Intelligent communication data management system and method based on 5G communication

Abstract

The application discloses an intelligent communication data management system and method based on 5G communication, and relates to the technical field of data management.The system comprises a data acquisition and analysis module, an abnormal event identification module, a load index trend analysis module, a real-time monitoring and early warning module and a path adjustment and optimization module; the data acquisition and analysis module collects communication and log data and constructs an abnormal event database; the abnormal event identification module extracts an edge computing node path of an abnormal event, calculates a radiation area and a load index of the abnormal event; the load index trend analysis module analyzes a load index trend curve, calculates a difference index, and matches an adjustment scheme according to the difference index; the real-time monitoring and early warning module collects real-time data, calculates an abnormal evaluation index, and outputs early warning information; and the path adjustment and optimization module calculates a difference vector according to the matched adjustment scheme, carries out similarity calculation on the difference vector and a difference vector of a current path, and finally outputs an adjustment scheme.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data management, in particular to an intelligent communication data management system and method based on 5G communication. BACKGROUND

[0002] With the rapid development of information technology and the increasing demand for data communication, traditional data management systems have gradually exposed their shortcomings in handling large-scale communication data. These shortcomings include slow processing speed, high latency, uneven resource allocation, and other issues, which seriously affect the efficiency of communication data and the intelligent level of the system. Especially in the 5G era, the high speed and low latency characteristics of communication networks make the traditional data management method even more unsuitable, and new solutions are needed to fully exploit the advantages of 5G technology.

[0003] 5G technology, with its significantly improved bandwidth and extremely low latency, has become a key driving force for modern communication networks. It not only supports higher data transmission rates, but also provides higher network reliability and lower latency, capable of meeting the needs of large-scale data transmission and real-time processing. However, how to effectively manage and utilize these high-speed, high-volume data poses new challenges to communication network systems.

[0004] Existing communication data management systems mostly use static resource allocation strategies and more traditional data processing methods, which are difficult to meet the requirements of real-time, flexibility and intelligence in the 5G network environment. Traditional systems usually lack real-time monitoring and dynamic adjustment capabilities for data traffic, and lack effective intelligent analysis and automated scheduling mechanisms. This leads to the system being difficult to respond quickly when the network load changes, low resource utilization efficiency, and insufficient ability to respond to data traffic fluctuations. SUMMARY

[0005] The purpose of the present application is to provide an intelligent communication data management system and method based on 5G communication to solve the problems raised in the background technology.

[0006] In order to solve the above technical problems, the present application provides the following technical solutions:

[0007] An intelligent communication data management method based on 5G communication, the method comprising the following steps:

[0008] Step S100. Deploy edge computing nodes at the network edge, connect the 5G network by configuring the edge computing nodes, collect communication data of the edge computing nodes; analyze the collected communication data in combination with log data, identify the corresponding abnormal events, and build an abnormal event database based on the identification results;

[0009] Step S200. According to the abnormal event database, obtain the edge computing node path corresponding to all abnormal events, and obtain the radiation area of the abnormal event based on the edge computing node path; according to the radiation area of the abnormal event, evaluate the load index of the edge computing node in the radiation area in the edge computing node path corresponding to the abnormal event;

[0010] Step S300. Analyze the load index trend curve of the edge computing node corresponding to the abnormal event, and obtain the load index trend curve of the edge computing node corresponding to the corresponding normal event combined with the log data; analyze the difference index between the load index trend curves corresponding to the abnormal event and the normal event, and match the difference index with the edge computing node path adjustment scheme corresponding to the corresponding abnormal event;

[0011] Step S400. Collect real-time communication data corresponding to the edge computing node, and obtain the corresponding current edge computing node path; match the abnormal event in the abnormal event database combined with the real-time communication data and the current edge computing node path, calculate the difference index of the current edge computing node path, and adjust the current edge computing node path based on the difference index.

[0012] Further, step S100 includes:

[0013] S101. The communication data includes traffic data, user behavior data and device state data; the log data refers to detailed records of edge computing node running state, event occurrence information and system operation; the communication data of the edge computing node in the selected time period is obtained, and the communication data of each edge computing node is analyzed discretely, so that each time stamp corresponds to a communication data segment;

[0014] S102. Obtain the log data in the selected time period, and according to the time recorded by the log data, correspond the communication data of each edge computing node with the corresponding log data, and according to the corresponding time stamp, correspond the communication data segment of each edge computing node with the corresponding log data segment, and take the corresponding communication data segment and log data segment as a data unit;

[0015] S103. Compare and analyze all data units of each edge computing node with the corresponding standard data unit respectively, so as to obtain the corresponding abnormal evaluation index Y, and the specific calculation formula is:

[0016] Y=(1 / n)Σ i∈[1,n] di / si,

[0017] wherein, n represents the number of data units corresponding to the edge computing node, di represents the similarity between the i th data unit and the corresponding standard data unit feature, and si represents the standard feature value of the i th standard data unit; the abnormality evaluation index Y is compared with a preset threshold Y0, the edge computing node in the data unit with the abnormality evaluation index Y greater than or equal to the preset threshold Y0 is taken as an abnormal node, and the corresponding abnormal event is identified according to the corresponding log data, so as to construct an abnormal event database.

[0018] Further, the step S200 comprises:

[0019] S201. According to the abnormal event database, for the abnormal events in the abnormal event database, the corresponding abnormal event records are extracted, and the edge computing node path corresponding to the abnormal event is obtained according to the edge computing node identifier associated with the corresponding abnormal event in the abnormal event record; for each abnormal event, the abnormal node Pj and the corresponding abnormality evaluation index Yj are extracted from the corresponding edge computing node path, and the influence range radius Rj is calculated according to the abnormality evaluation index of the abnormal node Pj, and Rj=k×Yj, wherein k represents the influence range coefficient, and Yj represents the abnormality evaluation index corresponding to the j th abnormal node;

[0020] S202. For each abnormal node Pj and the corresponding influence range radius Rj, a circular region is generated, and the circular region is represented as C(Pj, Rj), which represents a circular region with Pj as the center and Rj as the radius; for all circular regions in the same edge computing node path, the union region is calculated, so as to obtain the radiation region of the abnormal event;

[0021] S203. For each edge computing node Nt in the radiation region, the coverage CDt in the radiation region is calculated, and the specific calculation formula is: CDt=S(Nt, C) / C(Nt, Rt), wherein S(Nt, C) represents the intersection area of the influence range area of the edge computing node Nt and the corresponding radiation region, and C(Nt, Rt) represents the influence range area of the edge computing node Nt, and the calculation process of C(Nt, Rt) is consistent with that of C(Pj, Rj); for the edge computing node Nt in the radiation region, the corresponding load index FLIt is calculated, and the specific calculation formula is:

[0022] FLIt=LIt×(α×CDt+β×Y),

[0023] Wherein, LIt represents the initial load index of the edge computing node, and a and β represent weight factors.

[0024] Further, the step S300 comprises:

[0025] S301. Extract the edge computing nodes in the radiation area corresponding to the abnormal event, and obtain the load index trend curve A of the corresponding edge computing nodes in the selected time period; according to the log data, obtain the initial load index of the same edge computing node path in different time periods, thereby forming the load index trend curve B of the edge computing node of the normal event;

[0026] S302. Calculate the difference index between the load index trend curves of the abnormal event and the normal event, the difference index including mean square error MSE, mean absolute error MAE and correlation coefficient p; according to the above difference index, form the corresponding difference vector V, and V = [MSE, MAE, p], and match the difference vector V with the edge computing node path adjustment scheme corresponding to the abnormal event, and save it to the abnormal event database.

[0027] Further, step S400 includes:

[0028] S401. Collect real-time communication data corresponding to the edge computing node once every selected time period, analyze the real-time communication data, and thereby obtain the current edge computing node path; based on the current edge computing node path, calculate the abnormal evaluation index Y1 of each edge computing node of the current edge computing node path, and compare the abnormal evaluation index Y1 with the preset threshold Y0, if the abnormal evaluation index Y1 of each edge computing node of the current edge computing node path is less than the preset threshold Y0, no warning information is output;

[0029] S402. If there is an edge computing node of the current edge computing node path whose abnormal evaluation index Y1 is greater than or equal to the preset threshold Y0, output a warning information; and according to the current edge computing node path and the corresponding abnormal node, search in the abnormal event database, thereby matching the corresponding abnormal event; according to the abnormal node of the current edge computing node path, calculate the corresponding load index and obtain the corresponding load index trend curve A1, based on the load index trend curve A1 calculate the corresponding difference index, thereby forming the difference vector V1, and V1 = [MSE1, MAE1, p1];

[0030] S403. Obtain the difference vector V of the matched abnormal event, calculate the similarity between the difference vector V and the difference vector V1, take the one with the largest similarity as the final matching event, and output the edge computing node path adjustment scheme corresponding to the matching event as the reference adjustment scheme, and the system adjusts the current edge computing node path according to the reference adjustment scheme.

[0031] An intelligent communication data management system based on 5G communication, the system comprising: a data acquisition and analysis module, an abnormal event identification module, a load index trend analysis module, a real-time monitoring and warning module, and a path adjustment and optimization module.

[0032] The data collection and analysis module deploys a computing node at the network edge, collects communication data and log data, and performs discrete analysis to identify abnormal events and build an abnormal event database; the abnormal event identification module extracts the edge computing node path of the abnormal event based on the abnormal event database, calculates the radiation area and load index of the abnormal event; the load index trend analysis module analyzes the load index trend curves of abnormal events and normal events, calculates the difference index, and matches the adjustment scheme of the edge computing node path based on the difference index; the real-time monitoring and early warning module collects real-time communication data, calculates an abnormality evaluation index, compares it with a preset threshold, and outputs early warning information, and matches the abnormal event in the abnormal event database through real-time data; the path adjustment and optimization module calculates the difference vector based on the matched adjustment scheme of the abnormal event, performs similarity calculation with the difference vector of the current edge computing node path, and finally outputs the adjustment scheme to optimize the edge computing node path.

[0033] Further, the data collection and analysis module includes a data collection unit and a data analysis unit;

[0034] The data collection unit is responsible for collecting communication data from the edge computing node, including traffic data, user behavior data and device status data, collecting running logs from the edge computing node, including node running state, event occurrence information and system operation record; the data analysis unit divides the collected communication data into multiple data segments according to the timestamp, and pairs it with the corresponding log data to form a data unit, compares all data units of each edge computing node with the standard data unit, calculates the abnormality evaluation index, compares the abnormality evaluation index with the preset threshold, identifies the abnormal node and builds the abnormal event database.

[0035] Further, the abnormal event identification module includes an abnormal event record extraction unit, an abnormal node influence range calculation unit, and a node load calculation unit in the radiation area;

[0036] The abnormal event record extraction unit extracts the abnormal event record from the abnormal event database and extracts the associated edge computing node identifier to obtain the edge computing node path corresponding to the abnormal event; the abnormal node influence range calculation unit calculates the influence range radius according to the abnormality evaluation index of the abnormal node, generates a circular area to represent the influence range, and calculates the union of all circular areas to obtain the radiation area of the abnormal event; the node load calculation unit in the radiation area calculates the coverage and load index of each edge computing node in the radiation area, and updates the database of the abnormal event according to the calculation results.

[0037] Further, the load index trend analysis module includes a load index trend curve extraction unit and a load index difference analysis unit;

[0038] The load index trend curve extraction unit extracts the load index trend curves of the edge computing nodes corresponding to the abnormal events and normal events, respectively obtains the load index trend curve of the abnormal events and the load index trend curve of the normal events; the load index difference analysis unit calculates the difference indexes between the load index trend curves of the abnormal events and the normal events, including the mean square error, the mean absolute error and the correlation coefficient, generates a difference vector and matches the adjustment scheme of the abnormal events.

[0039] Further, the real-time monitoring and early warning module includes a real-time data acquisition unit, a real-time abnormality evaluation unit and a real-time early warning and matching unit;

[0040] The real-time data acquisition unit periodically acquires real-time communication data of the edge computing nodes and obtains the current edge computing node path; the real-time abnormality evaluation unit calculates an abnormality evaluation index based on the current edge computing node path, compares the abnormality evaluation index with a preset threshold value, and judges whether the prewarning information needs to be output according to the comparison result; the real-time early warning and matching unit outputs the prewarning information when it is detected that the abnormality evaluation index is greater than or equal to the preset threshold value, matches the abnormal events according to the current path and the records in the abnormal event database, calculates the corresponding load index trend curves and difference indexes, and generates the corresponding difference vector;

[0041] The path adjustment and optimization module includes an adjustment scheme matching unit and a path optimization execution unit;

[0042] The adjustment scheme matching unit acquires the difference vector of the abnormal events, performs similarity calculation on the difference vector obtained by the real-time monitoring, and selects the one with the highest similarity as the final matching event; the path optimization execution unit outputs the edge computing node path adjustment scheme corresponding to the matching event as a reference adjustment scheme, and the system adjusts the current edge computing node path according to the reference adjustment scheme.

[0043] Compared with the prior art, the present application has the following advantages:

[0044] The application can dynamically monitor the network state by deploying edge computing nodes at the network edge and collecting communication data and log data in real time. This real-time data collection and analysis mechanism significantly improves the response speed to network load changes, enabling the system to identify and handle abnormal events in a timely manner, improving the real-time performance and agility of the network. The application introduces an abnormal event database and intelligent analysis mechanism, which can intelligently identify and predict potential problems by analyzing load index trends, difference indicators and other information of abnormal events. This intelligent analysis not only effectively improves the detection accuracy of abnormal events, but also provides targeted adjustment solutions to optimize the allocation and use of network resources. Unlike traditional static resource allocation strategies, the application's dynamic adjustment mechanism based on real-time communication data and load indices can automatically adjust resource allocation according to the current network state. This dynamic adjustment capability improves the flexibility of the network, ensuring that the system can quickly respond when load fluctuations or abnormal events occur, thereby improving the stability and reliability of the overall network. The application can more accurately assess the load of edge computing nodes by calculating load indices and coverage. Combined with the influence range and radiation area of abnormal events, the load of edge computing nodes can be comprehensively analyzed to optimize resource allocation. This optimization method can effectively avoid resource waste and improve the efficiency of network resource utilization. Through continuous monitoring and analysis of real-time communication data, the application can provide accurate early warning at the initial stage of abnormal events to prevent potential problems from expanding. At the same time, based on the matching mechanism of historical data and real-time data, the most suitable adjustment scheme can be provided for the current edge computing node path, thereby effectively improving network performance and reducing maintenance costs. BRIEF DESCRIPTION OF DRAWINGS

[0045] The accompanying drawings are included to provide a further understanding of the application, and constitute a part of the specification, which together with the embodiments of the application, serve to explain the application, and do not constitute a limitation on the application. In the drawings:

[0046] Figure 1 is a schematic diagram of an intelligent communication data management system module based on 5G communication. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the application, not all embodiments. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of the application.

[0048] Please refer to Figure 1 The application provides technical solutions:

[0049] The application discloses an intelligent communication data management system based on 5G communication, which comprises a data collection and analysis module, an abnormal event identification module, a load index trend analysis module, a real-time monitoring and early warning module and a path adjustment and optimization module.

[0050] The data collection and analysis module deploys a computing node at the network edge, collects communication data and log data, and performs discrete analysis to identify abnormal events and construct an abnormal event database; the abnormal event identification module extracts the edge computing node path of the abnormal event based on the abnormal event database, calculates the radiation area and load index of the abnormal event; the load index trend analysis module analyzes the load index trend curves of abnormal events and normal events, calculates a difference index, and matches an adjustment scheme for the edge computing node path based on the difference index; the real-time monitoring and early warning module collects real-time communication data, calculates an abnormality evaluation index, compares it with a preset threshold, and outputs early warning information, and matches the abnormal event in the abnormal event database through real-time data; the path adjustment and optimization module calculates a difference vector based on the matched adjustment scheme of the abnormal event, performs similarity calculation on the difference vector of the current edge computing node path, and finally outputs an adjustment scheme to optimize the edge computing node path.

[0051] The data collection and analysis module comprises a data collection unit and a data analysis unit.

[0052] The data collection unit is responsible for collecting communication data from the edge computing node, and the communication data comprises traffic data, user behavior data and device status data; the data collection unit collects the running log from the edge computing node, including the node running state, event occurrence information and system operation record; the data analysis unit divides the collected communication data into multiple data segments according to the time stamp, and pairs the data segments with the corresponding log data to form a data unit; the data analysis unit compares all data units of each edge computing node with a standard data unit, calculates an abnormality evaluation index, compares the abnormality evaluation index with a preset threshold, identifies abnormal nodes and constructs an abnormal event database.

[0053] The abnormal event identification module comprises an abnormal event record extraction unit, an abnormal node influence range calculation unit and a node load calculation unit in the radiation area.

[0054] The abnormal event record extraction unit extracts the abnormal event record from the abnormal event database and extracts the associated edge computing node identifier to obtain the edge computing node path corresponding to the abnormal event; the abnormal node influence range calculation unit calculates the influence range radius according to the abnormality evaluation index of the abnormal node, generates a circular area to represent the influence range, and calculates the union of all circular areas to obtain the radiation area of the abnormal event; the node load calculation unit in the radiation area calculates the coverage and load index of each edge computing node in the radiation area, and updates the database of the abnormal event according to the calculation results.

[0055] The load index trend analysis module includes a load index trend curve extraction unit and a load index difference analysis unit.

[0056] The load index trend curve extraction unit extracts the load index trend curve of the edge computing node corresponding to the abnormal event and the normal event, respectively obtains the load index trend curve of the abnormal event and the load index trend curve of the normal event; the load index difference analysis unit calculates the difference index between the load index trend curves of the abnormal event and the normal event, including mean square error, mean absolute error and correlation coefficient, generates a difference vector and matches the adjustment scheme of the abnormal event.

[0057] The real-time monitoring and early warning module includes a real-time data acquisition unit, a real-time abnormality evaluation unit, and a real-time early warning and matching unit.

[0058] The real-time data acquisition unit periodically acquires real-time communication data of the edge computing node and obtains the current edge computing node path; the real-time abnormality evaluation unit calculates an abnormality evaluation index based on the current edge computing node path, compares it with a preset threshold, and determines whether to output a warning information according to the comparison result; the real-time early warning and matching unit outputs a warning information when detecting that the abnormality evaluation index is greater than or equal to the preset threshold, matches the abnormal event according to the current path and the record in the abnormal event database, calculates the corresponding load index trend curve and difference index, and generates the corresponding difference vector;

[0059] The adjustment scheme matching unit obtains the difference vector of the abnormal event, calculates the similarity with the difference vector obtained by real-time monitoring, and selects the one with the highest similarity as the final matching event; the path optimization execution unit outputs the edge computing node path adjustment scheme corresponding to the matching event as a reference adjustment scheme, and the system adjusts the current edge computing node path according to the reference adjustment scheme.

[0060] An intelligent communication data management method based on 5G communication, the method comprising the following steps:

[0061] Step S100. Deploy an edge computing node at the network edge, connect the 5G network by configuring the edge computing node, collect the communication data of the edge computing node; analyze the collected communication data in combination with the log data, identify the corresponding abnormal event, and build an abnormal event database based on the identification result;

[0062] Step S200. According to the abnormal event database, obtain the edge computing node path corresponding to all abnormal events, and obtain the radiation area of the abnormal event based on the edge computing node path; evaluate the load index of the edge computing node in the radiation area of the abnormal event corresponding to the edge computing node path.

[0063] Step S300. Analyze the load index trend curve of the edge computing node corresponding to the abnormal event, and obtain the load index trend curve of the edge computing node corresponding to the normal event in combination with the log data; analyze the difference index between the load index trend curves corresponding to the abnormal event and the normal event, and match the difference index with the edge computing node path adjustment scheme corresponding to the abnormal event;

[0064] Step S400. Collect real-time communication data corresponding to the edge computing node, and obtain the current edge computing node path; match the abnormal event in the abnormal event database in combination with the real-time communication data and the current edge computing node path, calculate the difference index of the current edge computing node path, and adjust the current edge computing node path based on the difference index.

[0065] Step S100 includes:

[0066] S101. The communication data includes traffic data, user behavior data and device state data; the log data refers to detailed records of edge computing node running state, event occurrence information and system operation; the communication data of the edge computing node in the selected time period is obtained, and the communication data of each edge computing node is analyzed discretely, so that each time stamp corresponds to a communication data segment;

[0067] S102. Obtain the log data in the selected time period, and correspond the communication data of each edge computing node with the corresponding log data according to the time recorded by the log data, and correspond the communication data segment of each edge computing node with the corresponding log data segment according to the corresponding time stamp, and take the corresponding communication data segment and log data segment as a data unit;

[0068] S103. Compare and analyze all data units of each edge computing node with the corresponding standard data unit respectively, so as to obtain the corresponding abnormal evaluation index Y, and the specific calculation formula is:

[0069] Y = (1 / n)∑ i∈[1,n] di / si,

[0070] Wherein, n represents the number of data units corresponding to the edge computing node, di represents the similarity between the i-th data unit and the corresponding standard data unit feature, and si represents the standard characteristic value of the i-th standard data unit; compare the abnormal evaluation index Y with the preset threshold Y0, and take the edge computing node in the data unit whose abnormal evaluation index Y is greater than or equal to the preset threshold Y0 as an abnormal node, and identify the corresponding abnormal event according to the corresponding log data, so as to construct an abnormal event database.

[0071] Step S200 comprises:

[0072] S201. According to the abnormal event database, for the abnormal event in the abnormal event database, the corresponding abnormal event record is extracted, and the edge computing node path corresponding to the abnormal event is obtained according to the edge computing node identifier associated with the corresponding abnormal event in the abnormal event record; for each abnormal event, the abnormal node Pj and the corresponding abnormal evaluation index Yj are extracted from the corresponding edge computing node path, and the influence range radius Rj is calculated according to the abnormal evaluation index of the abnormal node Pj, and Rj=k×Yj, wherein k represents the influence range coefficient; Yj represents the abnormal evaluation index corresponding to the jth abnormal node;

[0073] S202. For each abnormal node Pj and the corresponding influence range radius Rj, a circular area is generated, and the circular area is represented as C(Pj, Rj), which represents a circular area with Pj as the center and Rj as the radius; for all circular areas in the same edge computing node path, the union area of the circular areas is calculated, thereby obtaining the radiation area of the abnormal event;

[0074] S203. For each edge computing node Nt in the radiation area, the coverage CDt of the edge computing node Nt in the radiation area is calculated, and the specific calculation formula is: CDt=S(Nt, C) / C(Nt, Rt), wherein S(Nt, C) represents the intersection area of the influence range area of the edge computing node Nt and the corresponding radiation area, and C(Nt, Rt) represents the influence range area of the edge computing node Nt, and the calculation process of C(Nt, Rt) is consistent with the calculation process of C(Pj, Rj); for the edge computing node Nt in the radiation area, the corresponding load index FLIt is calculated, and the specific calculation formula is:

[0075] FLIt=LIt×(α×CDt+β×Y),

[0076] Wherein, LIt represents the initial load index of the edge computing node, and α and β represent weight factors.

[0077] In this embodiment, it is assumed that a certain edge computing node path has two abnormal nodes, and the corresponding abnormal node identifiers and the abnormal evaluation indexes thereof are:

[0078] Node P1: Y=1.5, Node P2: Y=2.0;

[0079] It is assumed that k=4, including nodes Nt1, Nt2, Nt3, and the initial load index (LIt) and the influence range data are known.

[0080] The influence range radius Rj (assuming the unit is km) is calculated:

[0081] Abnormality evaluation index of node P1 (Y1=1.5), R_1=4x1.5=6(km);

[0082] Abnormality evaluation index of node P2 (Y2=2.0, R_2=4x2.0=8(km);

[0083] Generate a circular region:

[0084] Circular region C(P1, R1) of node P1 and circular region C(P2, R2) of node P2; calculate the union region of C(P1, R1) and C(P2, R2), and calculate the coverage and load index according to the formula.

[0085] Step S300 includes:

[0086] S301. Extract the edge computing node in the radiation region corresponding to the abnormal event, and obtain the load index trend curve A of the corresponding edge computing node in the selected time period; according to the log data, obtain the initial load index of the same edge computing node path in different time periods, so as to constitute the load index trend curve B of the edge computing node of the normal event;

[0087] S302. Calculate the difference index between the load index trend curves of the abnormal event and the normal event, and the difference index includes mean square error MSE, mean absolute error MAE and correlation coefficient p; according to the above difference index, a corresponding difference vector V is constituted, and V=[MSE, MAE, p], and the difference vector V is matched with the edge computing node path adjustment scheme corresponding to the abnormal event, and saved to the abnormal event database.

[0088] Step S400 includes:

[0089] S401. Every selected time period, collect real-time communication data corresponding to the edge computing node once, analyze the real-time communication data, so as to obtain the current edge computing node path; based on the current edge computing node path, calculate the abnormality evaluation index Y1 of each edge computing node of the current edge computing node path, and compare the abnormality evaluation index Y1 with the preset threshold Y0, if the abnormality evaluation index Y1 of each edge computing node of the current edge computing node path is less than the preset threshold Y0, no warning information is outputted;

[0090] S402. If the abnormality evaluation index Y1 of the edge computing node of the current edge computing node path is greater than or equal to the preset threshold Y0, an early warning information is output; and according to the current edge computing node path and the corresponding abnormal node, a search is performed in the abnormal event database, so that a corresponding abnormal event is matched; according to the abnormal node of the current edge computing node path, a corresponding load index is calculated and a corresponding load index trend curve A1 is obtained, a corresponding difference index is calculated based on the load index trend curve A1, so as to form a difference vector V1, and V1 = [MSE1, MAE1, p1];

[0091] S403. The difference vector V of the matched abnormal event is obtained, the difference vector V and the difference vector V1 are similarity calculated, the maximum similarity is taken as the final matching event, and the adjustment scheme of the matching event corresponding to the edge computing node path is taken as the reference adjustment scheme and output, and the system adjusts the current edge computing node path according to the reference adjustment scheme.

[0092] It should be noted that in this paper, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment.

[0093] Finally, it should be noted that: the above only describes the preferred embodiments of the present application, and does not limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or make equivalent replacements to some technical features. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent communication data management method based on 5G communication, characterized in that: The method comprises the following steps: Step S100. Deploying an edge computing node at a network edge, connecting the 5G network by configuring the edge computing node, collecting communication data of the edge computing node; combining log data, analyzing the collected communication data to identify corresponding abnormal events, and constructing an abnormal event database based on the identification result; Step S200. According to the abnormal event database, the edge computing node path corresponding to all abnormal events is obtained, and the radiation area of the abnormal event is obtained based on the edge computing node path; according to the radiation area of the abnormal event, the load index of the edge computing node in the radiation area in the edge computing node path corresponding to the abnormal event is evaluated; The step S200 comprises: S201. According to the abnormal event database, the corresponding abnormal event record is extracted for the abnormal event in the abnormal event database, and the edge computing node path corresponding to the abnormal event is obtained according to the edge computing node identifier associated with the corresponding abnormal event in the abnormal event record; for each abnormal event, the abnormal node Pj and the corresponding abnormal evaluation index Yj are extracted from the corresponding edge computing node path, and the corresponding influence range radius Rj is calculated according to the abnormal evaluation index of the abnormal node Pj, and Rj=k×Yj, wherein k represents the influence range coefficient; Yj represents the abnormal evaluation index corresponding to the jth abnormal node; S202. For each abnormal node Pj and the corresponding influence range radius Rj, a circular area is generated, and the circular area is represented as C(Pj,Rj), which represents a circular area with Pj as the center and Rj as the radius; for all circular areas in the same edge computing node path, the union area of the circular areas is calculated, thereby obtaining the radiation area of the abnormal event; S203. For each edge computing node Nt in the radiation area, the coverage CDt in the radiation area is calculated, and the specific calculation formula is: CDt=S(Nt,C) / C(Nt,Rt), wherein S(Nt,C) represents the intersection area of the influence range area of the edge computing node Nt and the corresponding radiation area, and C(Nt,Rt) represents the influence range area of the edge computing node Nt; for the edge computing node Nt in the radiation area, the corresponding load index FLIt is calculated, and the specific calculation formula is: FLIt=LIt×(α×CDt+β×Y), Wherein LIt represents the initial load index of the edge computing node, and α and β represent weight factors; Step S300. Analyzing the load index trend curve of the edge computing node corresponding to the abnormal event, and combining the log data to obtain the load index trend curve of the edge computing node corresponding to the corresponding normal event; analyze the difference index between the load index trend curves corresponding to the abnormal event and the normal event, and match the difference index with the edge computing node path adjustment scheme corresponding to the corresponding abnormal event. Step S400. Collecting real-time communication data corresponding to the edge computing node, and obtaining the current edge computing node path; matching the abnormal event in the abnormal event database combined with the real-time communication data and the current edge computing node path, calculating the difference index of the current edge computing node path, and adjusting the current edge computing node path based on the difference index. 2.The intelligent communication data management method based on 5G communication of claim 1, wherein: The step S100 comprises: S101. The communication data includes traffic data, user behavior data and device state data; the log data refers to detailed records of edge computing node running state, event occurrence information and system operation; the communication data of the edge computing node in the selected time period is obtained, and the communication data of each edge computing node is analyzed discretely, so that each timestamp corresponds to a communication data segment; S102. Obtain the log data in the selected time period, and correspond the communication data of each edge computing node with the corresponding log data according to the time recorded by the log data, and correspond the communication data segment of each edge computing node with the corresponding log data segment according to the corresponding timestamp, and take the corresponding communication data segment and log data segment as a data unit; S103. Compare and analyze all data units of each edge computing node with the corresponding standard data unit respectively, so as to obtain the corresponding abnormal evaluation index Y, and the specific calculation formula is: Y = (1 / n)∑ i∈[1,n] di / si, Wherein, n represents the number of data units corresponding to the edge computing node, di represents the similarity between the i th data unit and the corresponding standard data unit feature, and si represents the standard characteristic value of the i th standard data unit; compare the abnormal evaluation index Y with the preset threshold Y0, and take the edge computing node in the data unit with the abnormal evaluation index Y greater than or equal to the preset threshold Y0 as the abnormal node, and identify the corresponding abnormal event according to the corresponding log data, so as to build the abnormal event database. 3.The intelligent communication data management method based on 5G communication of claim 1, wherein: The step S300 comprises: S301. Extract the edge computing node in the radiation area corresponding to the abnormal event, and obtain the load index trend curve A of the corresponding edge computing node in the selected time period; according to the log data, obtain the initial load index of the same edge computing node path in different time periods, so as to constitute the load index trend curve B of the edge computing node of the normal event; S302. Calculate the difference index between the abnormal event and the normal event load index trend curve, the difference index includes mean square error MSE, mean absolute error MAE and correlation coefficient p; according to the above difference index, the corresponding difference vector V is constituted, and V = [MSE, MAE, p], and the difference vector V is matched with the edge computing node path adjustment scheme corresponding to the abnormal event, and saved in the abnormal event database. 4.The intelligent communication data management method based on 5G communication of claim 3, wherein: The step S400 comprises: S401. Every selected time period, collect real-time communication data corresponding to the edge computing node, analyze the real-time communication data, and obtain the current edge computing node path; based on the current edge computing node path, calculate the abnormality evaluation index Y1 of each edge computing node of the current edge computing node path, and compare the abnormality evaluation index Y1 with the preset threshold Y0, if the abnormality evaluation index Y1 of each edge computing node of the current edge computing node path is less than the preset threshold Y0, no warning information is outputted; S402. If the abnormality evaluation index Y1 of an edge computing node of the current edge computing node path is greater than or equal to the preset threshold Y0, output the warning information; and according to the current edge computing node path and the corresponding abnormal node, search in the abnormal event database to match the corresponding abnormal event; according to the abnormal node of the current edge computing node path, calculate the corresponding load index and obtain the corresponding load index trend curve A1, based on the load index trend curve A1, calculate the corresponding difference index to form a difference vector V1, and V1=[MSE1, MAE1, ρ1]; S403. Obtain the difference vector V of the matched abnormal event, calculate the similarity of the difference vector V and the difference vector V1, take the maximum similarity as the final matching event, and output the adjustment scheme of the matching event corresponding edge computing node path as the reference adjustment scheme, and the system adjusts the current edge computing node path according to the reference adjustment scheme.

5. An intelligent communication data management system based on 5G communication, applied to the intelligent communication data management method based on 5G communication in any one of claims 1-4, characterized in that: The system comprises a data collection and analysis module, an abnormal event identification module, a load index trend analysis module, a real-time monitoring and warning module, and a path adjustment and optimization module. The data collection and analysis module deploys computing nodes at the network edge, collects communication data and log data, and performs discrete analysis to identify abnormal events and build an abnormal event database; the abnormal event identification module extracts the edge computing node path of the abnormal event based on the abnormal event database, calculates the radiation area and load index of the abnormal event; the load index trend analysis module analyzes the load index trend curve of the abnormal event and the normal event, calculates the difference index, and matches the adjustment scheme of the edge computing node path based on the difference index; the real-time monitoring and warning module collects real-time communication data, calculates the abnormality evaluation index, compares with the preset threshold, and outputs the warning information, and matches the abnormal event in the abnormal event database through real-time data; the path adjustment and optimization module calculates the difference vector based on the adjustment scheme of the matched abnormal event, calculates the similarity of the difference vector and the difference vector of the current edge computing node path, and finally outputs the adjustment scheme to optimize the edge computing node path. 6.The intelligent communication data management system based on 5G communication of claim 5, wherein: The data collection and analysis module comprises a data collection unit and a data analysis unit. The data collection unit is responsible for collecting communication data from the edge computing nodes, including traffic data, user behavior data and device status data, collecting running logs from the edge computing nodes, including node running status, event occurrence information and system operation records; the data analysis unit divides the collected communication data into multiple data segments according to the time stamp, and pairs them with the corresponding log data to form data units, compares all data units of each edge computing node with standard data units, calculates an abnormal evaluation index, compares the abnormal evaluation index with a preset threshold value, identifies abnormal nodes and constructs an abnormal event database. 7.The intelligent communication data management system based on 5G communication of claim 5, wherein: The abnormal event identification module includes an abnormal event record extraction unit, an abnormal node influence range calculation unit and a node load calculation unit in the radiation area; The abnormal event record extraction unit extracts abnormal event records from the abnormal event database and extracts associated edge computing node identifiers to obtain the edge computing node path corresponding to the abnormal event; the abnormal node influence range calculation unit calculates the influence range radius according to the abnormal evaluation index of the abnormal node, generates a circular area to represent the influence range, and calculates the union of all circular areas to obtain the radiation area of the abnormal event; The node load calculation unit in the radiation area calculates the coverage and load index of each edge computing node in the radiation area, and updates the database of the abnormal event according to the calculation results. 8.The intelligent communication data management system based on 5G communication of claim 5, wherein: The load index trend analysis module includes a load index trend curve extraction unit and a load index difference analysis unit; The load index trend curve extraction unit extracts the load index trend curves of the edge computing nodes corresponding to the abnormal events and normal events, respectively, to obtain the load index trend curves of the abnormal events and normal events; The load index difference analysis unit calculates the difference index between the load index trend curves of the abnormal events and normal events, including mean square error, mean absolute error and correlation coefficient, generates a difference vector and matches it with the adjustment scheme of the abnormal event. 9.The intelligent communication data management system based on 5G communication of claim 5, wherein: The real-time monitoring and early warning module includes a real-time data collection unit, a real-time abnormal evaluation unit and a real-time early warning and matching unit; The real-time data collection unit periodically collects real-time communication data of the edge computing nodes and obtains the current edge computing node path; The real-time abnormal evaluation unit calculates the abnormal evaluation index based on the current edge computing node path, compares it with the preset threshold value, and determines whether to output the warning information according to the comparison result; The real-time early warning and matching unit outputs the warning information when it detects that the abnormal evaluation index is greater than or equal to the preset threshold value, matches the abnormal event according to the current path and the records in the abnormal event database, calculates the corresponding load index trend curve and difference index, and generates the corresponding difference vector; The path adjustment and optimization module includes an adjustment scheme matching unit and a path optimization execution unit; The adjustment scheme matching unit obtains the difference vector of the abnormal event, performs similarity calculation with the difference vector obtained by real-time monitoring, and selects the one with the highest similarity as the final matching event; The path optimization execution unit outputs the edge computing node path adjustment scheme corresponding to the matching event as a reference adjustment scheme, and the system adjusts the current edge computing node path according to the reference adjustment scheme.

Images