A method for adapting domestic chip network element and network management system
By obtaining and analyzing historical communication requirements information, screening and testing network element equipment, and establishing a network that matches future needs, the problem of network element equipment failure in the existing technology affecting network performance, and improving network quality and user experience.
Patent Information
- Application Number
- CN202410661860.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-27
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-05-27
AI Technical Summary
The existing detection methods cannot effectively perform network slicing based on network element equipment failures or performance problems, resulting in a degradation in network performance and unable to meet the actual needs of users, bringing users a poor user experience.
By obtaining historical communication requirements information of the target area, predicting future needs, setting status reporting components and edge computing components, collecting heartbeat data and environment information of network elements, filtering alternative network elements, performing compatibility testing and chain building requests, and finally establishing A network that matches future communication needs.
The quality of networking has been improved, ensuring that the networking matches the future communication needs of the target area, and improving user experience.
Smart Images

Figure CN118632267B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of communication technology, and in particular to an adaptation method for a domestic chip network element and a network management system. Background Art
[0002] With the development of communication technology and the needs of people's lives, communication networks are becoming more and more complex. Especially in mobile communication networks, in order to achieve the purpose of controlling network element devices, network element devices need to be connected to the network management system, and the network management system can manage the connected network element devices.
[0003] With the rapid development of communication technology, the fifth generation of mobile communication technology (5th Generation Mobile Communication Technology, referred to as 5G) has gradually begun to be commercialized and has formed a certain scale. 5G technology can use network slicing technology for networking. Network slicing technology is an on-demand networking method. Network slicing technology can perform network slicing for different application scenarios and services. It isolates the original physical network and decomposes it into several logical networks. Each slice is configured differently, making the network used by users smoother and providing users with high-quality services.
[0004] However, the creation of network slices requires the support of base station equipment (i.e., network elements). The number of base station equipment in operation worldwide has exceeded 10 million. Base station equipment is distributed in thousands of computer rooms or signal towers in various cities and villages. Base station equipment will inevitably encounter various faults during operation. In the existing detection methods, no matter how the health status of the network element is, how to perform network slicing based on faulty or poorly performing network elements will greatly reduce the performance of the network, fail to meet the actual networking and business needs of users, and bring a bad user experience.
[0005] Therefore, it is necessary to provide an adaptation method for domestic chip network elements and network management systems to improve the quality of networking. Summary of the invention
[0006] The present invention provides an adaptation method for a domestic chip network element and a network management system, comprising: obtaining historical communication demand information of a target area; predicting future communication demand information of the target area based on the historical communication demand information of the target area; setting a status reporting component at each network element; setting multiple edge computing components, wherein one of the edge computing components is used to receive heartbeat data reported by at least one network element and environmental information reported by at least one status reporting component, and based on the heartbeat data reported by the network element and the environmental information reported by the status reporting component set at the network element, determine the status of the network element, and report the status of the network element to the network management; determine multiple candidate network elements based on the target area; the network management reports the status of each candidate network element based on the heartbeat data reported by the network element and the environmental information reported by the status reporting component set at the network element; The network manager determines the status of the candidate network element, performs a first screening on the multiple candidate network elements, and determines multiple first candidate network elements; the network manager sends trigger information to the edge computing component corresponding to each of the first candidate network elements; the edge computing component exchanges the trigger information to the corresponding first candidate network element; the first candidate network element sends a link establishment request to the network manager based on the trigger information, wherein the link establishment request includes the version number of the first candidate network element; the network manager performs a second screening on the multiple first candidate network elements based on the link establishment request sent by each of the first candidate network elements, and determines multiple second candidate network elements; the network manager establishes a network based on the multiple second candidate network elements and the future communication demand information of the target area.
[0007] Furthermore, based on the historical communication demand information of the target area, the future communication demand information of the target area is predicted, including: obtaining the historical communication demand information of multiple locations in the target area at multiple historical time points; clustering the multiple locations based on the historical communication demand information of each of the locations at multiple historical time points to determine multiple sub-areas of the target area; for each of the sub-areas, fitting the historical communication demand information of the sub-area at multiple historical time points based on the least squares method to determine the historical communication demand information of the sub-area at multiple historical time points; based on the historical communication demand information of each of the sub-areas at multiple historical time points, predicting the future communication demand of each of the sub-areas.
[0008] Furthermore, based on the historical communication demands of each of the locations at multiple historical time points, the multiple locations are clustered, including: for each of the locations, based on the historical communication demand information of the location at multiple historical time points, determining the historical communication demand fluctuation parameters of the location in multiple historical time periods; for any two of the locations, calculating the physical distance between the two locations and the similarity of historical communication demand fluctuations; and clustering the multiple locations based on the physical distance between any two locations and the similarity of historical communication demand fluctuations through a K-means clustering algorithm.
[0009] Furthermore, the multiple locations are clustered based on the physical distance between any two of the locations and the similarity of historical communication demand fluctuations by using a K-means clustering algorithm, including: S11, randomly selecting K locations as initial cluster centers, and executing S12; S12, for each of the locations, determining a candidate cluster center based on the physical distance between the location and each of the cluster centers, calculating the cluster distance between the location and each of the candidate cluster centers based on the similarity of historical communication demand fluctuations between the location and each of the candidate cluster centers, and clustering the location Assign the candidate cluster center with the shortest clustering distance to it, and execute S13; S13, for each location clustering cluster, calculate the mean similarity of historical communication demand fluctuations corresponding to each location included in the location clustering cluster, and take the location with the largest mean similarity of historical communication demand fluctuations as the new cluster center, and execute S14; S14, determine whether the cluster center determined by this clustering has changed from the cluster center determined by the previous clustering, if so, execute S12, if not, execute S15; S15, determine whether the predetermined number of iterations has been reached, if not, execute S12, if so, complete clustering.
[0010] Furthermore, based on the historical communication demand information of each sub-region at multiple historical time points, the future communication demand of each sub-region is predicted, including: for any two sub-regions, based on the historical communication demand information of the two sub-regions at multiple historical time points, the correlation coefficient between the two sub-regions is calculated; for each sub-region, based on the correlation coefficient between any two sub-regions, the associated sub-region of each sub-region is determined; for each sub-region, based on the historical communication demand information of the sub-region at multiple historical time points and the historical communication demand information of the associated sub-region of the sub-region at multiple historical time points, the future communication demand of the sub-region is predicted by a demand prediction model.
[0011] Furthermore, the heartbeat data includes at least CPU temperature, CPU usage, physical memory usage, hard disk read and write data flow, network inbound and outbound flow, load, throughput, delay and packet loss rate; the environmental information includes at least ambient temperature, ambient humidity and ambient electromagnetic interference.
[0012] Furthermore, the edge computing component determines the state of the network element based on the heartbeat data reported by the network element and the environmental information reported by the state reporting component set at the network element, including: determining the operation abnormality score of the network element through a first state determination model based on the CPU temperature of the network element, CPU utilization, physical memory utilization, hard disk read and write data traffic, network inbound and outbound traffic, load, throughput, delay and packet loss rate; determining the environmental abnormality score of the network element through a second state determination model based on the ambient temperature, ambient humidity and ambient electromagnetic interference of the network element; determining the state of the network element based on the operation abnormality score and the environmental abnormality score of the network element.
[0013] Furthermore, the network manager performs a second screening of the multiple first candidate network elements based on the link establishment request sent by each of the first candidate network elements to determine multiple second candidate network elements, including: the network manager performs a compatibility test on each of the first candidate network elements according to the version number of each of the first candidate network elements to obtain a compatibility test result; and performs a second screening of the multiple first candidate network elements according to the compatibility test result to determine multiple second candidate network elements.
[0014] Furthermore, the network manager establishes a network based on the multiple second candidate network elements and the future communication demand information of the target area, including: the network manager determines the network element density of each sub-area at multiple future time points based on the future communication demand of each sub-area; the network manager establishes a dynamic network based on the network element density of each sub-area at multiple future time points and the multiple second candidate network elements.
[0015] Furthermore, the network manager establishes a dynamic network based on the network element density of each of the sub-areas at multiple future time points and the multiple second candidate network elements, including: the network manager establishes multiple candidate dynamic networks based on the network element density of each of the sub-areas at multiple future time points and the multiple second candidate network elements; establishes a networking screening system; based on the networking screening system, screens the multiple candidate dynamic networks to determine the dynamic network.
[0016] Compared with the prior art, the adaptation method of a domestic chip network element and a network management system provided in this specification has at least the following beneficial effects:
[0017] 1. By obtaining the historical communication demand information of the target area, the future communication demand information of the target area is predicted. By setting the status reporting component and the edge computing component, the heartbeat data and environmental information of the network element are collected to provide data support for determining the status of the network element, thereby providing a basis for the screening of the network element. On this basis, the network manager sends the trigger information to the edge computing component corresponding to each first candidate network element. The edge computing component exchanges the trigger information to the corresponding first candidate network element. The first candidate network element sends a link establishment request to the network manager based on the trigger information, and performs a compatibility test on the first candidate network element to achieve the second round of screening. Finally, the network manager combines the multiple second candidate network elements after the two rounds of screening and the future communication demand information of the target area to establish a network, so that the final established network matches the future communication demand of the target area and has a higher network quality.
[0018] 2. Based on the historical communication demand information of each location in the target area at multiple historical time points, multiple locations are clustered to determine multiple sub-areas of the target area, and then the historical communication demand information of the sub-areas at multiple historical time points is fitted by the least squares method to determine more accurate historical communication demand information of the sub-areas at multiple historical time points;
[0019] 3. Based on the historical communication demand information of the two sub-regions at multiple historical time points, the correlation coefficient between the two sub-regions is calculated to provide more dimensional data support for the subsequent prediction of the future communication demand of the sub-regions. Through the demand prediction model, based on the historical communication demand information of the sub-regions at multiple historical time points and the historical communication demand information of the associated sub-regions of the sub-regions at multiple historical time points, a more accurate prediction of the future communication demand of the sub-regions is made. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:
[0021] Figure 1 It is a flow chart of a method for adapting a domestic chip network element and a network management system shown in an embodiment of the present application;
[0022] Figure 2 It is a flow chart of clustering multiple locations shown in one embodiment of the present application. DETAILED DESCRIPTION
[0023] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly introduces the drawings required for describing the embodiments.
[0024] Figure 1is a flow chart of a method for adapting a domestic chip network element to a network management system shown in an embodiment of the present application, such as Figure 1 As shown, a method for adapting a domestic chip network element to a network management system may include the following steps.
[0025] Step 111: Acquire historical communication demand information of the target area.
[0026] The historical communication demand information of the target area may include historical communication demand information of multiple locations at multiple historical time points.
[0027] Step 112: predicting future communication demand information of the target area based on the historical communication demand information of the target area.
[0028] Specifically include:
[0029] Obtain historical communication demand information of multiple locations in a target area at multiple historical time points;
[0030] Based on the historical communication demand information of each location at multiple historical time points, multiple locations are clustered to determine multiple sub-areas of the target area;
[0031] For each sub-region, fitting the historical communication demand information of the sub-region at multiple historical time points based on the least square method to determine the historical communication demand information of the sub-region at multiple historical time points;
[0032] Based on the historical communication demand information of each sub-region at multiple historical time points, the future communication demand of each sub-region is predicted.
[0033] In some embodiments, clustering the multiple locations based on the historical communication demands of each location at multiple historical time points includes:
[0034] For each location, based on historical communication demand information of the location at multiple historical time points, determine historical communication demand fluctuation parameters of the location in multiple historical time periods;
[0035] For any two locations, calculate the physical distance between the two locations and the similarity of historical communication demand fluctuations;
[0036] Through the K-means clustering algorithm, multiple locations are clustered based on the physical distance between any two locations and the similarity of historical communication demand fluctuations.
[0037] For example, the historical communication demand fluctuation parameter of a location in each historical time period can be calculated based on the following formula:
[0038]
[0039] Among them, V (i,j)is the historical communication demand fluctuation parameter of the i-th location in the j-th historical time period, D ((i,j),t) is the historical communication demand of the ith location at the tth historical time point in the jth historical time period, and T is the total number of historical time points included in the jth historical time period.
[0040] The historical communication demand fluctuation similarity can be calculated based on the following formula:
[0041]
[0042] Among them, S (i,e) is the historical communication demand fluctuation similarity between the i-th position and the e-th position, M1 is a preset parameter, and M1>0, V (e,j) is the historical communication demand fluctuation parameter of the e-th location in the j-th historical time period, and J is the total number of historical time periods.
[0043] Figure 2 is a flow chart of clustering multiple locations shown in an embodiment of the present application, such as Figure 2 As shown, in some embodiments, a K-means clustering algorithm is used to cluster multiple locations based on the physical distance between any two locations and the similarity of historical communication demand fluctuations, including:
[0044] S11, randomly select K positions as initial cluster centers, and execute S12;
[0045] S12, for each location, based on the physical distance between the location and each cluster center, determine the candidate cluster center, based on the similarity of historical communication demand fluctuations between the location and each candidate cluster center, calculate the cluster distance between the location and each candidate cluster center, and assign the location to the candidate cluster center with the shortest cluster distance, and execute S13;
[0046] S13, for each location cluster, calculating the mean similarity of historical communication demand fluctuations corresponding to each location included in the location cluster, taking the location with the largest mean similarity of historical communication demand fluctuations as the new cluster center, and executing S14;
[0047] S14, determining whether the cluster center determined by the current clustering is different from the cluster center determined by the previous clustering, if so, executing S12, if not, executing S15;
[0048] S15, determine whether the predetermined number of iterations is reached, if not, execute S12, if yes, complete clustering.
[0049] Specifically, for each position, a cluster center whose physical distance from the cluster center is less than a preset physical distance threshold is used as a candidate cluster center.
[0050] The clustering distance between the location and each candidate cluster center is calculated based on the similarity of historical communication demand fluctuations between the location and each candidate cluster center according to the following formula:
[0051]
[0052] Among them, D (i,f) is the clustering distance between the ith position and the center of the fth candidate cluster, M2 is a preset parameter, and M2>0, S (i,f) is the historical communication demand fluctuation similarity between the ith location and the fth candidate cluster center.
[0053] The mean similarity of historical communication demand fluctuations corresponding to each location included in the location cluster is calculated according to the following formula:
[0054]
[0055] Among them, S (g,mean) is the mean similarity of historical communication demand fluctuations corresponding to the g-th location included in the location cluster, S (g, h ) is the similarity of historical communication demand fluctuations between the g-th location and the h-th location included in the location cluster, and H is the total number of locations included in the location cluster.
[0056] In some embodiments, predicting the future communication demand of each sub-region based on the historical communication demand information of each sub-region at multiple historical time points includes:
[0057] For any two sub-areas, based on historical communication demand information of the two sub-areas at multiple historical time points, a correlation coefficient between the two sub-areas is calculated;
[0058] For each sub-region, based on the correlation coefficient between any two sub-regions, determining the associated sub-region of each sub-region;
[0059] For each sub-area, the future communication demand of the sub-area is predicted through a demand prediction model based on the historical communication demand information of the sub-area at multiple historical time points and the historical communication demand information of the associated sub-areas of the sub-area at multiple historical time points, wherein the demand prediction model can be a Convolutional Neural Network (CNN) model.
[0060] For example, the correlation coefficient between two sub-regions can be calculated based on the following formula:
[0061]
[0062] Among them, r(x,y) is the correlation coefficient between the x-th sub-region and the y-th sub-region, Dx is the historical communication demand information of the x-th sub-region at multiple historical time points, D y is the historical communication demand information of the y-th sub-region at multiple historical time points, COV(D x ,D y ) is the covariance of the historical communication demand information of the x-th sub-region at multiple historical time points and the historical communication demand information of the y-th sub-region at multiple historical time points, Var(D x ) is the variance of the historical communication demand information of the x-th sub-region at multiple historical time points, Var(D y ) is the variance of the historical communication demand information of the y-th sub-region at multiple historical time points.
[0063] Step 113: Set a status reporting component at each network element.
[0064] The status reporting component may be used to collect information about the installation environment of the network element, and the status reporting component may include a variety of sensors.
[0065] Step 114, set up multiple edge computing components, where one edge computing component is used to receive heartbeat data reported by at least one network element and environmental information reported by at least one status reporting component, and determine the status of the network element based on the heartbeat data reported by the network element and the environmental information reported by the status reporting component set at the network element, and report the status of the network element to the network manager.
[0066] In some embodiments, the heartbeat data includes at least CPU temperature, CPU usage, physical memory usage, hard disk read and write data flow, network in and out flow, load, throughput, delay and packet loss rate. Environmental information includes at least ambient temperature, ambient humidity and ambient electromagnetic interference.
[0067] In some embodiments, the edge computing component determines the state of the network element based on the heartbeat data reported by the network element and the environmental information reported by the state reporting component set at the network element, including:
[0068] Determine the operation abnormality score of the network element based on the CPU temperature, CPU utilization, physical memory utilization, hard disk read and write data flow, network inbound and outbound flow, load, throughput, delay and packet loss rate of the network element through the first state determination model;
[0069] Determine the environmental anomaly score of the network element based on the environmental temperature, environmental humidity and environmental electromagnetic interference of the network element through the second state determination model, wherein both the first state determination model and the second state determination model can be convolutional neural network (CNN) models;
[0070] The state of the network element is determined based on the operation anomaly score and the environment anomaly score of the network element. For example, the operation anomaly score and the environment anomaly score of the network element may be weighted summed to determine the abnormal state score of the network element.
[0071] Step 115: determine a plurality of candidate network elements based on the target area.
[0072] For example, a network element whose communication range covered by the network element overlaps with the target area is selected as a candidate network element.
[0073] Step 116: The network manager performs a first screening of multiple candidate network elements based on the status of each candidate network element to determine multiple first candidate network elements.
[0074] For example, a candidate network element whose abnormal state score is less than a preset abnormal state score threshold is used as the first candidate network element.
[0075] Step 117: The network management sends trigger information to the edge computing component corresponding to each first candidate network element.
[0076] Step 118: The edge computing component interacts the trigger information to the corresponding first candidate network element.
[0077] Step 119: The first candidate network element sends a link establishment request to the network manager based on the trigger information, wherein the link establishment request includes the version number of the first candidate network element.
[0078] Step 120: The network manager performs a second screening on the multiple first candidate network elements based on the link establishment request sent by each first candidate network element, and determines multiple second candidate network elements.
[0079] Specifically include:
[0080] The network management performs a compatibility test on each first candidate network element according to the version number of each first candidate network element, and obtains a compatibility test result;
[0081] According to the compatibility test result, a second screening is performed on the plurality of first candidate network elements to determine a plurality of second candidate network elements. Specifically, compatibility exists among the plurality of second candidate network elements after the screening.
[0082] Step 121: The network management establishes a network based on multiple second candidate network elements and future communication demand information of the target area.
[0083] Specifically include:
[0084] The network management determines the network element density of each sub-area at multiple future time points based on the future communication demand of each sub-area. Specifically, the higher the future communication demand of the sub-area at a certain future time point, the higher the network element density of the sub-area at the future time point.
[0085] The network management establishes a dynamic network based on the network element density of each sub-area at multiple future time points and multiple second candidate network elements.
[0086] In some embodiments, the network management establishes a dynamic network based on the network element density of each sub-area at multiple future time points and multiple second candidate network elements, including:
[0087] The network management establishes multiple candidate dynamic networks based on the network element density of each sub-area at multiple future time points and multiple second candidate network elements, wherein the candidate dynamic networks may include networks at multiple future time points;
[0088] Establishing a network screening system, wherein the network screening system may include multiple network screening indicators, such as coverage indicator, data transmission rate indicator, power consumption indicator, delay indicator, packet loss rate indicator, bandwidth utilization indicator, etc.;
[0089] Based on the network screening system, multiple candidate dynamic networks are screened to determine the dynamic network.
[0090] Specifically, for each candidate dynamic network, the scores of the candidate dynamic network in multiple networking screening indicators can be determined, and the scores of the candidate dynamic network in multiple networking screening indicators can be weighted and summed to determine the matching score of the candidate dynamic network, and the candidate dynamic network with the largest matching score can be used as the dynamic network.
[0091] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.
Claims
1. A method for adapting a domestic chip network element to a network management system, characterized in that: include: Obtain historical communication demand information for the target area; Predicting future communication demand information of the target area based on historical communication demand information of the target area; A status reporting component is provided at each network element; Setting multiple edge computing components, wherein one edge computing component is used to receive heartbeat data reported by at least one network element and environmental information reported by at least one status reporting component, and determine the status of the network element based on the heartbeat data reported by the network element and the environmental information reported by the status reporting component set at the network element, and report the status of the network element to the network management; Determine a plurality of candidate network elements based on the target area; The network manager performs a first screening on the multiple candidate network elements based on the status of each of the candidate network elements to determine multiple first candidate network elements; The network manager sends trigger information to the edge computing component corresponding to each of the first candidate network elements; The edge computing component exchanges the trigger information to the corresponding first candidate network element; The first candidate network element sends a link establishment request to the network manager based on the trigger information, wherein the link establishment request includes a version number of the first candidate network element; The network manager performs a second screening on the multiple first candidate network elements based on the link establishment request sent by each of the first candidate network elements to determine multiple second candidate network elements; The network management establishes a network based on the multiple second candidate network elements and the future communication demand information of the target area; Predicting future communication demand information of the target area based on historical communication demand information of the target area includes: Acquire historical communication demand information of multiple locations in the target area at multiple historical time points; Clustering the multiple locations based on historical communication demand information of each of the locations at multiple historical time points to determine multiple sub-areas of the target area; For each of the sub-regions, fitting the historical communication demand information of the sub-region at multiple historical time points based on the least squares method to determine the historical communication demand information of the sub-region at multiple historical time points; Predicting future communication demand of each of the sub-areas based on historical communication demand information of each of the sub-areas at multiple historical time points; Clustering the multiple locations based on historical communication demands of each of the locations at multiple historical time points includes: For each of the locations, determining historical communication demand fluctuation parameters of the location in multiple historical time periods based on historical communication demand information of the location at multiple historical time points; For any two of the locations, calculating the physical distance between the two locations and the similarity of historical communication demand fluctuations; Clustering the plurality of locations based on the physical distance between any two of the locations and the similarity of historical communication demand fluctuations by using a K-means clustering algorithm; Clustering the multiple locations based on the physical distance between any two of the locations and the similarity of historical communication demand fluctuations by using a K-means clustering algorithm includes: S11, randomly select K positions as initial cluster centers, and execute S12; S12, for each of the positions, based on the physical distance between the position and each of the cluster centers, determine a candidate cluster center, based on the similarity of historical communication demand fluctuations between the position and each of the candidate cluster centers, calculate the cluster distance between the position and each of the candidate cluster centers, and assign the position to the candidate cluster center with the shortest cluster distance, and execute S13; S13, for each location cluster, calculating the mean similarity of historical communication demand fluctuations corresponding to each location included in the location cluster, taking the location with the largest mean similarity of historical communication demand fluctuations as the new cluster center, and executing S14; S14, determining whether the cluster center determined by the current clustering is different from the cluster center determined by the previous clustering, if so, executing S12, if not, executing S15; S15, determining whether the predetermined number of iterations has been reached, if not, executing S12, if yes, completing clustering; Predicting future communication demand of each sub-area based on historical communication demand information of each sub-area at multiple historical time points includes: For any two of the sub-areas, calculating a correlation coefficient between the two sub-areas based on historical communication demand information of the two sub-areas at multiple historical time points; For each of the sub-regions, determining a correlated sub-region of each of the sub-regions based on a correlation coefficient between any two of the sub-regions; For each of the sub-areas, future communication demand of the sub-area is predicted by a demand prediction model based on historical communication demand information of the sub-area at multiple historical time points and historical communication demand information of associated sub-areas of the sub-area at multiple historical time points.
2. According to claim 1, a method for adapting a domestic chip network element to a network management system is characterized in that: The heartbeat data includes at least CPU temperature, CPU usage, physical memory usage, hard disk read and write data flow, network inbound and outbound flow, load, throughput, delay and packet loss rate; The environmental information at least includes environmental temperature, environmental humidity and environmental electromagnetic interference.
3. The method for adapting a domestic chip network element to a network management system according to claim 2, characterized in that: The edge computing component determines the state of the network element based on the heartbeat data reported by the network element and the environmental information reported by the state reporting component set at the network element, including: Determine the operation abnormality score of the network element based on the CPU temperature, CPU utilization, physical memory utilization, hard disk read and write data flow, network inbound and outbound flow, load, throughput, delay and packet loss rate of the network element through the first state determination model; Determine the environmental anomaly score of the network element based on the environmental temperature, environmental humidity and environmental electromagnetic interference of the network element through a second state determination model; The state of the network element is determined based on the operation anomaly score and the environment anomaly score of the network element.
4. The method for adapting a domestic chip network element to a network management system according to claim 1, characterized in that: The network manager performs a second screening on the multiple first candidate network elements based on the link establishment request sent by each of the first candidate network elements to determine multiple second candidate network elements, including: The network manager performs a compatibility test on each of the first candidate network elements according to the version number of each of the first candidate network elements, and obtains a compatibility test result; According to the compatibility test result, the plurality of first candidate network elements are screened for the second time to determine a plurality of second candidate network elements.
5. The method for adapting a domestic chip network element to a network management system according to claim 1, characterized in that: The network management establishes a network based on the multiple second candidate network elements and the future communication demand information of the target area, including: The network management determines the network element density of each of the sub-areas at multiple future time points based on the future communication demand of each of the sub-areas; The network management establishes a dynamic network based on the network element density of each of the sub-areas at multiple future time points and the multiple second candidate network elements.
6. The method for adapting a domestic chip network element to a network management system according to claim 5, characterized in that: The network management establishes a dynamic network based on the network element density of each sub-area at multiple future time points and the multiple second candidate network elements, including: The network manager establishes a plurality of candidate dynamic networks based on the network element density of each of the sub-areas at a plurality of future time points and the plurality of second candidate network elements; Establish a network screening system; Based on the network screening system, the multiple candidate dynamic networks are screened to determine the dynamic network.
Citation Information
Patent Citations
Network quality prediction method, service adjustment method, related equipment and storage medium
CN109756911A
Management and control method, network system, equipment and storage medium
CN113300865A
