Method, device and equipment for identifying abnormal data of base station and storage medium

Abstract

The invention discloses a base station abnormal data identification method and device, equipment and a storage medium. The method comprises the following steps: determining a base station position map corresponding to each base station in a plurality of base stations associated with a base station position data set, wherein the base station position map indicates all adjacent base stations capable of establishing a switching relationship with the base station; determining a base station switching atlas corresponding to each base station in a plurality of base stations associated with the base station switching data set, the base station switching atlas indicating an actual switching relationship between each target base station and the base station after the plurality of terminal devices perform base station switching by taking the base station as a source base station; respectively comparing the base station position map of the base station with the base station switching map of the base station, and determining an abnormal base station; and based on the abnormal base station, determining base station position abnormal data from the base station position data set and determining base station switching abnormal data from the base station switching data set. According to the method provided by the invention, the accuracy of the base station position data and the base station switching data is improved.

Description

Technical Field

[0001] The present application relates to the field of information processing technology, and in particular to a method, device, equipment and storage medium for identifying abnormal data of a base station. Background Art

[0002] With the continuous advancement of technology, base station location data and base station switching data play an increasingly important role in modern society, and are widely used in many fields such as navigation services, location sharing, disaster management, commercial applications, RFID systems, communication quality optimization, network planning, data analysis and mining, and intelligent transportation.

[0003] However, the base station location data and the base station switching data are collected from different data sources. Due to inaccurate positioning of the base station or network failures, the accuracy of the acquired base station location data and base station switching data is low. Summary of the invention

[0004] Based on the above problems, the present application provides a method, device, equipment and storage medium for identifying abnormal data of base stations, the purpose of which is to improve the accuracy of base station location data and base station switching data.

[0005] The embodiments of the present application disclose the following technical solutions: In a first aspect, the present application provides a method for identifying abnormal data of a base station, comprising: Determine a base station location map corresponding to each base station in a plurality of base stations associated with the base station location data set, the base station location map indicating all neighboring base stations that can establish a handover relationship with the base station; Determine a base station switching map corresponding to each of the multiple base stations associated with the base station switching data set, the base station switching map indicating an actual switching relationship between each target base station and the base station after multiple terminal devices perform base station switching with the base station as a source base station; Comparing the base station location map of the base station with the base station switching map of the base station respectively, and determining the abnormal base station; Based on the abnormal base station, base station position abnormal data is determined from the base station position data set and base station switching abnormal data is determined from the base station switching data set.

[0006] In this embodiment, a base station location map corresponding to each of the multiple base stations is determined based on the base station location data set, which can clearly show the switching relationship between the base station and all adjacent base stations that can establish a switching relationship; and a base station switching map corresponding to each of the multiple base stations is determined based on the base station switching data set, which can clearly show the actual switching relationship between each target base station and the base station after multiple terminal devices perform base station switching with the base station as the source base station. By comparing the constructed base station location map and the base station switching map, abnormal base stations can be accurately identified, so that the abnormal base station location data and the switching data from the base station can be accurately corresponded according to the abnormal base station, thereby improving the accuracy of the base station location data and the base station switching data.

[0007] Optionally, in the method as described above, determining a base station location map corresponding to each base station in a plurality of base stations associated with the base station location data set includes: Based on the base station longitude and latitude information and network standard information in the base station location data set, determining an adjacent base station corresponding to the base station from multiple base stations; Determine the weight corresponding to the adjacent base station according to the distance between the adjacent base station and the base station; With the base station as the center and the weights corresponding to the adjacent base stations as the edge weights, a base station location map between the base station and the adjacent base stations is constructed; the edge weights are used to characterize the distances between adjacent base stations in the base station location map, which can more clearly display the possible switching relationships between base stations, so as to improve the accuracy of abnormal base station positioning.

[0008] Optionally, in the method as described above, the neighboring base station includes an attached base station and a neighboring base station, and based on the base station longitude and latitude information and network standard information in the base station location data set, determining a neighboring base station corresponding to the base station from a plurality of base stations includes: Determine the attachment base station of the base station according to the latitude and longitude information of the base station; the distance between the attachment base station and the base station is less than a preset threshold; Based on the latitude and longitude information of the base station and the network standard information, the adjacent base stations are determined from multiple base stations, so as to more accurately understand the relative positions and relationships between different base stations and improve the accuracy of constructing the base station location map.

[0009] Optionally, the method as described above, based on the latitude and longitude information of the base station and the network standard information, determines an adjacent base station from multiple base stations, including: The proximity coefficient corresponding to each of the multiple base stations is calculated based on the latitude and longitude information of the base station and the network standard information; the proximity coefficient is used to characterize the possibility of switching between each of the multiple base stations and the base station; the proximity coefficient is negatively correlated with the distance and is related to the network standard information; Taking the base station as the center, multiple detection ranges are divided according to preset angles; By determining the base station with the largest proximity coefficient and a distance less than or equal to the farthest communication distance of the base station from each detection range in multiple detection ranges as a neighboring base station, the neighboring base stations that may have a switching relationship with the base station can be accurately determined from all base stations, thereby improving the accuracy of constructing the base station location map and avoiding missing base stations that may have a switching relationship.

[0010] Optionally, in the method described above, the proximity coefficient is related to the network standard information and includes: The network standard is inversely proportional to the maximum communication distance of the base station and directly proportional to the weight factor; The proximity coefficient is positively correlated with the farthest communication distance of the base station and is positively correlated with the weight factor.

[0011] Optionally, in the method as described above, determining a base station switching map corresponding to each base station in a plurality of base stations associated with the base station switching data set includes: According to the user identifiers in the base station switching data, the base station switching data is classified to obtain data sets corresponding to the user identifiers; The base station switching data in the data set are sorted in ascending order of time to obtain data sequences corresponding to the user identifiers; Performing statistical analysis on the data in the data sequence to determine the statistical results corresponding to the target base station and the base station in multiple feature dimensions respectively; the feature dimensions include the maximum value, minimum value, average value and variance of the switching frequency within a preset time; Obtaining a handover feature vector between the target base station and the base station according to statistical results corresponding to the target base station and the base station in multiple feature dimensions; Taking the base station as the center and the switching feature vector as the edge weight, a base station switching map corresponding to the base station is constructed, which can clearly display the actual switching relationship between base stations and improve the accuracy of abnormal base station identification.

[0012] Optionally, in the method described above, respectively comparing the base station location map of the base station with the base station switching map of the base station to determine the abnormal base station includes: A preconfigured base station abnormal switching detection and recognition model is used to compare the base station location map with the base station switching relationship in the base station switching map, and the base stations with inconsistent comparisons are determined to be abnormal base stations; the preconfigured base station abnormal switching detection and recognition model is obtained by comparing and training the historical base station location map with the historical base station switching map using a rule model according to preset rules, thereby improving the accuracy of abnormal base station detection.

[0013] In a second aspect, the present application provides a device for identifying abnormal data of a base station, comprising: A base station location map construction module, used to determine a base station location map corresponding to each base station in a plurality of base stations associated with the base station location data set, the base station location map indicating all neighboring base stations that can establish a handover relationship with the base station; A base station switching map construction module is used to determine a base station switching map corresponding to each of the multiple base stations associated with the base station switching data set, the base station switching map indicating the actual switching relationship between each target base station and the base station after multiple terminal devices perform base station switching with the base station as the source base station; An abnormal base station determination module is used to compare the base station location map of the base station with the base station switching map of the base station to determine the abnormal base station; The abnormal data determination module is used to determine the base station position abnormal data from the base station position data set and determine the base station switching abnormal data from the base station switching data set based on the abnormal base station.

[0014] In a third aspect, the present application provides an electronic device, the device comprising: a processor, and a memory communicatively connected to the processor; Memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method for identifying abnormal data of a base station described in any one of the above embodiments.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the method for identifying abnormal base station data described in any of the above embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0017] Figure 1A A schematic diagram of an application scenario provided for an embodiment of the present application; Figure 1B A flowchart of an embodiment of a method for identifying abnormal data of a base station provided in an embodiment of the present application; Figure 2 A flowchart of another embodiment of a method for identifying abnormal data of a base station provided in an embodiment of the present application; Figure 3 A flowchart of another embodiment of a method for identifying abnormal data of a base station provided in an embodiment of the present application; Figure 4 A flowchart of another embodiment of a method for identifying abnormal data of a base station provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of an embodiment of a device for identifying abnormal data of a base station provided in an embodiment of the present application; Figure 6 A schematic diagram of the structure of an embodiment of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] As described above, the current base station location data and base station switching data are not very accurate due to inaccurate positioning of the base station or network failures.

[0019] After research, the inventors have proposed a method, device, equipment and storage medium for identifying abnormal data of base stations to improve the accuracy of base station location data and base station switching data.

[0020] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0021] See also Figure 1A , which is a schematic diagram of an application scenario provided by an embodiment of the present application. Figure 1A As shown, taking base station A as an example, based on the base station location data, it is determined that the adjacent base stations that can establish a switching relationship with base station A in the base station location map corresponding to base station A are base station B and base station C. After the terminal device performs base station switching with base station A as the source base station, based on the base station switching data of all terminal devices, it is determined that in the base station switching map, base station A and base station B have an actual switching relationship, base station A and base station D also have an actual switching relationship, and base station A and base station C do not have an actual switching relationship. By comparing the base station location map of base station A with the base station switching map, it is obtained that the base station location data of base station C is abnormal data, and the base station switching data corresponding to base station D is abnormal data.

[0022] Then, according to the base station switching map corresponding to each of the multiple base stations associated with the base station switching data set, the base station switching map indicates the actual switching relationship between each target base station and the base station after multiple terminal devices perform base station switching with the base station as the source base station. See also Figure 1B, which is a flow chart of an embodiment of a method for identifying abnormal data of a base station provided in an embodiment of the present application. Figure 1B As shown, the method includes: S101: Determine a base station location map corresponding to each base station in a plurality of base stations associated with a base station location data set, wherein the base station location map indicates all neighboring base stations that can establish a switching relationship with the base station.

[0023] In this embodiment, taking base station A as an example, based on the base station longitude and latitude information in the acquired base station location data, the base station whose distance from base station A is less than a preset threshold is determined from all base stations as the attached base station corresponding to base station A, and the edge weight of the attached base station is set to 1. Then, based on the base station longitude and latitude information and network standard information in the base station location data, the adjacent base stations that may establish a switching relationship with base station A are determined from the remaining base stations except the attached base station, and the edge weight value corresponding to the adjacent base station is calculated. With base station A as the in-vertex, the attached base station and the adjacent base station as the out-vertex, a base station location map is constructed. Among them, the weight of the edge between the out-vertex and the in-vertex is the edge weight.

[0024] It is understandable that, since base stations between different operators do not communicate with each other, it is necessary to construct corresponding base station location maps for different operators. If there are shared base stations between multiple operators, the shared base stations exist in the base station location maps of multiple operators.

[0025] S102: Determine a base station switching map corresponding to each base station in a plurality of base stations associated with the base station switching data set, the base station switching map indicating actual switching relationships between each target base station and the base station after a plurality of terminal devices perform base station switching with the base station as the source base station.

[0026] In this embodiment, according to the user identifier in the acquired base station switching data, the base station switching data set is classified according to the user identifier, and then for each user identifier, the corresponding base station switching data is sorted in chronological order, so as to obtain the data sequence corresponding to each user identifier. Taking base station B as an example, the base station having a switching relationship with base station B is determined by the data in the data sequence corresponding to each user identifier, and the data sequence corresponding to each user identifier is statistically analyzed based on the base station having a switching relationship, so as to obtain the statistical results corresponding to different feature dimensions, and thus the switching feature vector containing the above statistical results is used as the edge weight. With base station B as the input vertex and the base station with the switching relationship as the output vertex, a base station switching graph is constructed. Among them, the weight of the edge between the output vertex and the input vertex is the switching feature vector.

[0027] S103: Compare the base station location map of the base station with the base station switching map of the base station respectively to determine the abnormal base station.

[0028] In this embodiment, a preconfigured base station abnormal switching detection and identification model is used to compare the base station location map with the base station switching relationship in the base station switching map, and base stations with inconsistent comparisons are determined to be abnormal base stations.

[0029] S104: Based on the abnormal base station, determine base station location abnormal data from the base station location data set and determine base station switching abnormal data from the base station switching data set.

[0030] In this embodiment, according to the identifier corresponding to the determined abnormal base station, the base station location abnormal data is determined from the base station location data set and the base station switching abnormal data is determined from the base station switching data set.

[0031] It is understandable that in order to reduce storage space and improve the query performance of the graph, the base station location map and the base station switching map can use the inverse adjacency list as the storage structure of the graph.

[0032] In this embodiment, a base station location map corresponding to each of the multiple base stations is determined based on the base station location data set, which can clearly show the switching relationship between the base station and all adjacent base stations that can establish a switching relationship; and a base station switching map corresponding to each of the multiple base stations is determined based on the base station switching data set, which can clearly show the actual switching relationship between each target base station and the base station after multiple terminal devices perform base station switching with the base station as the source base station. By comparing the constructed base station location map and the base station switching map, abnormal base stations can be accurately identified, so that the abnormal base station location data and the switching data from the base station can be accurately corresponded according to the abnormal base station, thereby improving the accuracy of the base station location data and the base station switching data.

[0033] See also Figure 2 , which is a flow chart of another embodiment of a method for identifying abnormal data of a base station provided in an embodiment of the present application. Figure 1B , Figure 2 A specific implementation method of "determining a base station location map corresponding to each base station in a plurality of base stations associated with a base station location data set" in S101 is shown, specifically in combination with Figure 2 Introduced in S1011-S1013.

[0034] S1011: Determine an adjacent base station corresponding to the base station from a plurality of base stations based on the base station longitude and latitude information and network standard information in the base station location data set.

[0035] In this embodiment, the acquired base station location data includes base station identification, network standard information, base station longitude and latitude information, and province, city, and district. The attached base station of the base station is determined based on the longitude and latitude information of the base station; the distance between the attached base station and the base station is less than a preset threshold; based on the longitude and latitude information of the base station and the network standard information, the adjacent base station is determined from multiple base stations.

[0036] It is understandable that the latitude and longitude information of the base stations in the base station location data set can be matched with the location of the province, city or district to which it belongs according to the geographical location on the map.

[0037] S1012: Determine a weight corresponding to an adjacent base station according to a distance between the adjacent base stations.

[0038] In this embodiment, for example, if the adjacent base station is an attached base station, the weight corresponding to the adjacent base station defaults to 1. If the adjacent base station is a nearby base station, the distance between the adjacent base station and the base station is normalized using a Sigmoid activation function to determine the weight corresponding to the adjacent base station.

[0039] S1013: With the base station as the center and the weights corresponding to the adjacent base stations as edge weights, a base station location map between the base station and the adjacent base stations is constructed.

[0040] In this embodiment, taking base station A as an example, base station A is taken as the inbound vertex, and the attached base stations and adjacent base stations are taken as outbound vertices to construct a base station location map. The weight of the edge between the outbound vertex and the inbound vertex is the edge weight; the edge weight is used to represent the distance between adjacent base stations and base stations in the base station location map.

[0041] In this embodiment, based on the latitude and longitude information of the base station and the network standard information in the base station location data set, the adjacent base stations corresponding to the base station are determined from multiple base stations to improve the accuracy of positioning the adjacent base stations, thereby more accurately obtaining the adjacent base stations that may have a switching relationship; then, based on the distance between the adjacent base stations, the weights corresponding to the adjacent base stations are determined, which can more accurately reflect the spatial position relationship between the base stations, and construct a base station location map between the base station and the adjacent base stations with the base station as the center and the weights corresponding to the adjacent base stations as the edge weights, which can more clearly show the possible switching relationships between base stations, so as to improve the accuracy of locating abnormal base stations.

[0042] Combination Figure 2 As shown, after executing S1011-S1013, referring to the previous embodiment, S102-S104 are executed to achieve the determination of base station location abnormal data and base station switching abnormal data.

[0043] Further, based on the above embodiment, the adjacent base station includes an attached base station and a neighboring base station, and a specific implementation of "determining an adjacent base station corresponding to the base station from multiple base stations based on the base station longitude and latitude information and network standard information in the base station location data set" in S1011 includes: The attachment base station of the base station is determined according to the latitude and longitude information of the base station; and the distance between the attachment base station and the base station is less than a preset threshold.

[0044] In this embodiment, taking base station A as an example, according to the latitude and longitude information corresponding to all base stations and the latitude and longitude information of base station A, base stations whose distance to base station A is less than a preset distance of 10m are calculated as attachment base stations.

[0045] Based on the base station latitude and longitude information and network standard information, a neighboring base station is determined from multiple base stations.

[0046] In this embodiment, based on the latitude and longitude information of the base station and the network standard information, the proximity coefficients between all base stations except the attached base station and the base station A are calculated. Then, with the base station A as the center, multiple detection ranges are divided according to preset angles, and the base station with the largest proximity coefficient and a distance from the base station less than or equal to the farthest communication distance of the base station is determined from each detection range as the adjacent base station.

[0047] In this embodiment, the attached base station of the base station is determined based on the latitude and longitude information of the base station; the distance between the attached base station and the base station is less than a preset threshold; and then based on the latitude and longitude information of the base station and the network standard information, the adjacent base station is determined from multiple base stations to more accurately understand the relative positions and relationships between different base stations, thereby improving the accuracy of constructing a base station location map.

[0048] Further, based on the above embodiment, a specific implementation of “determining a neighboring base station from multiple base stations based on the base station latitude and longitude information and the network standard information” includes: The proximity coefficient corresponding to each base station in the plurality of base stations is calculated based on the base station longitude and latitude information and the network standard information.

[0049] The proximity coefficient is used to characterize the possibility of handover between each base station in a plurality of base stations; the proximity coefficient is negatively correlated with the distance and is related to the network standard information.

[0050] In this embodiment, with base station A as the center, since the base station signal strength is inversely proportional to the base station distance; for different network standards, the communication priority of the base station with a network standard of 5G is higher than that of the base station with a network standard of 4G. The proximity coefficient can be expressed by the following expression:

[0051] The proximity coefficient is positively correlated with the farthest communication distance of the base station and is positively correlated with the weight factor.

[0052] Among them, Zscore represents the proximity coefficient between the current base station and the central base station; T is the maximum communication distance threshold of the current base station. Specifically, the network standard is inversely proportional to the maximum communication distance of the base station; for example, the maximum communication distance threshold of a 2G base station may be 10km, the maximum communication distance threshold of a 3G base station may be 5km, the maximum communication distance threshold of a 4G base station may be 3km, and the maximum communication distance threshold of a 5G base station may be 500m.

[0053] H is the distance between the current base station and the central base station; K1 is the initialization factor and can be set to a default value; for example, K1=10.

[0054] K2 is a weight factor. Specifically, the network standard is proportional to the weight factor. For example, the weight factor of a 2G base station can be set to 0.5; the weight factor of a 3G base station can be set to 0.5; the weight factor of a 4G base station can be set to 1; and the weight factor of a 5G base station can be set to 2.

[0055] With the base station as the center, multiple detection ranges are divided according to preset angles.

[0056] In this embodiment, for example, if the preset angle is 5°, a polar coordinate system is constructed with base station A as the center and 0° in the east direction, and every 5° in the counterclockwise direction is a detection range.

[0057] A base station with a maximum proximity coefficient and a distance less than or equal to the farthest communication distance of the base station is determined from each detection range in the multiple detection ranges as a neighboring base station.

[0058] In this embodiment, according to the calculated proximity coefficient of each base station, a base station with the largest proximity coefficient and a distance from the base station less than or equal to the farthest communication distance of the base station is determined from each detection range as a neighboring base station. If the distance between the base station with the largest proximity coefficient and base station A exceeds the farthest communication distance threshold of the base station, it is determined that there is no neighboring base station in the detection range.

[0059] In this embodiment, the proximity coefficient corresponding to each of the multiple base stations is calculated based on the base station latitude and longitude information and the network standard information, so that the adjacent base stations can be identified more accurately. With the base station as the center, multiple detection ranges are divided according to preset angles to identify adjacent base stations by region and reduce signal blind spots; then, from each detection range in the multiple detection ranges, the base station with the largest proximity coefficient and a distance less than or equal to the farthest communication distance of the base station is determined as the adjacent base station, which can accurately determine the adjacent base stations that may have a switching relationship with the base station from all base stations, improve the accuracy of constructing the base station location map, and avoid missing base stations that may have a switching relationship.

[0060] See also Figure 3 , which is a flow chart of another embodiment of a method for identifying abnormal data of a base station provided in an embodiment of the present application. Figure 1B , Figure 3 A specific implementation method of "determining a base station switching map corresponding to each of the multiple base stations associated with the base station switching data set" in S102 is shown, specifically in combination with Figure 3 Introduced in S1021-S1025.

[0061] S1021: Classify the base station switching data according to the user identifiers in the base station switching data to obtain data sets corresponding to the user identifiers.

[0062] In this embodiment, the acquired base station switching data is the relevant information extracted from the event data in the standardized signaling log data, including the start time, user ID, base station longitude and latitude information, current base station ID, and previous base station ID. According to the user ID corresponding to each user, the base station switching data is classified to obtain a data set corresponding to each user ID.

[0063] S1022: Sort the base station switching data in the data set in ascending order of time to obtain data sequences corresponding to the user identifiers.

[0064] In this embodiment, taking the data set corresponding to the user identification A as an example, the data set corresponding to the user identification A is sorted in ascending order of start time to obtain a data sequence corresponding to the user identification A.

[0065] It is understandable that after obtaining the data sequences corresponding to the user identifiers, the data can be preprocessed. Specifically, the preprocessing operation includes deduplication of the same data in the data sequence corresponding to the user identifier and filling or filtering the abnormal data in the data sequence. Among them, filling or filtering the abnormal data in the data sequence includes performing the following operations on each target data in the data sequence: (1) If the latitude and longitude information of the previous base station corresponding to the target data in the data sequence is the same as the latitude and longitude information of the current base station corresponding to the previous data, the target data is normal data.

[0066] (2) If the latitude and longitude information of the previous base station corresponding to the target data in the data sequence is different from the latitude and longitude information of the current base station corresponding to the previous data, and the time interval between the start time corresponding to the target data and the start time corresponding to the previous data is less than or equal to 60s, it is considered that the data between the two data is lost, and a new data is filled between the target data and the previous data. The latitude and longitude information of the current base station in the new data is the latitude and longitude information of the previous base station corresponding to the target data, and the latitude and longitude information of the previous base station is the latitude and longitude information of the current base station corresponding to the previous data.

[0067] (3) If the latitude and longitude information of the previous base station corresponding to the target data in the data sequence is different from the latitude and longitude information of the current base station corresponding to the previous data, and the time interval between the start time corresponding to the target data and the start time corresponding to the previous data is greater than 60s, then the target data is considered to be normal data.

[0068] (4) If the latitude and longitude information of the previous base station corresponding to the target data in the data sequence is missing, and the time interval between the target data and the previous data is less than or equal to 60s, the latitude and longitude information of the previous base station corresponding to the target data is filled with the latitude and longitude information of the current base station corresponding to the previous data.

[0069] (5) If the latitude and longitude information of the previous base station corresponding to the target data in the data sequence is missing, and the time interval between the target data and the previous data is greater than 60s, the target data is filtered as abnormal data.

[0070] S1023: Perform statistical analysis on the data in the data sequence to respectively determine statistical results corresponding to the target base station and the base station in multiple feature dimensions.

[0071] The characteristic dimensions include the maximum value, minimum value, average value and variance of the switching frequency within a preset time.

[0072] In this embodiment, for each user identifier, the number of handovers between the target base station and the base station is determined according to the data sequence, and a handover relationship with a higher handover frequency between the target base station and the base station is selected. For example, if the target base station has 10 handover relationships with other base stations, the handover frequencies corresponding to the 10 handover relationships are sorted, and 9 handover relationships corresponding to higher handover frequencies are retained.

[0073] It is understandable that, in the process of statistics, if the longitude and latitude information of the current base station corresponding to the current data is the same as the longitude and latitude information of the previous base station, it is not considered that there is a switching relationship.

[0074] Based on the determined switching relationship, the statistical results corresponding to each target base station and base station in multiple feature dimensions are respectively counted, such as the number of switches between the target base station and the base station, the switching frequency between the target base station and the base station in different time periods, the number of switches between the target base station and the base station in different time periods, the maximum, minimum, average and variance of the number of switches between the target base station and the base station within 30 minutes, etc.

[0075] S1024: Obtain a switching feature vector between the target base station and the base station according to statistical results corresponding to the target base station and the base station in multiple feature dimensions.

[0076] In this embodiment, the statistical results corresponding to the target base station and the base station in multiple feature dimensions are used as the handover feature vector between the target base station and the base station.

[0077] S1025: Taking the base station as the center and the switching feature vector as the edge weight, a base station switching graph corresponding to the base station is constructed.

[0078] In this embodiment, the target base station is taken as base station A, and base station A is taken as the input vertex and the base station is taken as the output vertex to construct a base station switching graph. The edge weight between the output vertex and the input vertex is the switching feature vector; the edge weight is used to characterize the switching relationship between the target base station and the base station in the base station switching graph.

[0079] Combination Figure 3 As shown, before executing S1021-S1025, refer to the previous embodiment and execute S101 to determine the base station location map; after executing S1021-S1025, refer to the previous embodiment and execute S103-S104 to determine the base station location abnormality data and the base station switching abnormality data.

[0080] In this embodiment, according to the user identifier in the base station switching data, the base station switching data is classified to obtain a data set corresponding to each user identifier. The base station switching data in the data set is sorted in order of time from small to large to obtain a data sequence corresponding to each user identifier, so as to clearly show the base station switching situation of the user at different time points; then the data in the data sequence is statistically analyzed to determine the statistical results corresponding to the target base station and the base station in multiple feature dimensions, which can fully reflect the switching characteristics of the switching between the target base station and the base station and be used to determine the switching feature vector between the target base station and the base station. With the base station as the center and the switching feature vector as the edge weight, a base station switching map corresponding to the base station is constructed, which can clearly show the actual switching relationship between the base stations and improve the accuracy of abnormal base station identification.

[0081] See also Figure 4, which is a flow chart of another embodiment of a method for identifying abnormal data of a base station provided in an embodiment of the present application. Figure 1B , Figure 4 A specific implementation method of "respectively comparing the base station location map of the base station with the base station switching map of the base station to determine the abnormal base station" in S103 is shown, specifically combined with Figure 4 S1031 is introduced in this article.

[0082] S1031: Using a pre-configured base station abnormal switching detection and identification model, the base station location map is compared with the base station switching relationship in the base station switching map, and the base stations with inconsistent comparisons are determined to be abnormal base stations.

[0083] Among them, the pre-configured base station abnormal switching detection and recognition model is obtained by comparing and training the historical base station location map with the historical base station switching map using a rule model according to preset rules.

[0084] In this embodiment, the preconfigured base station abnormal switching detection and identification model is obtained by comparing and training the historical base station location map with the historical base station switching map using a rule model according to preset rules. Since the base station location map already contains all possible switching relationships, when the switching relationship in the base station switching map cannot be successfully compared in the base station location map, the base station is considered to be an abnormal base station. Specifically, abnormal base stations may include base stations with inaccurate locations, base stations with missing switching relationships, base stations with wrong switching, and the like.

[0085] In this embodiment, a preconfigured base station abnormal switching detection and identification model is used to compare the base station location map with the base station switching relationship in the base station switching map, thereby avoiding errors that may be caused by a single data source, and being able to accurately identify base stations with inconsistent comparisons, thereby improving the accuracy of abnormal base station detection.

[0086] Combination Figure 4 As shown, before executing S1031, refer to the previous embodiment and execute S101-S102 to determine the base station location map and the base station switching map; after executing S1031, refer to the previous embodiment and execute S104 to determine the base station location abnormality data and the base station switching abnormality data.

[0087] Furthermore, based on the above embodiment, the method may further include: According to the manual verification results of the base station location anomaly data and the base station switching anomaly data, the base station location anomaly data and the base station switching anomaly data are labeled; and the features are extracted from the labeled data. Then, machine learning models, such as random forest, logistic regression, XGBoost, and LightGBM, are used to train the feature vectors and annotations, and the grid search algorithm is used to tune the parameters to obtain a prediction model; specifically, the prediction model is used to optimize the distance threshold setting during the construction of the base station location map.

[0088] See also Figure 5 , which is a schematic diagram of the structure of an embodiment of a device for identifying abnormal data of a base station provided in an embodiment of the present application. Figure 5 As shown, the device 50 includes: a base station location map construction module 51, a base station switching map construction module 52, an abnormal base station determination module 53 and an abnormal data determination module 54.

[0089] The base station location map construction module 51 is used to determine a base station location map corresponding to each base station in the plurality of base stations associated with the base station location data set, and the base station location map indicates all neighboring base stations that can establish a handover relationship with the base station; A base station switching map construction module 52 is used to determine a base station switching map corresponding to each of the multiple base stations associated with the base station switching data set, the base station switching map indicating the actual switching relationship between each target base station and the base station after multiple terminal devices perform base station switching with the base station as the source base station; The abnormal base station determination module 53 is used to compare the base station location map of the base station with the base station switching map of the base station to determine the abnormal base station; The abnormal data determination module 54 is used to determine the base station location abnormal data from the base station location data set and determine the base station switching abnormal data from the base station switching data set based on the abnormal base station.

[0090] An apparatus for identifying abnormal data of a base station provided in an embodiment of the present application can execute the technical solution shown in the above method embodiment, and its implementation principle and beneficial effects are similar, which will not be repeated here.

[0091] Further, based on the above embodiments, the base station location map construction module 51 is specifically used to determine the adjacent base stations corresponding to the base station from multiple base stations based on the base station longitude and latitude information and network standard information in the base station location data set; determine the weights corresponding to the adjacent base stations according to the distances between the adjacent base stations; construct a base station location map between the base station and the adjacent base stations with the base station as the center and the weights corresponding to the adjacent base stations as the edge weights; the edge weights are used to characterize the distances between adjacent base stations in the base station location map.

[0092] An apparatus for identifying abnormal data of a base station provided in an embodiment of the present application can execute the technical solution shown in the above method embodiment, and its implementation principle and beneficial effects are similar, which will not be repeated here.

[0093] Further, based on the above embodiments, the base station location map construction module 51 is specifically used to determine the attached base station of the base station according to the latitude and longitude information of the base station; the distance between the attached base station and the base station is less than a preset threshold; based on the latitude and longitude information of the base station and the network standard information, determine the adjacent base station from multiple base stations.

[0094] An apparatus for identifying abnormal data of a base station provided in an embodiment of the present application can execute the technical solution shown in the above method embodiment, and its implementation principle and beneficial effects are similar, which will not be repeated here.

[0095] Further, based on the above embodiments, the base station location map construction module 51 is specifically used to calculate the proximity coefficient corresponding to each base station in a plurality of base stations based on the base station longitude and latitude information and the network standard information; the proximity coefficient is used to characterize the possibility of switching between each base station in a plurality of base stations; the proximity coefficient is negatively correlated with the distance and is related to the network standard information; with the base station as the center, a plurality of detection ranges are divided according to preset angles; and from each detection range in the plurality of detection ranges, a base station with the largest proximity coefficient and a distance to the base station that is less than or equal to the farthest communication distance of the base station is determined as a neighboring base station.

[0096] An apparatus for identifying abnormal data of a base station provided in an embodiment of the present application can execute the technical solution shown in the above method embodiment, and its implementation principle and beneficial effects are similar, which will not be repeated here.

[0097] Further, based on the above embodiments, the base station switching map construction module 52 is specifically used to classify the base station switching data according to the user identifiers in the base station switching data to obtain data sets corresponding to each user identifier; sort the base station switching data in the data set in order of time from small to large to obtain data sequences corresponding to each user identifier; perform statistical analysis on the data in the data sequence to determine the statistical results corresponding to the target base station and the base station in multiple feature dimensions respectively; the feature dimensions include the maximum value, minimum value, average value and variance of the switching frequency within a preset time; based on the statistical results corresponding to the target base station and the base station in multiple feature dimensions respectively, obtain the switching feature vector between the target base station and the base station; with the base station as the center and the switching feature vector as the edge weight, construct a base station switching map corresponding to the base station.

[0098] An apparatus for identifying abnormal data of a base station provided in an embodiment of the present application can execute the technical solution shown in the above method embodiment, and its implementation principle and beneficial effects are similar, which will not be repeated here.

[0099] Further, based on the above embodiments, the abnormal base station determination module 53 is specifically used to use a pre-configured base station abnormal switching detection and identification model to compare the base station location map with the base station switching relationship in the base station switching map, and determine that the base station with inconsistent comparison is an abnormal base station; the pre-configured base station abnormal switching detection and identification model is obtained by comparing and training the historical base station location map with the historical base station switching map using a rule model according to preset rules.

[0100] An apparatus for identifying abnormal data of a base station provided in an embodiment of the present application can execute the technical solution shown in the above method embodiment, and its implementation principle and beneficial effects are similar, which will not be repeated here.

[0101] See also Figure 6 , which is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. The electronic device 60 may include: a processor 61 and a memory 62.

[0102] The processor 61 is in communication connection with the memory 62 , and the memory 62 is used to store computer-executable instructions; the processor 61 is configured to execute the technical solution in any of the aforementioned method embodiments by executing the computer-executable instructions stored in the memory 62 .

[0103] Optionally, the memory 62 may be independent or integrated with the processor 61. Optionally, when the memory 62 is a device independent of the processor 61, the electronic device 60 may further include: a bus for connecting the above devices.

[0104] The electronic device is used to execute the technical solution in any of the aforementioned method embodiments, and its implementation principle and technical effect are similar and will not be repeated here.

[0105] An embodiment of the present application also provides a computer-readable storage medium, which stores computer execution instructions. When the computer execution instructions are executed by a processor, they are used to implement the above-mentioned method. The implementation principle and technical effect are similar and will not be repeated here.

[0106] It should be noted that each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described above is merely schematic, in which the unit described as a separate component may or may not be physically separated, and the component prompted as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.

[0107] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for identifying abnormal data of a base station, characterized in that: include: Determine a base station location map corresponding to each base station in a plurality of base stations associated with the base station location data set, the base station location map indicating all neighboring base stations that can establish a handover relationship with the base station; Determine a base station switching map corresponding to each of the multiple base stations associated with the base station switching data set, the base station switching map indicating an actual switching relationship between each target base station and the base station after multiple terminal devices perform base station switching with the base station as a source base station; Comparing the base station location map of the base station with the base station switching map of the base station respectively to determine the abnormal base station; Based on the abnormal base station, base station location abnormal data is determined from the base station location data set and base station switching abnormal data is determined from the base station switching data set.

2. The method according to claim 1, characterized in that The determining of a base station location map corresponding to each base station in a plurality of base stations associated with the base station location data set includes: Determine, from the plurality of base stations, an adjacent base station corresponding to the base station based on the base station longitude and latitude information and the network standard information in the base station location data set; Determining a weight corresponding to the adjacent base station according to a distance between the adjacent base station and the base station; With the base station as the center and the weights corresponding to the adjacent base stations as edge weights, a base station location map between the base station and the adjacent base stations is constructed; the edge weights are used to characterize the distance between the adjacent base stations and the base station in the base station location map.

3. The method according to claim 2, characterized in that The adjacent base stations include attached base stations and adjacent base stations, and the determining, from the plurality of base stations based on the base station longitude and latitude information and network standard information in the base station location data set, an adjacent base station corresponding to the base station includes: Determine the attachment base station of the base station according to the latitude and longitude information of the base station; the distance between the attachment base station and the base station is less than a preset threshold; Based on the latitude and longitude information of the base station and the network standard information, a neighboring base station is determined from the multiple base stations.

4. The method according to claim 3, characterized in that The determining of a neighboring base station from the plurality of base stations based on the base station latitude and longitude information and the network standard information includes: Calculating a proximity coefficient corresponding to each of the multiple base stations based on the latitude and longitude information of the base station and the network standard information; the proximity coefficient is used to characterize the possibility of switching between each of the multiple base stations and the base station; the proximity coefficient is negatively correlated with the distance and is related to the network standard information; Taking the base station as the center, a plurality of detection ranges are divided according to preset angles; A base station having the largest proximity coefficient and a distance less than or equal to the farthest communication distance of a base station is determined from each detection range of the multiple detection ranges as a neighboring base station.

5. The method according to claim 4, characterized in that The proximity coefficient is related to the network standard information and includes: The network standard is inversely proportional to the maximum communication distance of the base station and directly proportional to the weight factor; The proximity coefficient is positively correlated with the farthest communication distance of the base station and is positively correlated with the weight factor.

6. The method according to claim 1, characterized in that The determining of a base station switching map corresponding to each base station of the plurality of base stations associated with the base station switching data set includes: Classifying the base station switching data according to the user identifiers in the base station switching data to obtain data sets corresponding to the user identifiers; The base station switching data in the data set are sorted in ascending order of time to obtain data sequences corresponding to the user identifiers; Performing statistical analysis on the data in the data sequence to respectively determine statistical results corresponding to the target base station and the base station in multiple feature dimensions; the feature dimensions include a maximum value, a minimum value, an average value, and a variance of the switching frequency within a preset time; Obtaining a handover feature vector between the target base station and the base station according to statistical results corresponding to the target base station and the base station in multiple feature dimensions; With the base station as the center and the switching feature vector as the edge weight, a base station switching graph corresponding to the base station is constructed.

7. The method according to claim 1, characterized in that The comparing the base station location map of the base station with the base station switching map of the base station to determine the abnormal base station includes: A preconfigured base station abnormal switching detection and recognition model is used to compare the base station location map with the base station switching relationship in the base station switching map, and the base station with inconsistent comparison is determined to be an abnormal base station; the preconfigured base station abnormal switching detection and recognition model is obtained by comparing and training the historical base station location map with the historical base station switching map using a rule model according to preset rules.

8. A device for identifying abnormal data of a base station, characterized in that: include: A base station location map construction module, used to determine a base station location map corresponding to each base station in a plurality of base stations associated with the base station location data set, wherein the base station location map indicates all neighboring base stations that can establish a handover relationship with the base station; A base station switching map construction module, used to determine a base station switching map corresponding to each base station in a plurality of base stations associated with a base station switching data set, wherein the base station switching map indicates an actual switching relationship between each target base station and the base station after a plurality of terminal devices perform base station switching with the base station as a source base station; An abnormal base station determination module is used to compare the base station location map of the base station with the base station switching map of the base station to determine the abnormal base station; The abnormal data determination module is used to determine the base station location abnormal data from the base station location data set and determine the base station switching abnormal data from the base station switching data set based on the abnormal base station.

9. An electronic device, characterized in that: The device comprises: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

Images