internet of things device; matching the optimal path of the data stream to obtain a target data stream path; performing data transfer and transfer state monitoring to obtain a transfer state interaction diagram; extracting local and global circulation state characteristics to obtain a circulation state characteristic set; according to the data transfer anomaly analysis method, the data transfer anomaly analysis is carried out through the data transfer anomaly analysis model, the data transfer anomaly analysis result is obtained, and the corresponding data transfer anomaly processing scheme is matched according to the data transfer anomaly analysis result.
Description
Technical Field
The application relates to the technical field of data processing of the internet of things, in particular to a data transfer method, a device, equipment and a storage medium based on an internet of things platform.
Background
In the current internet of things (IoT) era, tremendous data traffic has been generated with the rapid growth of internet of things devices and the widespread spread of applications. These data streams contain not only a large amount of real-time data collected from the sensors, but also video, audio and other multimedia information. The information needs to be processed and transmitted efficiently and safely on the internet of things platform. However, due to the distributed nature and resource limitations of the internet of things environment, how to ensure efficient and safe flow of data streams between multiple distributed data centers has become an important research issue. Especially when facing large-scale internet of things application, optimization of data flow efficiency and security is particularly critical.
On the one hand, the internet of things platform has to handle massive data streams from widely distributed devices, which have a high degree of dynamics and diversity. It is a challenge to effectively sort, encrypt, and route these data streams to ensure that they can be accurately processed and analyzed in a timely manner while meeting real-time and security requirements. Existing data processing and streaming methods often fail to fully meet these requirements, especially in terms of handling complex data stream classification, encryption, and optimized routing, and lack sufficient flexibility and adaptation. On the other hand, with the increase of data traffic and the continuous progress of attack technology, data security problems are increasingly prominent. How to ensure the security and privacy of data and prevent data leakage and tampering while ensuring the data transfer efficiency is another important problem that needs to be solved by the internet of things platform. In addition, the detection and processing of data flow anomalies on the internet of things platform also face a great challenge, and the traditional anomaly detection method is difficult to adapt to the complexity and the dynamics of the internet of things environment.
Disclosure of Invention
The application provides a data transfer method, a device, equipment and a storage medium based on an internet of things platform, which are used for improving the accuracy of data transfer by combining the internet of things platform with a distributed data transfer technology.
In a first aspect, the present application provides a data transfer method based on an internet of things platform, where the data transfer method based on the internet of things platform includes:
acquiring a plurality of distributed data centers of an Internet of things platform, and creating a data flow target network of each distributed data center based on a multi-target optimization algorithm;
Receiving initial data stream sets of a plurality of target internet of things devices through the internet of things platform, and carrying out data stream classification and multi-level data encryption on the initial data stream sets to obtain target data streams of each target internet of things device;
performing data flow optimal path matching on the target data flow of each target internet of things device based on the data flow target network to obtain a target data flow path of each target data flow;
performing data circulation and circulation state monitoring on the target data flows through the target data circulation paths to obtain circulation state interaction diagrams of each target data flow;
local and global circulation state feature extraction is respectively carried out on the circulation state interaction graph of each target data stream through a node segmentation method and a graph rolling network with a multi-head attention mechanism, so that a circulation state feature set of each circulation state interaction graph is obtained;
inputting the circulation state feature set of each circulation state interaction diagram into a preset data circulation abnormality analysis model to perform data circulation abnormality analysis, obtaining a data circulation abnormality analysis result, and matching a corresponding data circulation abnormality processing scheme according to the data circulation abnormality analysis result.
In a second aspect, the present application provides a data transfer device based on an internet of things platform, where the data transfer device based on the internet of things platform includes:
the acquisition module is used for acquiring a plurality of distributed data centers of the Internet of things platform and creating a data flow target network of each distributed data center based on a multi-target optimization algorithm;
the classification module is used for receiving initial data flow sets of a plurality of target internet of things devices through the internet of things platform, and carrying out data flow classification and multi-level data encryption on the initial data flow sets to obtain target data flows of each target internet of things device;
The matching module is used for carrying out data flow optimal path matching on the target data flow of each target internet of things device based on the data flow target network to obtain a target data flow path of each target data flow;
The monitoring module is used for carrying out data flow and flow state monitoring on the target data flow through the target data flow path to obtain a flow state interaction diagram of each target data flow;
The feature extraction module is used for extracting local and global circulation state features of the circulation state interaction graph of each target data stream through a node segmentation method and a graph rolling network with a multi-head attention mechanism to obtain a circulation state feature set of each circulation state interaction graph;
The analysis module is used for inputting the circulation state feature set of each circulation state interaction diagram into a preset data circulation abnormality analysis model to perform data circulation abnormality analysis, obtaining a data circulation abnormality analysis result, and matching a corresponding data circulation abnormality processing scheme according to the data circulation abnormality analysis result.
The third aspect of the present application provides a data circulation device based on an internet of things platform, comprising: a memory and at least one processor, the memory having instructions stored therein; and the at least one processor calls the instruction in the memory so that the data flow equipment based on the Internet of things platform executes the data flow method based on the Internet of things platform.
A fourth aspect of the present application provides a computer readable storage medium having instructions stored therein, which when run on a computer, cause the computer to perform the above-described data transfer method based on an internet of things platform.
According to the technical scheme provided by the application, the data flow path can be dynamically optimized according to the actual capacity of the data center and the data flow requirement by creating the data flow target network for each distributed data center by adopting the multi-target optimization algorithm. This not only minimizes the delay of data transmission, but also effectively increases the speed and efficiency of data processing. In a mass data environment generated by the Internet of things equipment, rapid processing and response of real-time data can be ensured, and the method is very important for supporting key applications such as real-time monitoring and remote control. And the initial data stream set is subjected to multi-level data encryption processing, so that strong safety guarantee is provided for data transmission in the Internet of things platform. The multi-level encryption ensures that even if the data is intercepted in the transmission process, the data is difficult to be decrypted by an unauthorized third party, and the problems of data leakage and privacy invasion are effectively prevented. In addition, by classifying the data flow and matching the encryption intensity, the encryption strategy can be adjusted according to the sensitivity of the data, so that the data security is further enhanced. The node segmentation method and the graph rolling network with the multi-head attention mechanism are adopted to analyze the states of the data flow, and the abnormal states possibly occurring in the data flow process can be intelligently identified and predicted by combining a preset data flow abnormal analysis model. Once an abnormality is detected, the corresponding abnormality processing scheme can be automatically matched, and the problem can be responded and solved in time, so that the stable operation and the service continuity of the Internet of things system are ensured. The intelligent abnormality detection and processing mechanism not only improves the autonomy and the robustness of the system, but also reduces the requirement of manual intervention and maintenance cost.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained based on these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of an embodiment of a data circulation method based on an internet of things platform according to an embodiment of the present application;
fig. 2 is a schematic diagram of an embodiment of a data circulation device based on an internet of things platform in an embodiment of the present application.
Detailed Description
The embodiment of the application provides a data circulation method, a device, equipment and a storage medium based on an Internet of things platform. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
For easy understanding, the following describes a specific flow of an embodiment of the present application, referring to fig. 1, and an embodiment of a data circulation method based on an internet of things platform in the embodiment of the present application includes:
Step S101, acquiring a plurality of distributed data centers of an Internet of things platform, and creating a data flow target network of each distributed data center based on a multi-target optimization algorithm;
It can be understood that the execution body of the application may be a data transfer device based on an internet of things platform, and may also be a terminal or a server, which is not limited herein. The embodiment of the application is described by taking a server as an execution main body as an example.
Specifically, a plurality of distributed data centers of the internet of things platform are acquired, and the data centers are distributed at different geographic positions and are responsible for processing mass data from the internet of things equipment. And carrying out feasible path calculation on each distributed data center by adopting a multi-objective optimization algorithm. A set of possible paths for each data center is defined and calculated, wherein the set of possible paths is determined by evaluating multiple objective functions under different paths, ensuring that optimizations in path selection can be evaluated and selected from multiple dimensions. Based on the set of feasible paths, an initial network of data flows is built for each distributed data center, and the network serves as an infrastructure of the data flows. In order to further optimize the data flow efficiency, the data flow initial network of each distributed data center is subjected to flow distribution optimization through a network flow distribution model. The decision variables and unit data flow costs are calculated to minimize data flow costs while ensuring that each node is able to handle the amount of data allocated to it. And according to the network flow distribution strategy, carrying out network variable weight optimization on the data flow initial network of each distributed data center. And adjusting the weight of each connection in the network based on the flow distribution result to form a target network of data flow. The target network optimizes the data circulation path and also considers the dynamic change of the data flow, so that the data can be circulated between the distributed data centers efficiently and safely through the optimal path.
Step S102, receiving initial data stream sets of a plurality of target Internet of things devices through an Internet of things platform, and carrying out data stream classification and multi-level data encryption on the initial data stream sets to obtain target data streams of each target Internet of things device;
Specifically, the internet of things platform is used as a data aggregation point to receive an initial data stream set from each target internet of things device. Data is coming in from the source and contains various information such as environment monitoring data, user behavior data and the like. The initial set of data streams is data stream classified to determine how the data will be processed and forwarded. By classification, the data stream of each target internet of things device is identified and separated. And carrying out data hierarchy division on the initial data flow of each target Internet of things device, subdividing each initial data flow into a plurality of target hierarchy data flows, and the division is convenient for implementing different levels of data processing and protection measures aiming at different data sensitivity and application requirements. And matching corresponding data encryption intensity strategies according to the characteristics of each target level data stream. Different data streams are given different intensities of encryption processing according to their importance and sensitivity, thereby ensuring security during data stream. After the encryption process, each initial data stream is converted into a plurality of encrypted data streams. And carrying out data integrity verification on a plurality of encrypted data streams of each initial data stream, and ensuring that the integrity of the data in the encryption and transmission processes is not destroyed. The verification result will instruct whether to output the encrypted data stream as the target data stream to the target internet of things device.
Step S103, performing data flow optimal path matching on the target data flow of each target Internet of things device based on the data flow target network to obtain a target data flow path of each target data flow;
Specifically, data flow demand analysis is performed on the target data flow of each target internet of things device, so that characteristics and demands of each data flow, such as data volume, transmission speed requirements, security level and the like, are accurately grasped, and a data flow demand set of each target data flow is formed. And carrying out relation matching between the data stream and the distributed data center based on the data stream demand set. And the target data flow of each target internet of things device corresponds to the data flow target network of the distributed data center, so that each data flow can find a proper data processing and transmission environment, and the data flow target network corresponding to each target data flow is obtained. And traversing path nodes and analyzing data flow paths of the data flow target network to obtain a plurality of candidate data flow paths of each target data flow, wherein the candidate paths provide a plurality of possible options for final optimal path selection. Because various path dependencies and constraint conditions exist in the data flow process, such as network bandwidth limitation, data transmission delay, security requirements and the like, path dependency and constraint condition analysis is performed on each target data flow according to the requirement set of each target data flow so as to ensure that the finally selected path can meet all relevant requirements. Based on the path dependency and constraint conditions, a path optimization solving method is adopted to comprehensively evaluate and select candidate data transfer paths of each target data flow, and a target data transfer path meeting transmission efficiency and meeting safety requirements is determined for each target data flow.
Step S104, carrying out data circulation and circulation state monitoring on the target data flows through the target data circulation paths to obtain circulation state interaction diagrams of each target data flow;
Specifically, the target data stream is subjected to data stream through the target data stream path, and the stream state of the target data stream path is monitored, including monitoring a plurality of stream state characteristic data such as the speed, stability and safety of the data stream, so that each link of the data stream in the transmission process can reach the expected performance standard. And analyzing the captured feature data of each circulation state through a single-factor cloud model, and calculating the target digital feature of each feature data. The single-factor cloud model has unique advantages in terms of uncertainty and ambiguity handling, can effectively combine qualitative analysis with quantitative analysis, and provides an accurate numerical representation for each circulation state characteristic data. In order to facilitate subsequent processing and analysis, the target digital features are normalized to generate normalized digital features. The normalization process normalizes the data range so that efficient comparison and comprehensive analysis can be performed between different feature data. And acquiring the influence intensity sequence of the feature data based on the influence intensity of each circulation state feature data. The determination of the influence intensity reflects the importance of each characteristic data in the data circulation process. And carrying out serialization conversion on the normalized digital features according to the influence intensity sequence to generate a target feature sequence of each target data stream. The multidimensional characteristic data is converted into a one-dimensional sequence, so that subsequent path weight setting and analysis are facilitated. And setting path weight of the target data circulation path through the target feature sequence to generate a path weight distribution diagram. The weight distribution diagram intuitively reflects the importance and the priority of different path segments in data flow, and the data flow target network is subjected to directed interaction diagram conversion based on the path weight distribution diagram, so that the flow state interaction diagram of each target data flow is finally obtained.
Step S105, local and global circulation state feature extraction is carried out on the circulation state interaction diagram of each target data flow through a node segmentation method and a diagram rolling network with a multi-head attention mechanism, so as to obtain a circulation state feature set of each circulation state interaction diagram;
specifically, local feature extraction is performed on the circulation state interaction diagram of each target data flow by a node segmentation method, and the complex circulation state interaction diagram is decomposed into smaller parts, so that the features of the data circulation state, such as circulation speed, stability and the like, are revealed on a local level, and a plurality of local circulation state features of each circulation state interaction diagram are obtained. And carrying out global circulation state feature extraction on the circulation state interaction diagram of each target data flow through a diagram rolling network with a multi-head attention mechanism. The multi-head attention mechanism can learn information in parallel from different subspaces, enhance the capturing capability of the model on global features of the data circulation state, understand complex interactions among different nodes and obtain the global circulation state features of each circulation state interaction diagram. And carrying out mean value calculation on the local circulation state features, wherein the mean value reflects the average state of the local features, calculating standard deviation based on the local feature mean value, and the local standard deviation can reveal the discrete degree among the local features and provide a quantization index for understanding the diversity of the local circulation states. And similarly, calculating the mean value and standard deviation of the global features to obtain the general trend of the global features, and evaluating the variability among the global features. And integrating the local and global circulation state features by taking the local and global standard deviation as a weight, and constructing a comprehensive circulation state feature set by integrating the local and global view angles. The feature set integrates microscopic and macroscopic features of data flow, and also considers the difference and consistency between features.
And S106, inputting the circulation state feature set of each circulation state interaction diagram into a preset data circulation abnormality analysis model to perform data circulation abnormality analysis, obtaining a data circulation abnormality analysis result, and matching a corresponding data circulation abnormality processing scheme according to the data circulation abnormality analysis result.
Specifically, the circulation state feature set of each circulation state interaction diagram is input into a preset data circulation abnormality analysis model, and the model is composed of an input layer, a circulation neural network layer and two full-connection layers. And carrying out feature coding on the circulation state feature set of each circulation state interaction diagram through the input layer, and converting the complex feature information into circulation state feature coding vectors. And inputting the circulation state feature coding vector of each circulation state interaction diagram into the circulation neural network layer. The cyclic neural network (RNN) can effectively capture long-term dependency in time sequence data, and the model can deeply understand dynamic relations and change trends among circulation state features, so that a target state feature coding vector of each circulation state interaction diagram is obtained. And inputting the target state characteristic coding vector of each circulation state interaction diagram into two full-connection layers to conduct data circulation abnormal probability prediction. The full connection layer can process the feature coding vector, extract key information through nonlinear transformation and finally output the abnormal probability of data flow. The probability reflects the possibility of the current data flow state relative to the abnormal state, and provides a quantization basis for subsequent abnormal judgment. And comparing the data flow abnormality probability of each flow state interaction diagram with a preset target probability threshold value to judge which data flow states are likely to have abnormality, so as to output a data flow abnormality analysis result. And automatically matching the corresponding data flow abnormality processing scheme according to the data flow abnormality analysis result. These schemes may include data flow rerouting, increasing data redundancy, adjusting data transmission speeds, etc., with the aim of taking effective measures against predicted anomalies to ensure the stability and reliability of the data flow process.
In the embodiment of the application, the data flow path can be dynamically optimized according to the actual capability of the data center and the data flow requirement by adopting the multi-objective optimization algorithm to create the data flow target network for each distributed data center. This not only minimizes the delay of data transmission, but also effectively increases the speed and efficiency of data processing. In a mass data environment generated by the Internet of things equipment, rapid processing and response of real-time data can be ensured, and the method is very important for supporting key applications such as real-time monitoring and remote control. And the initial data stream set is subjected to multi-level data encryption processing, so that strong safety guarantee is provided for data transmission in the Internet of things platform. The multi-level encryption ensures that even if the data is intercepted in the transmission process, the data is difficult to be decrypted by an unauthorized third party, and the problems of data leakage and privacy invasion are effectively prevented. In addition, by classifying the data flow and matching the encryption intensity, the encryption strategy can be adjusted according to the sensitivity of the data, so that the data security is further enhanced. The node segmentation method and the graph rolling network with the multi-head attention mechanism are adopted to analyze the states of the data flow, and the abnormal states possibly occurring in the data flow process can be intelligently identified and predicted by combining a preset data flow abnormal analysis model. Once an abnormality is detected, the corresponding abnormality processing scheme can be automatically matched, and the problem can be responded and solved in time, so that the stable operation and the service continuity of the Internet of things system are ensured. The intelligent abnormality detection and processing mechanism not only improves the autonomy and the robustness of the system, but also reduces the requirement of manual intervention and maintenance cost.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
(1) Acquiring a plurality of distributed data centers of an Internet of things platform;
(2) The feasible path calculation is respectively carried out on each distributed data center based on a multi-objective optimization algorithm to obtain a feasible path set of each distributed data center, ，Is a set of feasible paths,/>Represents a viable path, wherein/>Is the current viable path,/>Is a further possible path which is a path,Is/>Individual objective function vs. feasible path/>Evaluation of/>Is a set of objective functions;
(3) Carrying out data flow network construction on the feasible path set of each distributed data center to obtain a data flow initial network of each distributed data center;
(4) Respectively carrying out network flow distribution optimization on the data flow conversion initial network of each distributed data center through a network flow distribution model to obtain a network flow distribution strategy, wherein the network flow distribution model comprises the following components: ，/> Representing slave node/> To node/>Per unit data streaming cost,/>Is a decision variable representing the slave node/>To node/>Data traffic of/>Node/>Total data volume to be processed,/>The total number of nodes and the total number of connections in the data flow initial network;
(5) And carrying out network variable weight optimization on the data flow initial network of each distributed data center according to the network flow distribution strategy to obtain the data flow target network of each distributed data center.
Specifically, a plurality of distributed data centers of the internet of things platform are obtained. The data centers form an infrastructure for processing and storing data of the Internet of things, are distributed in different geographic positions, and respectively bear the tasks of data collection, processing and forwarding. And carrying out feasible path calculation on each data center through a multi-objective optimization algorithm, and finding out an optimal path set of data flow under the condition that a plurality of constraint conditions are met. The algorithm calculates a series of feasible data flow paths by evaluating the network environment and the data processing capacity of each data center, and the paths not only need to consider the efficiency and the cost of data transmission, but also consider multidimensional indexes such as data safety, reliability and the like. For example, assume that there are three distributed data centers A, B and C, which differ in data transfer cost and processing power. Multiple possible paths between each pair of data centers from a through B, A through C, B through C, etc. are evaluated using a multi-objective optimization algorithm. For each path, the algorithm calculates its performance under different objective functions, such as time of transmission, cost and security. In this way, the algorithm can determine a set of feasible paths for each data center, each path in the set being an optimal solution under the constraint of a particular objective function. An initial network of data flows is constructed for each distributed data center based on the set of viable paths. The network reflects possible data flow paths between data centers and related attributes such as bandwidth, delay, cost, etc. of the paths. On the basis, the initial network is optimized through the network flow distribution model, so that the data transmission efficiency is ensured, and the overall cost is minimized. The model determines the data traffic on each path by a mathematical programming method so that the total cost is minimized while meeting the data processing requirements of each data center. In the calculation process, the model considers the unit data flow cost of each path, the data processing amount requirement of each data center and the total number of nodes and the total number of connections in the network, so as to obtain the optimal data flow distribution strategy on each path. And according to the obtained network flow distribution strategy, network weight change optimization is carried out on the data flow initial network of each distributed data center, and the data flow efficiency and the cost effectiveness are improved. The selection of the data flow paths is adjusted, the flow distribution on certain paths is changed or the selection of the data processing nodes is optimized so as to adapt to the change of the data flow demands, and therefore a data flow target network of each data center is formed.
In a specific embodiment, the process of executing step S102 may specifically include the following steps:
(1) Receiving an initial data stream set of a plurality of target internet of things devices through an internet of things platform;
(2) Classifying the initial data flow set to obtain an initial data flow of each target Internet of things device;
(3) Respectively carrying out data hierarchy division on the initial data stream of each target Internet of things device to obtain a plurality of target hierarchy data streams of each initial data stream;
(4) Respectively matching the data encryption intensity policies of each target level data stream, and carrying out data encryption processing on a plurality of target level data streams according to the data encryption intensity policies to obtain a plurality of encrypted data streams of each initial data stream;
(5) And respectively carrying out data integrity verification on a plurality of encrypted data streams of each initial data stream to obtain a data integrity verification result, and outputting a target data stream of each target Internet of things device according to the data integrity verification result.
Specifically, an initial data stream set of a plurality of target internet of things devices is received through an internet of things platform. The internet of things platform is used as a data convergence center and can receive data streams from different devices. The data streams are effectively classified and managed to ensure that the data can be properly processed and analyzed. Each type of data stream is hierarchically partitioned, and each initial data stream is subdivided into a plurality of target hierarchical data streams according to its importance, sensitivity or other relevant criteria. For example, the data stream generated by the video surveillance camera may be divided into different levels according to the sensitivity of the picture content (e.g., the monitored pictures of the public area and the private area), while the data stream of the temperature sensor may be divided according to the frequency and importance of the change of the readings. The data encryption intensity policy of each target level data stream is matched respectively, and an appropriate encryption algorithm and key intensity are selected for each level data stream to protect the data from unauthorized access. For example, for private area pictures in video surveillance, a higher intensity encryption policy, such as AES 256 bit encryption, may be required, while for low sensitivity data such as temperature readings, a lower intensity encryption policy may be employed to conserve computing resources. And carrying out data integrity verification on each encrypted data stream, ensuring that the data is not tampered or damaged in the encryption and transmission processes, and ensuring the authenticity and reliability of the data. For example, for an encrypted video surveillance data stream, its integrity is confirmed by computing and verifying a hash value of the data. If the hash value of the data is matched with the hash value calculated by the transmitting end, the data is proved not to be tampered in the transmission process, so that the encrypted data streams can be safely output to the corresponding processing modules or the storage systems as target data streams. And outputting the target data stream of each target Internet of things device according to the data integrity verification result.
In a specific embodiment, the process of executing step S103 may specifically include the following steps:
(1) Carrying out data flow demand analysis on the target data flow of each target internet of things device to obtain a data flow demand set of each target data flow;
(2) Performing relationship matching on the data flow target network of each distributed data center and the target data flow of each target internet of things device according to the data flow demand set to obtain a data flow target network corresponding to each target data flow;
(3) Traversing path nodes and analyzing data flow paths of each target data flow according to the data flow target network to obtain a plurality of candidate data flow paths of each target data flow;
(4) Analyzing path dependency and constraint conditions of each target data stream according to the data stream demand set to obtain the path dependency and constraint conditions;
(5) And carrying out path optimization solving on a plurality of candidate data flow paths of each target data flow according to the path dependency and constraint conditions to obtain the target data flow path of each target data flow.
Specifically, the requirement analysis is performed on the target data stream of each target internet of things device, and the characteristics and requirements of the data stream are understood. For example, a data stream of a temperature monitoring device may need to be transmitted in real time and be delay sensitive, while a status data stream of an intelligent door lock may have higher security requirements. By this analysis, a set of demand indicators, such as transmission speed, security level, data size, etc., is determined for each data stream, forming a data stream demand set. And according to the data flow demand set, performing relation matching on the data flow target network of each distributed data center and the target data flow of each target internet of things device, and considering factors such as the processing capacity of the data center, the bandwidth of the network, the priority of the data and the like. For example, for data streams with high real-time requirements, it is desirable to match a data center with a low latency network path; for data streams with high security requirements, a data center providing advanced data encryption and security protection needs to be selected. Through matching, each target data stream can be ensured to find a data stream target network meeting the requirements of the target data stream. And traversing path nodes and analyzing data flow paths of the data flow target network, and generating a plurality of candidate data flow paths for each target data flow. By examining the characteristics (e.g., delay, bandwidth, and cost) of the network connections and the various paths between data centers, multiple transmission options are provided for the data streams. And analyzing path dependency and constraint conditions of each target data flow according to the data flow demand set to determine which paths can meet the specific demands of the data flow, and considering the influence of potential problems such as network congestion, node faults and the like on data transmission. For example, for those data flows that are delay sensitive, those paths that may traverse congested network areas are excluded; for data streams that require extremely high data integrity, it is desirable to select paths that can provide end-to-end encryption. And carrying out optimization solution on the candidate data circulation paths of each target data flow according to the path dependency and constraint conditions, and determining the final target data circulation path. The path most suitable for each data flow requirement is calculated by algorithm by comprehensively considering various factors such as the efficiency, the cost, the safety, the reliability and the like of the path.
In a specific embodiment, the process of executing step S104 may specifically include the following steps:
(1) Performing data flow on the target data flow through a target data flow path, and performing flow state monitoring on the target data flow path to obtain a plurality of flow state characteristic data of each target data flow;
(2) Calculating target digital characteristics of each circulation state characteristic data through a single-factor cloud model respectively;
(3) Carrying out digital feature normalization on the target digital feature of each circulation state feature data to generate normalized digital features of each circulation state feature data;
(4) Acquiring the influence intensity sequences of a plurality of circulation state feature data, and carrying out serialization conversion on the normalized digital features based on the influence intensity sequences to generate a target feature sequence of each target power supply;
(5) Setting path weight of a target data circulation path through a target feature sequence to obtain a path weight distribution diagram;
(6) And performing directed interaction graph conversion on the data flow target network through the path weight distribution graph to obtain a flow state interaction graph of each target data flow.
Specifically, the target data stream is subjected to data stream through the target data stream path, and each link of data transmission such as transmission speed, delay, packet loss rate and the like is monitored. And analyzing the feature data of each circulation state through a single-factor cloud model, and calculating the target digital features of the feature data. The single factor cloud model is an effective tool capable of processing uncertainty information, and by combining qualitative analysis with quantitative analysis, a digital feature with statistical significance such as expectation, entropy and super entropy is generated for each circulation state feature. These digital features can comprehensively reflect the characteristics of the data stream state. And carrying out normalization processing on the target digital features of each circulation state feature data so as to eliminate the dimension influence among different feature data, so that the feature data are on the same level, and the comparison and analysis are convenient. And determining the influence intensity sequence of the plurality of circulation state characteristic data, and carrying out serialization conversion on the normalized digital characteristic based on the sequence. The order of impact strength is determined by analyzing the degree of impact of each feature on data flow efficiency and security in order to identify which features have the greatest impact on performance during data flow, thereby giving higher priority during optimization. And setting path weight of the target data flow path through the target feature sequence. The weights reflect the priority and importance of different paths and provide basis for decision making of data flow. For example, if the delay characteristics of a path are of greater impact, the weight of that path may be reduced to avoid data transmission through that path. And converting the weighted data flow information into a directed interaction graph. The directed interaction graph visually represents the individual paths of the data stream and the interactions between them, and also reflects the priorities of the different paths through path weights.
In a specific embodiment, the process of executing step S105 may specifically include the following steps:
(1) Extracting local circulation state characteristics of the circulation state interaction diagrams of each target data flow respectively through a node segmentation method to obtain a plurality of local circulation state characteristics of each circulation state interaction diagram;
(2) Global circulation state feature extraction is carried out on the circulation state interaction diagram of each target data flow through a diagram rolling network with a multi-head attention mechanism, so that a plurality of global circulation state features of each circulation state interaction diagram are obtained;
(3) Carrying out mean value calculation on the local circulation state features to obtain local feature mean values, and carrying out standard deviation calculation on the local circulation state features according to the local feature mean values to obtain local standard deviations;
(4) Calculating the average value of the plurality of global circulation state features to obtain a global feature average value, and calculating the standard deviation of the plurality of global circulation state features according to the global feature average value to obtain a global standard deviation;
(5) And carrying out feature integration on the plurality of local circulation state features and the plurality of global circulation state features according to the local standard deviation and the global standard deviation to obtain a circulation state feature set of each circulation state interaction diagram.
Specifically, the local circulation state feature extraction is carried out on the circulation state interaction diagram of each target data flow through the node segmentation method, and the local behaviors and state changes of each node in the data circulation process are identified and understood. The node segmentation method breaks down the complex flow state interaction graph into smaller, more manageable parts, thereby enabling the system to extract key local features for each part. For example, local features such as traffic flow, delay, and congestion at each intersection (node) are extracted by node segmentation of a flow state interaction graph of traffic flow data. Global flow state feature extraction is performed on the flow state interaction graph of each target data flow through a graph rolling network (GCN) with a multi-head attention mechanism. The GCN can capture complex relationships and dependencies between nodes, and the multi-head attention mechanism further enhances the recognition capability of the model for different node and path importance. For example, through GCN, traffic flow relationships between different intersections and congestion patterns of the entire urban traffic network are understood. And calculating the mean value and standard deviation of the local circulation state characteristics, and quantifying the concentration trend and the discrete degree of the local characteristics. The local feature mean reflects the average state of each node or region, while the local standard deviation reveals the variability of states between different nodes. Similarly, the global circulation state features are subjected to mean and standard deviation calculation to evaluate the overall state and consistency of the whole data circulation network. And integrating the features according to the local and global standard deviations to form a comprehensive circulation state feature set of each circulation state interaction diagram. This feature set includes both detailed understanding of individual nodes and local regions and fused insight into global behavior of the entire data flow network. After feature integration, this information is used to optimize the data flow path, for example, by increasing redundancy of data transmissions to reduce the impact of certain high variability nodes, or to adjust the path to avoid global congestion points.
In a specific embodiment, the process of executing step S106 may specifically include the following steps:
(1) Inputting the circulation state feature set of each circulation state interaction diagram into a preset data circulation abnormality analysis model, wherein the data circulation abnormality analysis model comprises: an input layer, a circulating neural network and two full-connection layers;
(2) The feature coding is carried out on the circulation state feature set of each circulation state interaction graph through the input layer, so that circulation state feature coding vectors of each circulation state interaction graph are obtained;
(3) Inputting the circulation state feature code vector of each circulation state interaction diagram into a circulation neural network to perform feature extraction to obtain a target state feature code vector of each circulation state interaction diagram;
(4) Inputting the target state feature coding vector of each circulation state interaction diagram into two full-connection layers to conduct data circulation abnormal probability prediction, and obtaining the data circulation abnormal probability of each circulation state interaction diagram;
(5) Comparing the data flow abnormality probability of each flow state interaction diagram with a target probability threshold value, and outputting a data flow abnormality analysis result;
(6) And matching the corresponding data flow exception handling scheme according to the data flow exception analysis result.
Specifically, the circulation state feature set of each circulation state interaction diagram is input into a preset data circulation abnormality analysis model. And respectively carrying out feature coding on the circulation state feature set of each circulation state interaction diagram through the input layer. In the internet of things system, the data flow state may include various characteristics such as transmission delay, packet loss rate, transmission rate, and the like of the data packet. By feature encoding, the features are converted into stream state feature encoding vectors, providing input data for subsequent analysis and prediction. And inputting the circulation state feature coding vector of each circulation state interaction diagram into a circulation neural network to perform feature extraction. The recurrent neural network is suitable for processing sequence data and can capture the dynamic characteristic of the change of the data circulation state along with time. Through the processing of the cyclic neural network, the model can extract the target state characteristic coding vector reflecting the dynamic state of the data flow from the coding vector. For example, if a data stream frequently encounters a delay within a specific time period, the recurrent neural network can identify this time-series pattern, providing a basis for the prediction of abnormal conditions. And inputting the target state characteristic coding vector into two fully-connected layers to predict the abnormal probability of the data flow. The full-connection layer can further process and analyze the features extracted by the cyclic neural network, and finally output the abnormal probability of data flow of each flow state interaction diagram. This probability value reflects the likelihood of an abnormality in the data streaming process in the current state. For example, if the probability of anomalies for a certain data flow is above a preset threshold, this may indicate a serious delay or data loss problem in the data flow process. The model compares the abnormal probability of the data flow of each flow state interaction diagram with a preset target probability threshold value to determine which data flow states are likely to have abnormality, so that a data flow abnormality analysis result is output. For example, setting the anomaly probability threshold value to 0.8, the model will identify potential anomaly states for those data flow states having an anomaly probability greater than 0.8. And matching and adopting a corresponding data flow abnormality processing scheme according to the data flow abnormality analysis result. These processing schemes may include adjusting data transmission paths, limiting data traffic, enhancing data encryption measures, etc., in order to solve the identified anomaly problem and ensure smooth and safe data flow. For example, for a data stream identified as being delay-anomalous, the system may choose to reroute the data stream through a path in the network that has a higher bandwidth to reduce transmission delay.
The data transfer method based on the internet of things platform in the embodiment of the present application is described above, and the data transfer device based on the internet of things platform in the embodiment of the present application is described below, referring to fig. 2, one embodiment of the data transfer device based on the internet of things platform in the embodiment of the present application includes:
An acquisition module 201, configured to acquire a plurality of distributed data centers of an internet of things platform, and create a data flow target network of each distributed data center based on a multi-target optimization algorithm;
The classification module 202 is configured to receive an initial data flow set of a plurality of target internet of things devices through the internet of things platform, and perform data flow classification and multi-level data encryption on the initial data flow set to obtain a target data flow of each target internet of things device;
the matching module 203 is configured to perform data flow optimal path matching on the target data flow of each target internet of things device based on the data flow target network, so as to obtain a target data flow path of each target data flow;
the monitoring module 204 is configured to perform data flow and flow state monitoring on the target data flows through the target data flow path, so as to obtain a flow state interaction diagram of each target data flow;
The feature extraction module 205 is configured to perform local and global feature extraction on the circulation state interaction graph of each target data stream through a node segmentation method and a graph convolution network with a multi-head attention mechanism, so as to obtain a circulation state feature set of each circulation state interaction graph;
The analysis module 206 is configured to input the circulation state feature set of each circulation state interaction diagram into a preset data circulation abnormality analysis model to perform data circulation abnormality analysis, obtain a data circulation abnormality analysis result, and match a corresponding data circulation abnormality processing scheme according to the data circulation abnormality analysis result.
By the cooperation of the above components, the data flow path can be dynamically optimized according to the actual capability of the data center and the data flow requirement by creating a data flow target network for each distributed data center by adopting a multi-target optimization algorithm. This not only minimizes the delay of data transmission, but also effectively increases the speed and efficiency of data processing. In a mass data environment generated by the Internet of things equipment, rapid processing and response of real-time data can be ensured, and the method is very important for supporting key applications such as real-time monitoring and remote control. And the initial data stream set is subjected to multi-level data encryption processing, so that strong safety guarantee is provided for data transmission in the Internet of things platform. The multi-level encryption ensures that even if the data is intercepted in the transmission process, the data is difficult to be decrypted by an unauthorized third party, and the problems of data leakage and privacy invasion are effectively prevented. In addition, by classifying the data flow and matching the encryption intensity, the encryption strategy can be adjusted according to the sensitivity of the data, so that the data security is further enhanced. The node segmentation method and the graph rolling network with the multi-head attention mechanism are adopted to analyze the states of the data flow, and the abnormal states possibly occurring in the data flow process can be intelligently identified and predicted by combining a preset data flow abnormal analysis model. Once an abnormality is detected, the corresponding abnormality processing scheme can be automatically matched, and the problem can be responded and solved in time, so that the stable operation and the service continuity of the Internet of things system are ensured. The intelligent abnormality detection and processing mechanism not only improves the autonomy and the robustness of the system, but also reduces the requirement of manual intervention and maintenance cost.
The application also provides a data transfer device based on the internet of things platform, which comprises a memory and a processor, wherein the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the processor executes the steps of the data transfer method based on the internet of things platform in the above embodiments.
The application also provides a computer readable storage medium, which can be a nonvolatile computer readable storage medium, and can also be a volatile computer readable storage medium, wherein the computer readable storage medium stores instructions, and when the instructions run on a computer, the instructions cause the computer to execute the steps of the data circulation method based on the internet of things platform.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, systems and units may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random acceS memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (10)
1. The data transfer method based on the Internet of things platform is characterized by comprising the following steps of:
acquiring a plurality of distributed data centers of an Internet of things platform, and creating a data flow target network of each distributed data center based on a multi-target optimization algorithm;
Receiving initial data stream sets of a plurality of target internet of things devices through the internet of things platform, and carrying out data stream classification and multi-level data encryption on the initial data stream sets to obtain target data streams of each target internet of things device;
performing data flow optimal path matching on the target data flow of each target internet of things device based on the data flow target network to obtain a target data flow path of each target data flow;
performing data circulation and circulation state monitoring on the target data flows through the target data circulation paths to obtain circulation state interaction diagrams of each target data flow;
local and global circulation state feature extraction is respectively carried out on the circulation state interaction graph of each target data stream through a node segmentation method and a graph rolling network with a multi-head attention mechanism, so that a circulation state feature set of each circulation state interaction graph is obtained;
inputting the circulation state feature set of each circulation state interaction diagram into a preset data circulation abnormality analysis model to perform data circulation abnormality analysis, obtaining a data circulation abnormality analysis result, and matching a corresponding data circulation abnormality processing scheme according to the data circulation abnormality analysis result.
2. The data circulation method based on the internet of things platform according to claim 1, wherein the acquiring the plurality of distributed data centers of the internet of things platform and creating the data circulation target network of each distributed data center based on the multi-target optimization algorithm comprises:
acquiring a plurality of distributed data centers of an Internet of things platform;
the feasible path calculation is respectively carried out on each distributed data center based on a multi-objective optimization algorithm to obtain a feasible path set of each distributed data center, ，Is a set of feasible paths,/>Represents a viable path, wherein/>Is the current viable path,/>Is a further possible path which is a path,Is/>Individual objective function vs. feasible path/>Evaluation of/>Is a set of objective functions;
carrying out data flow network construction on the feasible path set of each distributed data center to obtain a data flow initial network of each distributed data center;
Respectively carrying out network flow distribution optimization on the data flow conversion initial network of each distributed data center through a network flow distribution model to obtain a network flow distribution strategy, wherein the network flow distribution model comprises the following components: ，/> Representing slave node/> To node/>Per unit data streaming cost,/>Is a decision variable representing the slave node/>To node/>Data traffic of/>Node/>Total data volume to be processed,/>The total number of nodes and the total number of connections in the data flow initial network;
And carrying out network weight change optimization on the data flow initial network of each distributed data center according to the network flow distribution strategy to obtain a data flow target network of each distributed data center.
3. The method for data stream communication based on the internet of things platform according to claim 2, wherein the receiving, by the internet of things platform, the initial data stream set of the plurality of target internet of things devices, and performing data stream classification and multi-level data encryption on the initial data stream set, to obtain the target data stream of each target internet of things device, includes:
Receiving an initial data stream set of a plurality of target internet of things devices through the internet of things platform;
Carrying out data flow classification on the initial data flow set to obtain initial data flow of each target Internet of things device;
respectively carrying out data hierarchy division on the initial data stream of each target Internet of things device to obtain a plurality of target hierarchy data streams of each initial data stream;
Respectively matching the data encryption intensity strategies of each target level data stream, and carrying out data encryption processing on the plurality of target level data streams according to the data encryption intensity strategies to obtain a plurality of encrypted data streams of each initial data stream;
And respectively carrying out data integrity verification on a plurality of encrypted data streams of each initial data stream to obtain a data integrity verification result, and outputting a target data stream of each target Internet of things device according to the data integrity verification result.
4. The data circulation method based on the internet of things platform according to claim 3, wherein the performing data circulation optimal path matching on the target data flows of each target internet of things device based on the data circulation target network to obtain a target data circulation path of each target data flow comprises:
Carrying out data flow demand analysis on the target data flow of each target internet of things device to obtain a data flow demand set of each target data flow;
Performing relationship matching on the data flow target network of each distributed data center and the target data flow of each target internet of things device according to the data flow demand set to obtain a data flow target network corresponding to each target data flow;
traversing path nodes and analyzing data flow paths of each target data flow according to the data flow target network to obtain a plurality of candidate data flow paths of each target data flow;
analyzing path dependency and constraint conditions of each target data stream according to the data stream demand set to obtain the path dependency and constraint conditions;
And carrying out path optimization solving on a plurality of candidate data flow paths of each target data flow according to the path dependency and the constraint condition to obtain a target data flow path of each target data flow.
5. The method for data transfer based on the platform of the internet of things according to claim 1, wherein the performing data transfer and transfer state monitoring on the target data streams through the target data transfer path to obtain a transfer state interaction diagram of each target data stream comprises:
Performing data circulation on the target data flows through the target data circulation paths, and performing circulation state monitoring on the target data circulation paths to obtain a plurality of circulation state characteristic data of each target data flow;
Calculating target digital characteristics of each circulation state characteristic data through a single-factor cloud model respectively;
carrying out digital feature normalization on the target digital feature of each circulation state feature data to generate normalized digital features of each circulation state feature data;
Acquiring the influence intensity sequence of the plurality of circulation state feature data, and carrying out serialization conversion on the normalized digital features based on the influence intensity sequence to generate a target feature sequence of each target power supply;
Setting path weight of the target data stream path through the target feature sequence to obtain a path weight distribution diagram;
and performing directed interaction graph conversion on the data flow target network through the path weight distribution graph to obtain a flow state interaction graph of each target data flow.
6. The method for data transfer based on the platform of the internet of things according to claim 1, wherein the performing local and global transfer state feature extraction on the transfer state interaction graph of each target data stream by the node segmentation method and the graph rolling network with the multi-head attention mechanism to obtain a transfer state feature set of each transfer state interaction graph includes:
extracting local circulation state characteristics of the circulation state interaction diagrams of each target data flow respectively through a node segmentation method to obtain a plurality of local circulation state characteristics of each circulation state interaction diagram;
global circulation state feature extraction is carried out on the circulation state interaction diagram of each target data flow through a diagram rolling network with a multi-head attention mechanism, so that a plurality of global circulation state features of each circulation state interaction diagram are obtained;
carrying out mean value calculation on the local circulation state features to obtain local feature mean values, and carrying out standard deviation calculation on the local circulation state features according to the local feature mean values to obtain local standard deviations;
Performing mean value calculation on the global circulation state features to obtain a global feature mean value, and performing standard deviation calculation on the global circulation state features according to the global feature mean value to obtain a global standard deviation;
and carrying out feature integration on the local circulation state features and the global circulation state features according to the local standard deviation and the global standard deviation to obtain circulation state feature sets of each circulation state interaction diagram.
7. The internet of things platform-based data transfer method according to claim 6, wherein inputting the transfer state feature set of each transfer state interaction graph into a preset data transfer abnormality analysis model to perform data transfer abnormality analysis, obtaining a data transfer abnormality analysis result, and matching a corresponding data transfer abnormality processing scheme according to the data transfer abnormality analysis result, comprises:
inputting the circulation state feature set of each circulation state interaction diagram into a preset data circulation abnormality analysis model, wherein the data circulation abnormality analysis model comprises: an input layer, a circulating neural network and two full-connection layers;
Performing feature coding on the circulation state feature set of each circulation state interaction graph through the input layer to obtain circulation state feature coding vectors of each circulation state interaction graph;
inputting the circulation state feature coding vector of each circulation state interaction diagram into the circulation neural network to perform feature extraction to obtain a target state feature coding vector of each circulation state interaction diagram;
Inputting the target state feature coding vector of each circulation state interaction diagram into the two layers of full-connection layers to predict the abnormal probability of data circulation, so as to obtain the abnormal probability of data circulation of each circulation state interaction diagram;
comparing the data flow abnormality probability of each flow state interaction diagram with a target probability threshold value, and outputting a data flow abnormality analysis result;
and matching the corresponding data flow abnormality processing scheme according to the data flow abnormality analysis result.
8. The utility model provides a data circulation device based on thing networking platform which characterized in that, data circulation device based on thing networking platform includes:
the acquisition module is used for acquiring a plurality of distributed data centers of the Internet of things platform and creating a data flow target network of each distributed data center based on a multi-target optimization algorithm;
the classification module is used for receiving initial data flow sets of a plurality of target internet of things devices through the internet of things platform, and carrying out data flow classification and multi-level data encryption on the initial data flow sets to obtain target data flows of each target internet of things device;
The matching module is used for carrying out data flow optimal path matching on the target data flow of each target internet of things device based on the data flow target network to obtain a target data flow path of each target data flow;
The monitoring module is used for carrying out data flow and flow state monitoring on the target data flow through the target data flow path to obtain a flow state interaction diagram of each target data flow;
The feature extraction module is used for extracting local and global circulation state features of the circulation state interaction graph of each target data stream through a node segmentation method and a graph rolling network with a multi-head attention mechanism to obtain a circulation state feature set of each circulation state interaction graph;
The analysis module is used for inputting the circulation state feature set of each circulation state interaction diagram into a preset data circulation abnormality analysis model to perform data circulation abnormality analysis, obtaining a data circulation abnormality analysis result, and matching a corresponding data circulation abnormality processing scheme according to the data circulation abnormality analysis result.
9. Data circulation equipment based on thing networking platform, its characterized in that, data circulation equipment based on thing networking platform includes: a memory and at least one processor, the memory having instructions stored therein;
the at least one processor invokes the instructions in the memory to cause the internet of things platform based data transfer device to perform the internet of things platform based data transfer method of any one of claims 1-7.
10. A computer readable storage medium having instructions stored thereon, which when executed by a processor, implement the internet of things platform based data transfer method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410348418.5A CN117955746B (en) | 2024-03-26 | 2024-03-26 | Data circulation method, device, equipment and storage medium based on Internet of things platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410348418.5A CN117955746B (en) | 2024-03-26 | 2024-03-26 | Data circulation method, device, equipment and storage medium based on Internet of things platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117955746A true CN117955746A (en) | 2024-04-30 |
| CN117955746B CN117955746B (en) | 2024-05-28 |
Family
ID=90796528
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410348418.5A Active CN117955746B (en) | 2024-03-26 | 2024-03-26 | Data circulation method, device, equipment and storage medium based on Internet of things platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117955746B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100198799A1 (en) * | 2007-06-20 | 2010-08-05 | Sanjeev Krishnan | Method and Apparatus for Software Simulation |
| CN105208585A (en) * | 2014-06-23 | 2015-12-30 | 中兴通讯股份有限公司 | Scheduling information configuration method, configuration parameter processing method, scheduling information configuration device and configuration parameter processing device |
| CN109347834A (en) * | 2018-10-24 | 2019-02-15 | 广东工业大学 | Detection method, device and the equipment of abnormal data in Internet of Things edge calculations environment |
| CN109522206A (en) * | 2018-09-26 | 2019-03-26 | 平安科技（深圳）有限公司 | Abnormal data localization method, device, computer equipment and storage medium |
| US11681283B1 (en) * | 2022-05-17 | 2023-06-20 | Chengdu Qinchuan Iot Technology Co., Ltd. | Intelligent manufacturing industrial Internet of Things based on distributed control, control methods and media thereof |
| CN116487795A (en) * | 2023-06-25 | 2023-07-25 | 深圳市天富利信息技术有限公司 | New energy storage cabinet and energy storage system |
| CN117614741A (en) * | 2024-01-18 | 2024-02-27 | 中诚华隆计算机技术有限公司 | Network security vulnerability position detection method and system |
-
2024
- 2024-03-26 CN CN202410348418.5A patent/CN117955746B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100198799A1 (en) * | 2007-06-20 | 2010-08-05 | Sanjeev Krishnan | Method and Apparatus for Software Simulation |
| CN105208585A (en) * | 2014-06-23 | 2015-12-30 | 中兴通讯股份有限公司 | Scheduling information configuration method, configuration parameter processing method, scheduling information configuration device and configuration parameter processing device |
| CN109522206A (en) * | 2018-09-26 | 2019-03-26 | 平安科技（深圳）有限公司 | Abnormal data localization method, device, computer equipment and storage medium |
| CN109347834A (en) * | 2018-10-24 | 2019-02-15 | 广东工业大学 | Detection method, device and the equipment of abnormal data in Internet of Things edge calculations environment |
| US11681283B1 (en) * | 2022-05-17 | 2023-06-20 | Chengdu Qinchuan Iot Technology Co., Ltd. | Intelligent manufacturing industrial Internet of Things based on distributed control, control methods and media thereof |
| CN116487795A (en) * | 2023-06-25 | 2023-07-25 | 深圳市天富利信息技术有限公司 | New energy storage cabinet and energy storage system |
| CN117614741A (en) * | 2024-01-18 | 2024-02-27 | 中诚华隆计算机技术有限公司 | Network security vulnerability position detection method and system |
Non-Patent Citations (4)
| Title |
|---|
| 刘秀华;孙伟松;高嵩;: "基于物联网的自动报告管理系统及其应用研究", 现代商贸工业, no. 06, 15 March 2010 (2010-03-15) * |
| 吕继伟;: "基于泛在电力物联网的换流站在线监测系统优化综述", 电力工程技术, no. 06, 28 November 2019 (2019-11-28) * |
| 洪晓彬等人: "大型多层次数据库外层数据传输同步加密仿真", 《计算机仿真》, vol. 35, no. 6, 15 June 2018 (2018-06-15) * |
| 熊俊浩;盛金根;李永建;程小琴;孙鹏;: "基于知识特征的界面交互模型及事件激活原则", 计算机工程与应用, no. 26, 11 September 2012 (2012-09-11) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117955746B (en) | 2024-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kumar et al. | A Distributed framework for detecting DDoS attacks in smart contract‐based Blockchain‐IoT Systems by leveraging Fog computing | |
| CN111787011B (en) | Intelligent analysis and early warning system, method and storage medium for security threat of information system | |
| Nakhodchi et al. | Steeleye: An application-layer attack detection and attribution model in industrial control systems using semi-deep learning | |
| Mounica et al. | RETRACTED: Detecting Sybil Attack In Wireless Sensor Networks Using Machine Learning Algorithms | |
| Patil et al. | [Retracted] Prediction of IoT Traffic Using the Gated Recurrent Unit Neural Network‐(GRU‐NN‐) Based Predictive Model | |
| Vidal et al. | Framework for anticipatory self-protective 5G environments | |
| CN109871711B (en) | Ocean big data sharing and distributing risk control model and method | |
| CN117955746B (en) | Data circulation method, device, equipment and storage medium based on Internet of things platform | |
| CN117580046A (en) | Deep learning-based 5G network dynamic security capability scheduling method | |
| Yu et al. | Network security monitoring method based on deep learning | |
| Krundyshev | Neural network approach to assessing cybersecurity risks in large-scale dynamic networks | |
| Tang et al. | Intelligent awareness of delay-sensitive internet traffic in digital twin network | |
| Ahakonye et al. | Trees Bootstrap Aggregation for Detection and Characterization of IoT-SCADA Network Traffic | |
| Joldasbayev et al. | Development of an Intelligent Service Delivery System to Increase Efficiency of Software Defined Networks. | |
| Sezgin et al. | AID4I: An Intrusion Detection Framework for Industrial Internet of Things Using Automated Machine Learning. | |
| Gnanamonickam et al. | Intelligent Traffic Classification Feature Engineering Technique (ITCFFE) For SDN Networks Based On Neural Networks | |
| Dhingra et al. | RLET: a lightweight model for ubiquitous multi-class intrusion detection in sustainable and secured smart environment | |
| Wu et al. | An active learning framework using deep Q-network for zero-day attack detection | |
| Al-Muntaser et al. | Cybersecurity Advances in SCADA Systems | |
| CN117527622B (en) | Data processing method and system of network switch | |
| Tiwari et al. | Sentinel Shield: Leveraging ConvLSTM and Elephant Herd Optimization for Advanced Network Intrusion Detection | |
| Fong et al. | Data Stream Mining in Fog Computing Environment with Feature Selection Using Ensemble of Swarm Search Algorithms | |
| Grammatopoulou et al. | A multi-step and resilient predictive Q-learning algorithm for IoT: a case study in water supply networks | |
| Sivakumar et al. | Intrusion Detection System for Securing the SCADA Industrial Control System | |
| Tang et al. | High-speed mobile communication network and wireless sensor network convergence service traffic prediction model and security mechanism design |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |