A method for intelligent scheduling of Internet of Things link nodes

Abstract

The present invention discloses an intelligent scheduling method for an Internet of Things link node, which relates to the technical field of virtual resource scheduling, including: establishing an Internet of Things device link node network map, evaluating the scene initialization performance of each Internet of Things device link node according to the structural parameters of each Internet of Things device, and establishing an Internet of Things device link node network scene performance map; obtaining the transmission data stream of the real-time Internet of Things device link node, analyzing the transmission properties of the transmission data stream to classify and package data, and obtaining a number of transmission preference classification data packets; establishing a transit link node matrix; verifying the initialization performance of each Internet of Things device link node in the transit link node matrix and the adaptability index between a number of transmission preference classification data packets, and generating an Internet of Things link node data transmission scheme with the Internet of Things device link node having the maximum value of the adaptability index. The advantages of the present invention are: improving data transmission efficiency and increasing data transmission stability.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual resource scheduling, and in particular to an intelligent scheduling method for an Internet of Things link node. Background Art

[0002] Intelligent scheduling of IoT link nodes refers to the efficient and reasonable allocation and deployment of device link nodes in the IoT through advanced computer technology and intelligent management systems to achieve optimal utilization and configuration of resources.

[0003] However, the existing intelligent scheduling scheme for IoT link nodes is to transfer the transmission data to IoT devices that meet the performance redundancy of link nodes. However, when data transmission exceeds the performance bottleneck of IoT link nodes, the scheduling efficiency in this aspect is very limited. Summary of the invention

[0004] In order to solve the above technical problems, a method for intelligent scheduling of Internet of Things link nodes is provided. This technical solution solves the above-mentioned problem that the existing intelligent scheduling scheme of Internet of Things link nodes is to transfer the transmission data to the Internet of Things devices that meet the performance redundancy of the link nodes. However, when the data transmission exceeds the performance bottleneck of the Internet of Things link nodes, the scheduling efficiency in this aspect is very limited.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] An intelligent scheduling method for an Internet of Things link node, comprising:

[0007] Obtain all IoT device link nodes and establish a network graph of IoT device link nodes;

[0008] Based on the network graph of IoT device link nodes, the scenario initialization performance of each IoT device link node is evaluated according to the structural parameters of each IoT device, and a network scenario performance graph of IoT device link nodes is established;

[0009] Obtain the transmission data stream of the real-time IoT device link node, analyze the transmission properties of the transmission data stream to classify and package the data, and obtain several transmission preference classification data packets;

[0010] Determine the transmission target locations of several transmission preference classified data packets, screen out the optional link nodes in the IoT device link node network scenario performance map, and establish a transit link node matrix;

[0011] Verify the initialization performance of each IoT device link node in the transit link node matrix and the adaptability index between several transmission preference classification data packets, and generate the IoT link node data transmission plan based on the IoT device link node with the maximum value of the adaptability index.

[0012] Preferably, based on the network graph of the IoT device link nodes, according to the structural parameters of each IoT device, the scenario initialization performance of each IoT device link node is evaluated, and the establishment of the IoT device link node network scenario performance graph specifically includes:

[0013] Based on the link relationship of known IoT device link nodes, a network graph of IoT device link nodes is constructed;

[0014] Acquire the construction parameters of the IoT device link node; the construction parameters of the IoT device link node include: hardware layer construction parameters and software layer construction parameters;

[0015] Obtain a data transmission layer support protocol table of an IoT device link node to determine the transmission scenarios supported by the IoT device link node; the transmission layer support protocol table includes: application layer transmission protocol, network layer transmission protocol, and data link layer transmission protocol;

[0016] Based on the construction parameters of IoT device link nodes and the transmission scenarios supported by IoT device link nodes, a scenario performance monitoring model for IoT device link nodes is constructed to analyze the maximum performance of IoT device link nodes in different transmission scenarios, which is recorded as the scenario initialization performance of IoT device link nodes.

[0017] Based on the scene initialization performance of the IoT device link node, the performance attribute is assigned to the IoT device link node network map to obtain the IoT device link node network scene performance map;

[0018] The analysis of the maximum performance of IoT device link nodes in different transmission scenarios is specifically as follows:

[0019]

[0020] In the formula, To maximize the performance of IoT device link nodes in different transmission scenarios, The first node of the IoT device link construction parameters, The first node of the IoT device link Supported transmission scenarios, is the intercept, is the regression coefficient of the construction parameter, is the regression coefficient of the transmission scenario.

[0021] Preferably, the transmission data stream of the real-time IoT device link node is obtained, the transmission attributes of the transmission data stream are analyzed to classify and package the data, and a number of transmission preference classification data packets are obtained, specifically including:

[0022] Perform data preprocessing on the transmission data stream of the real-time IoT device link node;

[0023] Based on the transmission data stream of the real-time IoT device link node, the transmission data stream of the real-time IoT device link node is divided and bound according to the unit time to obtain the transmission data timing stream of the IoT device link node;

[0024] The transmission data time series flow of the IoT device link node is reduced in dimension by the principal component analysis algorithm to obtain the reduced-dimensional transmission data time series flow of the IoT device link node;

[0025] Based on the reduced-dimensional transmission data time series flow of the IoT device link node, the unit time is used as the observation window, and K-means clustering analysis is used to cluster the transmission attributes of the reduced-dimensional transmission data time series flow within the observation window to obtain several transmission preference classification data packets; the transmission attributes include: delay attribute and packet loss tolerance attribute;

[0026] Among them, the K-means cluster analysis is specifically as follows:

[0027]

[0028] In the formula, For the Clusters in The mean of observation windows, For the The observation windows are classified into A clustered data set, For data points The weight of For Data points of a data set Weighted sum, For the forgetting factor, For the Clusters in The mean of the observation windows.

[0029] Preferably, verifying the initialization performance of each IoT device link node in the transit link node matrix and the adaptability index between a number of transmission preference classification data packets, and generating an IoT link node data transmission scheme with the IoT device link node having the maximum adaptability index specifically includes:

[0030] Construct a transmission scenario adaptation decision tree for IoT device link nodes;

[0031] Based on the maximum performance of each IoT device link node in the transit link node matrix under different transmission scenarios, the maximum performance under different transmission scenarios is used as the segmentation leaf node feature data of the transmission scenario adaptation decision tree of the IoT device link node, and a transmission scenario adaptation random forest of the serially connected device link node is constructed;

[0032] Bring several transmission preference classification data packets per unit time into the transmission scenario adaptation random forest of the serial networking device link node, calculate the adaptability index between each transmission preference classification data packet and the segmentation leaf node feature data, take the maximum value of the adaptability index of each leaf node as the transmission path, and generate the data transmission plan for the IoT link node;

[0033] The calculation of the adaptability index between each transmission preference classification data packet and the segmentation leaf node feature data is specifically as follows:

[0034]

[0035] In the formula, Classify packets for transmission preferences And split leaf node feature data The compatibility index between For the Transmission preference classification packets, To split the leaf node feature data, is the probability of feature data taking value, For feature data The weight coefficient of is a logarithmic function.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The present invention proposes an intelligent scheduling scheme for IoT link nodes, which collects IoT device link node information, constructs a network map and evaluates the initialization performance map of each node, analyzes the preferences of historical transmission data streams, classifies data packets according to preferences, screens out IoT device link nodes according to the transmission target location of the data, constructs a transit link node matrix, verifies and screens out IoT device link nodes with the highest adaptability scores to the transmission preference classified data packets, and thus formulates an optimal data transmission scheme. The advantages of this scheme are: improving data transmission efficiency and increasing data transmission stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of an intelligent scheduling method for an Internet of Things link node;

[0039] Figure 2A flow chart of the method for establishing a performance graph of network scenarios for IoT device link nodes;

[0040] Figure 3 Flowchart of a method for classifying data packets to obtain a number of transmission preferences;

[0041] Figure 4 A flow chart of a method for generating a data transmission scheme for nodes in an IoT link. DETAILED DESCRIPTION

[0042] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0043] Reference Figure 1 As shown, a method for intelligent scheduling of IoT link nodes includes:

[0044] Obtain all IoT device link nodes and establish a network graph of IoT device link nodes;

[0045] Based on the network graph of IoT device link nodes, the scenario initialization performance of each IoT device link node is evaluated according to the structural parameters of each IoT device, and a network scenario performance graph of IoT device link nodes is established;

[0046] Obtain the transmission data stream of the real-time IoT device link node, analyze the transmission properties of the transmission data stream to classify and package the data, and obtain several transmission preference classification data packets;

[0047] Determine the transmission target locations of several transmission preference classified data packets, screen out the optional link nodes in the IoT device link node network scenario performance map, and establish a transit link node matrix;

[0048] Verify the initialization performance of each IoT device link node in the transit link node matrix and the adaptability index between several transmission preference classification data packets, and generate the IoT link node data transmission plan based on the IoT device link node with the maximum value of the adaptability index.

[0049] It is understandable that when communication congestion occurs in the link nodes of IoT devices, most of the reasons are that the communication protocol at the transmission level of the link nodes of IoT devices does not match the data to be transmitted, resulting in a large gap between the data processing speed and the data upload speed, causing the data to be transmitted to accumulate, resulting in performance bottlenecks in the link nodes of IoT devices. Therefore, data is divided and classified according to data preference, the current state of the link nodes of IoT devices is evaluated, the adaptability scores between the link nodes of IoT devices and the data to be transmitted are analyzed, and the link nodes with the highest adaptability scores are selected for scheduling and planning, so as to reduce performance bottlenecks and communication congestion in the link nodes of IoT devices.

[0050] This solution collects information about IoT device link nodes, builds a network map and evaluates the initialization performance map of each node, analyzes the preferences of historical transmission data streams, classifies data packets according to preferences, selects IoT device link nodes according to the data transmission target location, builds a transit link node matrix, verifies and selects IoT device link nodes with the highest adaptability scores for the transmission preference classified data packets, and thus develops the optimal data transmission solution. The advantages of this solution are: improving data transmission efficiency and increasing data transmission stability.

[0051] Reference Figure 2 As shown, based on the network graph of IoT device link nodes, according to the structural parameters of each IoT device, the scenario initialization performance of each IoT device link node is evaluated, and the network scenario performance graph of IoT device link nodes is established, which specifically includes:

[0052] Based on the link relationship of known IoT device link nodes, a network graph of IoT device link nodes is constructed;

[0053] Acquire the construction parameters of the IoT device link node; the construction parameters of the IoT device link node include: hardware layer construction parameters and software layer construction parameters;

[0054] Obtain a data transmission layer support protocol table of an IoT device link node to determine the transmission scenarios supported by the IoT device link node; the transmission layer support protocol table includes: application layer transmission protocol, network layer transmission protocol, and data link layer transmission protocol;

[0055] Based on the construction parameters of IoT device link nodes and the transmission scenarios supported by IoT device link nodes, a scenario performance monitoring model for IoT device link nodes is constructed to analyze the maximum performance of IoT device link nodes in different transmission scenarios, which is recorded as the scenario initialization performance of IoT device link nodes.

[0056] Based on the scene initialization performance of the IoT device link node, the performance attribute is assigned to the IoT device link node network map to obtain the IoT device link node network scene performance map;

[0057] The analysis of the maximum performance of IoT device link nodes in different transmission scenarios is specifically as follows:

[0058]

[0059] In the formula, To maximize the performance of IoT device link nodes in different transmission scenarios, The first node of the IoT device link construction parameters, The first node of the IoT device link Supported transmission scenarios, is the intercept, is the regression coefficient of the construction parameter, is the regression coefficient of the transmission scenario.

[0060] It is understandable that since the application layer transmission protocol, network layer transmission protocol, and data link layer transmission protocol of the IoT device link node usually include several transmission protocols, namely, transmission scenarios, and different transmission scenarios correspond to different construction parameter requirements of the IoT device link node, there will be performance differences when the IoT device link node is in different transmission scenarios. Therefore, by evaluating the performance status of the IoT device link node in different transmission scenarios, reference data is provided for subsequent data transmission.

[0061] Reference Figure 3 As shown, the transmission data stream of the real-time IoT device link node is obtained, the transmission attributes of the transmission data stream are analyzed to classify and package the data, and several transmission preference classification data packets are obtained, including:

[0062] Perform data preprocessing on the transmission data stream of the real-time IoT device link node;

[0063] Based on the transmission data stream of the real-time IoT device link node, the transmission data stream of the real-time IoT device link node is divided and bound according to the unit time to obtain the transmission data timing stream of the IoT device link node;

[0064] The transmission data time series flow of the IoT device link node is reduced in dimension by the principal component analysis algorithm to obtain the reduced-dimensional transmission data time series flow of the IoT device link node;

[0065] Based on the reduced-dimensional transmission data time series flow of the IoT device link node, the unit time is used as the observation window, and K-means clustering analysis is used to cluster the transmission attributes of the reduced-dimensional transmission data time series flow within the observation window to obtain several transmission preference classification data packets; the transmission attributes include: delay attribute and packet loss tolerance attribute;

[0066] Among them, the K-means cluster analysis is specifically as follows:

[0067]

[0068] In the formula, For the Clusters in The mean of observation windows, For the The observation windows are classified into A clustered data set, For data points The weight of For Data points of a data set Weighted sum, For the forgetting factor, For the Clusters in The mean of the observation windows.

[0069] It is understandable that due to the differences in transmission data faced by the link nodes of IoT devices, after the data is comprehensively clustered according to the transmission attributes, it is possible to perform correlation matching according to the transmission protocol of the link nodes of each IoT device to seek the optimal transmission method.

[0070] Reference Figure 4 As shown, the initialization performance of each IoT device link node in the transit link node matrix is ​​verified with the adaptability index between several transmission preference classification data packets, and the IoT device link node with the maximum adaptability index is used to generate the IoT link node data transmission scheme, which specifically includes:

[0071] Construct a transmission scenario adaptation decision tree for IoT device link nodes;

[0072] Based on the maximum performance of each IoT device link node in the transit link node matrix under different transmission scenarios, the maximum performance under different transmission scenarios is used as the segmentation leaf node feature data of the transmission scenario adaptation decision tree of the IoT device link node, and a transmission scenario adaptation random forest of the serially connected device link node is constructed;

[0073] Bring several transmission preference classification data packets per unit time into the transmission scenario adaptation random forest of the serial networking device link node, calculate the adaptability index between each transmission preference classification data packet and the segmentation leaf node feature data, take the maximum value of the adaptability index of each leaf node as the transmission path, and generate the data transmission plan for the IoT link node;

[0074] The calculation of the adaptability index between each transmission preference classification data packet and the segmentation leaf node feature data is specifically as follows:

[0075]

[0076] In the formula, Classify packets for transmission preferences And split leaf node feature data The compatibility index between For the Transmission preference classification packets, To split the leaf node feature data, is the probability of feature data taking value, For feature data The weight coefficient of is a logarithmic function.

[0077] It is understandable that due to the different transmission states of the data packets classified by transmission preference when facing different transmission protocols of the IoT device link nodes, the IoT link node data transmission scheme is generated by constructing decision trees and random forests, and the impact of different transmission scenarios, performance, transmission protocols and data packet transmission states on data transmission is considered. This scheme helps to improve the efficiency and reliability of data transmission and adapt to different IoT application scenarios and needs.

[0078] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for intelligent scheduling of Internet of Things link nodes, characterized in that: include: Obtain all IoT device link nodes and establish a network graph of IoT device link nodes; Based on the network graph of IoT device link nodes, the scenario initialization performance of each IoT device link node is evaluated according to the structural parameters of each IoT device, and a network scenario performance graph of IoT device link nodes is established; Obtain the transmission data stream of the real-time IoT device link node, analyze the transmission properties of the transmission data stream to classify and package the data, and obtain several transmission preference classification data packets; Determine the transmission target locations of several transmission preference classified data packets, screen out the optional link nodes in the IoT device link node network scenario performance map, and establish a transit link node matrix; Verify the initialization performance of each IoT device link node in the transit link node matrix and the adaptability index between several transmission preference classification data packets, and generate the IoT link node data transmission plan based on the IoT device link node with the maximum adaptability index; Among them, verify the initialization performance of each IoT device link node in the transit link node matrix and the adaptability index between several transmission preference classification data packets, and use the IoT device link node with the maximum adaptability index to generate the IoT link node data transmission plan, which specifically includes: Construct a transmission scenario adaptation decision tree for IoT device link nodes; Based on the maximum performance of each IoT device link node in the transit link node matrix under different transmission scenarios, the maximum performance under different transmission scenarios is used as the segmentation leaf node feature data of the transmission scenario adaptation decision tree of the IoT device link node, and a transmission scenario adaptation random forest of the serially connected device link node is constructed; Bring several transmission preference classification data packets per unit time into the transmission scenario adaptation random forest of the serial networking device link node, calculate the adaptability index between each transmission preference classification data packet and the segmentation leaf node feature data, take the maximum value of the adaptability index of each leaf node as the transmission path, and generate the data transmission plan for the IoT link node; The calculation of the adaptability index between each transmission preference classification data packet and the segmentation leaf node feature data is specifically as follows: In the formula, Classify packets for transmission preferences And split leaf node feature data The compatibility index between For the Transmission preference classification packets, To split the leaf node feature data, is the probability of feature data taking value, For feature data The weight coefficient of is a logarithmic function.

2. According to claim 1, a method for intelligent scheduling of Internet of Things link nodes is characterized in that: Based on the network graph of IoT device link nodes, the scenario initialization performance of each IoT device link node is evaluated according to the structural parameters of each IoT device, and the network scenario performance graph of IoT device link nodes is established, which specifically includes: Based on the link relationship of known IoT device link nodes, a network graph of IoT device link nodes is constructed; Acquire the construction parameters of the IoT device link node; the construction parameters of the IoT device link node include: hardware layer construction parameters and software layer construction parameters; Obtain a data transmission layer support protocol table of an IoT device link node to determine the transmission scenarios supported by the IoT device link node; the transmission layer support protocol table includes: application layer transmission protocol, network layer transmission protocol, and data link layer transmission protocol; Based on the construction parameters of IoT device link nodes and the transmission scenarios supported by IoT device link nodes, a scenario performance monitoring model for IoT device link nodes is constructed to analyze the maximum performance of IoT device link nodes in different transmission scenarios, which is recorded as the scenario initialization performance of IoT device link nodes. Based on the scenario initialization performance of the IoT device link node, performance attributes are assigned to the IoT device link node network map to obtain the IoT device link node network scenario performance map.

3. According to claim 2, a method for intelligent scheduling of Internet of Things link nodes is characterized in that: The analysis of the maximum performance of IoT device link nodes in different transmission scenarios is specifically as follows: In the formula, To maximize the performance of IoT device link nodes in different transmission scenarios, The first node of the IoT device link construction parameters, The first node of the IoT device link Supported transmission scenarios, is the intercept, is the regression coefficient of the construction parameter, is the regression coefficient of the transmission scenario.

4. The method for intelligent scheduling of Internet of Things link nodes according to claim 3 is characterized in that: Obtain the transmission data stream of the real-time IoT device link node, analyze the transmission properties of the transmission data stream to classify and package the data, and obtain several transmission preference classification data packets, including: Perform data preprocessing on the transmission data stream of the real-time IoT device link node; Based on the transmission data stream of the real-time IoT device link node, the transmission data stream of the real-time IoT device link node is divided and bound according to the unit time to obtain the transmission data timing stream of the IoT device link node; The transmission data timing stream of the IoT device link node is subjected to data dimension reduction by using the principal component analysis algorithm to obtain the dimension-reduced transmission data timing stream of the IoT device link node; Based on the reduced-dimensional transmission data timing flow of the IoT device link node, the unit time is used as the observation window, and K-means clustering analysis is used to cluster the transmission attributes of the reduced-dimensional transmission data timing flow within the observation window to obtain several transmission preference classification data packets; the transmission attributes include: delay attribute and packet loss tolerance attribute.

5. The method for intelligent scheduling of Internet of Things link nodes according to claim 4 is characterized in that: The specific use of K-means cluster analysis is as follows: In the formula, For the Clusters in The mean of observation windows, For the The observation windows are classified into A clustered data set, For data points The weight of For Data points of a data set Weighted sum, For the forgetting factor, For the Clusters in The mean of the observation windows.

Images