Methods, devices and computer equipment for determining line anomalies

Abstract

This invention provides a method, apparatus, and computer device for determining line anomalies. The method includes: extracting features from sample data collected from each floor to generate first feature data; inputting the first feature data and the generated first initial centroid into a centroid training model to output a first clustering result; performing dimensionality reduction processing on the first clustering result to generate a first target centroid; generating anomaly intervals based on the generated first dataset and the first target centroids; and determining whether there are anomalies in the feeder lines of the target floor based on the collected communication data of the target floor and the anomaly intervals. The computer device can quickly determine whether there are anomalies in the feeder lines of the target floor based on the generated anomaly intervals, saving testing costs, reducing testing difficulty, and improving testing efficiency.

Description

[Technical Field]

[0001] This invention relates to the field of antenna technology, and in particular to a method, apparatus, and computer device for determining line anomalies. [Background Technology]

[0002] With the increasing prevalence of mobile internet and the Internet of Things (IoT), the demand for mobile communication networks is growing stronger. To meet this increasing customer demand, mobile operators invest heavily in network construction each year, and this cost has become even greater with the rise and application of 5G technology. According to data service statistics from major operators' network departments, a large portion of mobile applications occur indoors, accounting for almost 75% of total business. Therefore, indoor coverage has become a fiercely contested area for mobile operators. However, indoor coverage scenarios are exceptionally complex. For example, interior renovations may damage mobile network cabling, affecting indoor signal coverage and creating significant challenges for subsequent maintenance.

[0003] To detect problems in indoor distributed antenna systems (DAS) in real time, related technologies incorporate an indoor monitoring device based on Radio Frequency Identification (RFID) technology into the existing passive indoor DAS. An RFID tag is attached to the antenna, and the RFID gateway periodically transmits a radio frequency signal in the 840MHz-845MHz band. This radio frequency signal is combined with information from the existing remote radio units (RRUs) via a combiner and then transmitted to the antenna end via a feeder line. The RFID tag at the antenna end receives the RFID radio frequency signal, charges itself based on the signal, and once it reaches a certain energy level, it reflects an energy signal that travels back through the original feeder line to the RFID gateway. The RFID gateway decodes this signal to determine if the wiring of each antenna is damaged. However, this method has low testing efficiency. When the wiring is concealed within the ceiling, it is difficult for on-site personnel to locate the antenna, requiring the use of a frequency sweeper, which increases the testing cost and difficulty. [Summary of the Invention]

[0004] In view of this, embodiments of the present invention provide a method, apparatus and computer device for determining line anomalies, in order to solve the problems of high testing cost, high testing difficulty and low testing efficiency in the prior art of line testing.

[0005] In a first aspect, embodiments of the present invention provide a method for determining line anomalies, the method comprising:

[0006] Feature extraction is performed on the sample data collected from each floor to generate the first feature data;

[0007] Input the first feature data and the generated first initial centroid into the centroid training model, and output the first clustering result;

[0008] The first clustering result is subjected to dimensionality reduction processing to generate the first target centroid;

[0009] Anomaly intervals are generated based on the first generated dataset and the first target centroid.

[0010] Based on the collected communication data of the target floor and the abnormal interval, it is determined whether there is an abnormality in the feeder line of the target floor. The target floor includes the floor selected from the at least one floor.

[0011] In one possible implementation, the sample data includes an RFID line loss data table for indoor distributed antenna systems (DAS), a minimum drive test (MDT) measurement data table for indoor DAS, a deep packet inspection (DPI) data table, and a Wi-Fi information data table.

[0012] The step of extracting features from the collected sample data from each floor to generate first feature data includes:

[0013] Connect the MDT measurement data table and the DPI acquisition data table to generate a second dataset;

[0014] The MDT measurement data table is filtered to generate a third dataset;

[0015] The third dataset and the Wi-Fi information data table are connected to generate the fourth dataset;

[0016] The first and fourth datasets are merged to generate the fifth dataset.

[0017] The first feature data is selected from the fifth dataset according to the preset selection rules.

[0018] In one possible implementation, generating the anomaly interval based on the generated first dataset and the first target centroid includes:

[0019] Based on the generated first dataset and the first target centroid, generate line anomaly indicators for sample data;

[0020] The minimum and maximum values ​​are selected from the line anomaly indicators of the sample data, and the minimum and maximum values ​​are used as the two endpoints of the anomaly interval.

[0021] An abnormal interval is generated based on the two endpoints of the abnormal interval.

[0022] In one possible implementation, the process of generating the line anomaly index based on the generated first dataset and the first target centroid includes:

[0023] The area of ​​the first centroid is generated by calculating the first centroid area according to the area formula.

[0024] The area of ​​the first centroid is divided according to the generated first dataset to determine the area of ​​the first normal centroid and the area of ​​the first abnormal centroid.

[0025] The smallest centroid area is selected from the first normal centroid area, and the smallest centroid area is used as the floor standard reference surface of the indoor distribution cell.

[0026] The standard reference surface of the floor and the area of ​​the first abnormal centroid are calculated according to the standard ratio formula to generate the line anomaly index of the sample data.

[0027] In one possible implementation, the RFID line loss data table includes antenna line loss for each floor covered by the indoor distributed antenna system (DAS) and the cell identification code of the DAS.

[0028] Before merging the generated first and fourth datasets to generate the fifth dataset, the process further includes:

[0029] Determine whether the antenna line loss of each floor covered by the indoor distributed antenna system is less than or equal to a set threshold.

[0030] If it is determined that the antenna line loss of a floor is less than or equal to a set threshold, then the floor is marked according to the intact enumeration value;

[0031] If it is determined that the antenna loss of a floor is greater than a set threshold, the floor is marked according to the defective enumeration value.

[0032] The first dataset is generated based on the cell identification code of the indoor distributed antenna system (DAS) cell and the line loss enumeration values ​​of each floor covered by the indoor DAS cell. The line loss enumeration values ​​of each floor covered by the indoor DAS cell include intact enumeration values ​​and / or incomplete enumeration values.

[0033] In one possible implementation, determining whether there is an anomaly in the feeder line of the target floor based on the collected communication data of the target floor and the abnormal interval includes:

[0034] Based on the generated sixth dataset and the generated second target centroid, generate the line anomaly index for the target floor;

[0035] Determine whether the abnormal line indicators of the target floor are within the abnormal range;

[0036] If the abnormality index of the line on the target floor is determined to be within the abnormal range, then it is determined that there is no abnormality in the feeder line of the target floor.

[0037] If it is determined that the abnormality index of the line on the target floor is not within the abnormal range, then it is determined that there is an abnormality in the feeder line of the target floor.

[0038] In one possible implementation, generating the line anomaly index for the target floor based on the generated sixth dataset and the generated second target centroid includes:

[0039] The area of ​​the second centroid is calculated based on the area formula to generate the area of ​​the second centroid.

[0040] The area of ​​the second centroid is divided according to the generated sixth dataset to determine the area of ​​the second normal centroid and the area of ​​the second abnormal centroid.

[0041] The smallest centroid area is selected from the second normal centroid area, and the smallest centroid area is used as the floor standard reference surface of the indoor distribution cell.

[0042] The standard reference surface of the floor and the area of ​​the second abnormal centroid are calculated according to the standard ratio formula to generate the line anomaly index of the target floor.

[0043] Secondly, embodiments of the present invention provide a line anomaly determination device, the device comprising:

[0044] The first generation module is used to extract features from the sample data collected from each floor and generate the first feature data.

[0045] The output module is used to input the first feature data and the generated first initial centroid into the centroid training model and output the first clustering result;

[0046] The second generation module is used to perform dimensionality reduction processing on the first clustering result to generate the first target centroid;

[0047] The third generation module generates anomaly intervals based on the generated first dataset and the first target centroid.

[0048] The judgment module determines whether there is an anomaly in the feeder line of the target floor based on the collected communication data of the target floor and the abnormal interval. The target floor includes the floor selected from the at least one floor.

[0049] Thirdly, embodiments of the present invention provide a computer device, including:

[0050] One or more processors; a memory; and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions that, when executed by the device, cause the device to perform the line anomaly determination method in the first aspect or any possible implementation of the first aspect.

[0051] Fourthly, embodiments of the present invention provide a computer-readable storage medium comprising a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to execute the line anomaly determination method in the first aspect or any possible implementation thereof.

[0052] In the technical solution provided by this invention, feature extraction is performed on the sample data collected from each floor to generate first feature data; the first feature data and the generated first initial centroid are input into a centroid training model to output a first clustering result; the first clustering result is subjected to dimensionality reduction processing to generate a first target centroid; anomaly intervals are generated based on the generated first dataset and the first target centroid; and the presence of anomalies in the feeder lines of the target floor is determined based on the collected communication data of the target floor and the anomaly intervals. In the technical solution provided by this invention, computer equipment can quickly determine whether there are anomalies in the feeder lines of the target floor based on the generated anomaly intervals, saving testing costs, reducing testing difficulty, and improving testing efficiency. [Attached Image Description]

[0053] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart of a method for determining line anomalies provided in an embodiment of the present invention;

[0055] Figure 2 A flowchart of a feature extraction method provided in an embodiment of the present invention;

[0056] Figure 3 A flowchart illustrating a method for generating a first dataset according to an embodiment of the present invention;

[0057] Figure 4 A flowchart of an abnormal interval generation method provided in an embodiment of the present invention;

[0058] Figure 5 A flowchart of a method for generating line anomaly indicators provided in an embodiment of the present invention;

[0059] Figure 6 A flowchart of a line anomaly detection method provided in an embodiment of the present invention;

[0060] Figure 7 A flowchart of another method for generating line anomaly indicators provided in an embodiment of the present invention;

[0061] Figure 8 A flowchart illustrating a method for generating a sixth dataset according to an embodiment of the present invention;

[0062] Figure 9 This is a schematic diagram of a line anomaly determination device provided in an embodiment of the present invention;

[0063] Figure 10 This is a schematic diagram of a computer device provided in an embodiment of the present invention.

Detailed Implementation Methods

[0064] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] It should be understood that the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0066] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0067] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0068] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0069] Figure 1 A flowchart of a method for determining line anomalies provided in an embodiment of the present invention is shown below. Figure 1 As shown, the method includes:

[0070] Step 101: Extract features from the sample data collected from each floor to generate the first feature data.

[0071] Each step in the embodiments of the present invention can be executed by a computer device.

[0072] In this step, the sample data includes RFID line loss data tables for indoor distributed antenna systems (DAS) cells, Minimization of Drive Test (MDT) measurement data tables for indoor DAS cells, Deep Packet Inspection (DPI) data tables, and Wi-Fi information data tables. Specifically, the computer equipment collects RFID line loss data tables for indoor distributed antenna systems (DAS) from the intelligent indoor DAS platform. The main fields of the RFID line loss data tables include the cell ID, antenna name, antenna line loss, antenna status, and the horizontal coordinates of the antenna in the plane. The computer equipment also collects MDT measurement data tables from the MDT platform. The main fields of the MDT measurement data tables include the cell ID, reference signal receiving power (RSRP), signal to interference plus noise ratio (SINR), neighboring cell IDs, neighboring cell RSRP, and the first International Mobile Subscriber Identity (IMSI). Furthermore, the computer equipment collects DPI acquisition data tables from the signaling platform. The main fields of the DPI acquisition data tables include the second IMSI and the first MAC address. Finally, the computer equipment collects Wi-Fi information data tables from the mobile resource management platform. The main fields of the Wi-Fi information data tables include the second MAC address, Wi-Fi name, and the floor number to which the Wi-Fi belongs.

[0073] Step 102: Input the first feature data and the generated first initial centroid into the centroid training model, and output the first clustering result.

[0074] In this step, the computer device performs random initialization processing on the first feature data to generate first initial centroids. The computer device then builds a centroid model based on a clustering algorithm and trains the centroid model using the first feature data to generate a centroid training model. After inputting the first feature data and the first initial centroids into the centroid training model, the centroid training model clusters the first feature data according to the first initial centroids, generating the first clustering result. For example, the clustering algorithm includes the K-Means clustering algorithm, which performs clustering based on Euclidean distance.

[0075] Step 103: Perform dimensionality reduction on the first clustering result to generate the first target centroid.

[0076] In this step, since the first clustering result is a vector with one row and multiple columns, it needs to be dimensionality-reduced to generate the first target centroid. The computer device iterates through the first clustering result using a heuristic approach, updating the position of the centroid after each iteration until the centroid stabilizes. The iteration then stops, and the centroid after the last iteration is used as the first target centroid. Optionally, the computer device can generate a new dataset based on the indoor distributed antenna system (DAS) cell identification code, the floor number of the Wi-Fi network, and the first target centroid.

[0077] Step 104: Generate anomaly intervals based on the generated first dataset and the first target centroid.

[0078] Step 105: Determine whether there is an anomaly in the feeder line of the target floor based on the collected communication data and abnormal intervals.

[0079] In the technical solution provided by this invention, feature extraction is performed on the sample data collected from each floor to generate first feature data; the first feature data and the generated first initial centroid are input into a centroid training model to output a first clustering result; the first clustering result is subjected to dimensionality reduction processing to generate a first target centroid; anomaly intervals are generated based on the generated first dataset and the first target centroid; and the presence of anomalies in the feeder lines of the target floor is determined based on the collected communication data of the target floor and the anomaly intervals. In the technical solution provided by this invention, computer equipment can quickly determine whether there are anomalies in the feeder lines of the target floor based on the generated anomaly intervals, saving testing costs, reducing testing difficulty, and improving testing efficiency.

[0080] Figure 2 A flowchart of a feature extraction method provided in an embodiment of the present invention is shown below. Figure 2 As shown, step 101 may specifically include:

[0081] Step 1011: Connect the MDT measurement data table and the DPI acquisition data table to generate a second dataset.

[0082] In this step, the main fields of the second dataset include the cell identification code of the indoor distributed antenna system (DAS), the RSRP of the indoor DAS, the SINR of the indoor DAS, the cell identification code of the neighboring cell, the RSRP of the neighboring cell, and the first MAC address.

[0083] As shown in Table 1 below, Table 1 presents the MDT measurement data.

[0084] Table 1

[0085] First IMSI First Data IMSI 1 Data 1 IMSI 2 Data 2 IMSI 3 Data 3

[0086] As shown in Table 1 above, the MDT measurement data table includes the first IMSI and the first data corresponding to the first IMSI. The first data includes the cell identification code, indoor cell RSRP, indoor cell SINR, neighboring cell cell identification code, and neighboring cell RSRP. The first IMSI includes IMSI 1, IMSI 2, and IMSI 3, and the first data includes data 1, data 2, and data 3. Specifically, the first data corresponding to IMSI 1 is data 1, the first data corresponding to IMSI 2 is data 2, and the first data corresponding to IMSI 3 is data 3.

[0087] As shown in Table 2 below, Table 2 shows the DPI acquisition data table.

[0088] Table 2

[0089] Second IMSI Second data IMSI 2 Data 4 IMSI 3 Data 5 IMSI 1 Data 6

[0090] As shown in Table 2 above, the DPI acquisition data table includes the second IMSI and the second data corresponding to the second IMSI. The second data includes the first MAC address. The second IMSI includes IMSI 2, IMSI 3, and IMSI 1, and the second data includes data 4, data 5, and data 6. Specifically, the second data corresponding to IMSI 2 is data 4, the second data corresponding to IMSI 3 is data 5, and the second data corresponding to IMSI 1 is data 6.

[0091] As shown in Tables 1 and 2 above, the first IMSI in Table 1 corresponds one-to-one with the second IMSI in Table 2.

[0092] As an optional approach, the computer device retrieves the second data corresponding to the first IMSI from the DPI acquisition data table, and concatenates the first data corresponding to the first IMSI and the second data corresponding to the first IMSI to generate a second dataset. For example, the computer device retrieves data 6 as the second data corresponding to IMSI 1, data 4 as the second data corresponding to IMSI 2, and data 5 as the second data corresponding to IMSI 3 from Table 2. It then concatenates the first data (data 1) corresponding to IMSI 1 with the second data (data 6) corresponding to IMSI 1, the first data (data 2) corresponding to IMSI 2 with the second data (data 4) corresponding to IMSI 2, and the first data (data 3) corresponding to IMSI 3 with the second data (data 5) corresponding to IMSI 3 to generate a second dataset.

[0093] As an alternative, the computer device retrieves the first data corresponding to the second IMSI from the MDT data collection table; it then concatenates the first data corresponding to the second IMSI and the second data corresponding to the second IMSI to generate a second dataset. For example, the computer device retrieves the first data corresponding to IMSI 2 as data 2, the first data corresponding to IMSI 3 as data 3, and the first data corresponding to IMSI 1 as data 1 from Table 1. It then concatenates the first data (data 2) corresponding to IMSI 2 with the second data (data 4) corresponding to IMSI 2, the first data (data 3) corresponding to IMSI 3 with the second data (data 5) corresponding to IMSI 3, and the first data (data 1) corresponding to IMSI 1 with the second data (data 6) corresponding to IMSI 1 to generate a second dataset.

[0094] Step 1012: Filter the MDT measurement data table to generate a third dataset.

[0095] In this step, the computer equipment selects data from the MDT measurement data corresponding to the cell identification codes of the indoor distributed antenna system (DAS) cells in the second dataset. Then, it generates a third dataset based on the data corresponding to the cell identification codes of the indoor DAS cells in the second dataset and the set invalid data. Invalid data includes data with empty fields and / or data where the RSRP of the indoor DAS cells is less than a preset threshold. Since areas where the RSRP of the indoor DAS cells is less than the preset threshold are weak coverage areas, there is no gradient gain for the algorithm to fit positive and negative samples; therefore, the RSRP data in areas where the RSRP of the indoor DAS cells is less than the preset threshold is considered invalid data.

[0096] Step 1013: Connect the third dataset and the Wi-Fi information data table to generate the fourth dataset.

[0097] In this step, the main fields of the fourth dataset include the cell identification code of the indoor distributed antenna system (DAS), the RSRP of the indoor DAS, the SINR of the indoor DAS, the cell identification code of the neighboring cell, the RSRP of the neighboring cell, the MAC address of the Wi-Fi (first MAC address or second MAC address), the Wi-Fi name, and the floor number to which the Wi-Fi belongs.

[0098] As an alternative, the computer device queries the Wi-Fi information data table to find the fourth data corresponding to the first MAC address, connects the third data corresponding to the first MAC address and the fourth data corresponding to the first MAC address to generate the first connection data, and generates the fourth dataset based on the first MAC address and the first connection data.

[0099] As an alternative, the computer device retrieves the third data corresponding to the second MAC address from the third dataset, concatenates the third data corresponding to the second MAC address with the fourth data corresponding to the second MAC address to generate second concatenation data, and generates the fourth dataset based on the second MAC address and the second concatenation data.

[0100] Step 1014: Merge the generated first and fourth datasets to generate the fifth dataset.

[0101] In this step, the main fields of the fifth dataset include the cell identification code of the indoor distributed antenna system (DAS), the RSRP of the indoor DAS, the SINR of the indoor DAS, the cell identification code of the neighboring cell, the RSRP of the neighboring cell, the MAC address of the Wi-Fi (first MAC address or second MAC address), the Wi-Fi name, the floor number to which the Wi-Fi belongs, and the line loss enumeration value.

[0102] Step 1015: Select the first feature data from the fifth dataset according to the preset selection rules.

[0103] In this step, the computer device selects a field from the fields of the fifth dataset as a feature, sorts the data in the fifth dataset according to the selected feature, and generates a sorted fifth dataset. Then, it selects the first feature data from the sorted fifth dataset according to a preset selection rule. If the number of data in the fifth dataset is less than the number of data set in the selection rule, the fifth dataset is padded with zeros to select the first feature data. For example, if the feature selected by the computer device is the neighboring cell RSRP, the selection rule is to sort the fifth dataset in descending order of neighboring cell RSRP, and select the data corresponding to the first 3 RSRPs from the sorted fifth dataset as the first feature data.

[0104] In the technical solution provided by this invention, the MDT measurement data table and the DPI acquisition data table are connected to generate a second dataset; the MDT measurement data table is filtered to generate a third dataset; the third dataset is connected to the Wi-Fi information data table to generate a fourth dataset; the generated first and fourth datasets are merged to generate a fifth dataset; and first feature data is selected from the fifth dataset according to a preset selection rule. The computer device extracts features from the sample data to obtain more representative data, thereby obtaining a more accurate anomaly range and improving the accuracy of the line anomaly judgment result.

[0105] Figure 3 A flowchart of a first dataset generation method provided in an embodiment of the present invention is shown below. Figure 3 As shown, before step 1014, the following steps are also included:

[0106] Step 201: Determine whether the antenna line loss of each floor covered by the indoor distributed antenna system is less than or equal to a set threshold. If the antenna line loss of the floor is less than or equal to the set threshold, proceed to step 202; if the antenna line loss of the floor is greater than the set threshold, proceed to step 203.

[0107] In this step, the computer equipment uses line loss enumeration values ​​to reflect the integrity of the feeder lines within a floor. These enumeration values ​​include either intact or incomplete enumeration values. If the antenna line loss of a floor is determined to be less than or equal to a set threshold, it indicates that there are no antennas with substandard line loss within the floor, and step 202 is executed. If the antenna line loss of a floor is determined to be greater than the set threshold, it indicates that there are antennas with substandard line loss within the floor, and step 203 is executed.

[0108] Step 202: Mark the floors according to the intact enumeration values.

[0109] Step 203: Mark the floors according to the incomplete enumeration values.

[0110] Step 204: Generate the first dataset based on the cell identification code of the indoor distributed antenna system (DAS) cell and the line loss enumeration values ​​of each floor covered by the indoor DAS cell.

[0111] In this step, the main fields of the first dataset include the cell identification code of the indoor distributed antenna system (DAS) and the line loss enumeration values ​​for each floor covered by the DAS. For example, if the DAS covers three floors, namely floor 1, floor 2, and floor 3, and the computer determines that the antenna line loss of floor 1 is greater than a set threshold, while the antenna line loss of floors 2 and 3 is less than or equal to the set threshold, then floor 1 is marked according to the incomplete enumeration value, and floors 2 and 3 are marked according to the complete enumeration value. In this case, the first dataset includes the cell identification code, complete enumeration value, and incomplete enumeration value of the DAS corresponding to the three floors. Alternatively, if the antenna line loss of floors 1, 2, and 3 are all less than or equal to the set threshold, then floors 1, 2, and 3 are marked according to the complete enumeration value. In this case, the first dataset includes the cell identification code and complete enumeration value of the DAS corresponding to the three floors. For example, if the antenna line loss of floors 1, 2, and 3 is greater than a set threshold, then floors 1, 2, and 3 are marked according to the incomplete enumeration value. At this time, the first dataset includes the cell identification code of the indoor distributed antenna system corresponding to the three floors and the incomplete enumeration value.

[0112] In the technical solution provided by this invention, a computer device determines whether the antenna line loss of each floor covered by an indoor distributed antenna system (DAS) is less than or equal to a set threshold. If the antenna line loss of a floor is determined to be less than or equal to the set threshold, the floor is marked according to the intact enumeration value. If the antenna line loss of a floor is determined to be greater than the set threshold, the floor is marked according to the incomplete enumeration value. A first dataset is generated based on the cell identification code of the indoor DAS and the line loss enumeration value of each floor covered by the indoor DAS. The computer device marks the integrity of the feeders within the floors covered by the indoor DAS to generate the first dataset, which facilitates the computer device in determining abnormal intervals based on the first dataset, thereby improving testing efficiency and the accuracy of line anomaly judgment results.

[0113] Figure 4 A flowchart of an abnormal interval generation method provided in an embodiment of the present invention is shown below. Figure 4 As shown, step 104 specifically includes:

[0114] Step 1041: Generate line anomaly indicators for sample data based on the generated first dataset and the first target centroid.

[0115] Step 1042: Select the minimum and maximum values ​​from the line anomaly indicators in the sample data, and use the minimum and maximum values ​​as the two endpoints of the anomaly interval.

[0116] Step 1043: Generate the abnormal interval based on the two endpoints of the abnormal interval.

[0117] In the technical solution provided by the embodiments of the present invention, the computer equipment determines the abnormal range based on the line abnormality index of the sample data, and quickly determines whether there is an abnormality in the feeder line of the target floor based on the abnormal range, which saves testing costs, reduces testing difficulty, and improves testing efficiency.

[0118] Figure 5 A flowchart of a method for generating line anomaly indicators provided in an embodiment of the present invention is shown below. Figure 5 As shown, step 1041 specifically includes:

[0119] Step S11: Calculate the first target centroid according to the area formula to generate the area of ​​the first centroid.

[0120] In this step, the computer equipment uses Principal Component Analysis (PCA) to calculate the area of ​​the first target centroid within the plane formed by the indoor distributed antenna system's RSRP and SINR dimensions, based on the area formula. The first target centroid includes multiple centroids from the first clustering result, where each cluster in the first clustering result includes one centroid. For example, the first target centroid includes a first centroid and a second centroid. The area formula includes:

[0121]

[0122] Where Sc_f represents the area of ​​the first centroid, x i The x-coordinate of the first centroid, y i The ordinate of the first centroid, x i+1 The x-coordinate of the second centroid, y i+1 The ordinate represents the second centroid.

[0123] Step S12: Divide the area of ​​the first centroid according to the generated first dataset to determine the area of ​​the first normal centroid and the area of ​​the first abnormal centroid.

[0124] In this step, the area of ​​the first centroid is divided according to the line loss enumeration values ​​in the first dataset. The area of ​​the first normal centroid includes the area formed by centroids whose line loss enumeration values ​​are intact, corresponding to the floors. The area of ​​the first abnormal centroid includes the area formed by centroids whose line loss enumeration values ​​are incomplete, corresponding to the floors.

[0125] Step S13: Select the smallest centroid area from the first normal centroid area and use the smallest centroid area as the floor standard reference surface of the indoor distribution cell.

[0126] In this step, the standard reference plane for the floor includes:

[0127] Sc_f_ref = min(Sc_f_g)

[0128] Where Sc_f_ref represents the floor standard reference plane, and Sc_f_g represents the area of ​​the first normal centroid.

[0129] The computer equipment conducted experiments using the average centroid area and the minimum centroid area of ​​the first normal centroid area as the standard reference surface for the floor. The experimental results showed that using the minimum centroid area of ​​the first normal centroid area as the standard reference surface for the floor had better experimental results. Therefore, in this embodiment of the invention, the minimum centroid area is used as the standard reference surface for the floor for calculation.

[0130] Step S14: Calculate the floor standard reference surface and the area of ​​the first abnormal centroid according to the standard ratio formula to generate the line anomaly index of the sample data.

[0131] In this step, the standard ratio formula includes:

[0132]

[0133] Wherein, sRc_f_d represents the line anomaly index of the sample data, Sc_f_d represents the area of ​​the first anomaly centroid, and Sc_f_ref represents the floor standard reference surface.

[0134] In the technical solution provided by the embodiments of the present invention, the computer equipment can determine the abnormal range based on the line abnormality index of the sample data, and quickly determine whether there is an abnormality in the feeder line of the target floor based on the abnormal range, thereby saving testing costs, reducing testing difficulty, and improving testing efficiency.

[0135] Figure 6 A flowchart of a line anomaly detection method provided by an embodiment of the present invention is shown below. Figure 6 As shown, step 105 specifically includes:

[0136] Step 1051: Extract features from the collected communication data of the target floor to generate second feature data.

[0137] In this step, the communication data for the target floor includes the RFID data sheet for the target floor, the MDT measurement data sheet for the target floor, the DPI data sheet for the target floor, and the Wi-Fi information data sheet for the target floor.

[0138] Step 1052: Input the second feature data and the generated second initial centroid into the centroid training model, and output the second clustering result.

[0139] In this step, the computer device performs random initialization processing on the second feature data to generate the second initial centroid.

[0140] Step 1053: Perform dimensionality reduction on the second clustering result to generate the second target centroid.

[0141] Step 1054: Generate the line anomaly index for the target floor based on the generated sixth dataset and the second target centroid.

[0142] Step 1055: Determine whether the abnormal line indicator of the target floor is within the abnormal range. If it is determined that the abnormal line indicator of the target floor is within the abnormal range, proceed to step 1056; if it is determined that the abnormal line indicator of the target floor is not within the abnormal range, proceed to step 1057.

[0143] Step 1056: Determine that there are no abnormalities in the feeder lines of the target floor.

[0144] Step 1057: Determine that there is an anomaly in the feeder line of the target floor.

[0145] In the technical solution provided by the embodiments of the present invention, the computer equipment can quickly determine whether there is an anomaly in the feeder line of the target floor based on the line anomaly index of the target floor and the generated anomaly interval, thereby saving testing costs, reducing testing difficulty, and improving testing efficiency.

[0146] In the technical solution provided by this invention, the computer equipment can save 50% of the number of RFID tags affixed to existing RFID-based indoor monitoring systems. Typically, an indoor distribution system requires deployment on two floors, necessitating the deployment of an existing RFID indoor monitoring system. However, in the technical solution provided by this invention, only one floor needs to be equipped with the RFID indoor monitoring system. The computer equipment can use big data analysis to infer the feeder line status of the other floor, saving testing costs and improving testing efficiency.

[0147] Figure 7 A flowchart of another method for generating line anomaly indicators provided in an embodiment of the present invention is shown below. Figure 7 As shown, step 1054 specifically includes:

[0148] Step S21: Calculate the second target centroid according to the area formula to generate the area of ​​the second centroid.

[0149] In this embodiment of the invention, the description of steps S21 to S24 can be found in the description of steps S11 to S14, and will not be repeated here.

[0150] Step S22: Divide the area of ​​the second centroid according to the generated sixth dataset to determine the area of ​​the second normal centroid and the area of ​​the second abnormal centroid.

[0151] Step S23: Select the smallest centroid area from the second normal centroid area and use the smallest centroid area as the floor standard reference surface of the indoor distribution cell.

[0152] Step S24: Calculate the standard reference surface and the area of ​​the second abnormal centroid of the floor according to the standard ratio formula to generate the line anomaly index of the target floor.

[0153] In the technical solution provided by the embodiments of the present invention, the computer equipment can quickly determine whether there is an anomaly in the feeder line of the target floor based on the line anomaly index of the target floor and the generated anomaly interval, thereby saving testing costs, reducing testing difficulty, and improving testing efficiency.

[0154] Figure 8 A flowchart of a sixth dataset generation method provided in an embodiment of the present invention is shown below. Figure 8 As shown, before step 1054, the following steps are also included:

[0155] Step 301: Determine whether the antenna line loss of the target floor is less than or equal to the set threshold. If the antenna line loss of the target floor is less than or equal to the set threshold, proceed to step 302. If the antenna line loss of the target floor is greater than the set threshold, proceed to step 303.

[0156] In this embodiment of the invention, the description of step 301 can be found in the description of step 201, and will not be repeated here.

[0157] Step 302: Mark the target floor according to the intact enumeration value.

[0158] Step 303: Mark the target floor based on the incomplete enumeration values.

[0159] Step 304: Generate the sixth dataset based on the cell identification code of the indoor distributed antenna system corresponding to the target floor and the line loss enumeration value of the target floor.

[0160] In this step, the main fields of the sixth dataset include the cell identification code of the indoor distributed antenna system (DAS) corresponding to the target floor and the line loss enumeration value of the target floor. The line loss enumeration value of the target floor includes either intact or incomplete enumeration values.

[0161] In the technical solution provided by the embodiments of the present invention, the computer device marks the integrity of the feeder lines of the target floor to generate a sixth dataset, which facilitates the computer device to generate line anomaly indicators of the target floor based on the sixth dataset, and to determine whether there are any anomalies in the feeder lines of the target floor based on the line anomaly indicators of the target floor, thereby improving the testing efficiency.

[0162] Figure 9 This is a schematic diagram of a line anomaly determination device provided in an embodiment of the present invention, as shown below. Figure 9 As shown, the device includes a first generation module 11, an output module 12, a second generation module 13, a third generation module 14, and a judgment module 15. The first generation module 11 is connected to the output module 12, the output module 12 is connected to the second generation module 13, the second generation module 13 is connected to the third generation module 14, and the third generation module 14 is connected to the judgment module 15.

[0163] The first generation module 11 extracts features from the collected sample data of each floor to generate first feature data. The output module 12 inputs the first feature data and the generated first initial centroid into the centroid training model and outputs the first clustering result. The second generation module 13 performs dimensionality reduction on the first clustering result to generate the first target centroid. The third generation module 14 generates anomaly intervals based on the generated first dataset and the first target centroid. The judgment module 15 judges whether there are anomalies in the feeder lines of the target floors based on the collected communication data of the target floors and the anomaly intervals. The target floors include floors selected from at least one floor.

[0164] In this embodiment of the invention, the first generation module 11 specifically includes a first generation unit 111, a second generation unit 112, a third generation unit 113, a fourth generation unit 114, and a first selection unit 115. The first generation unit 111 is connected to the second generation unit 112, the second generation unit 112 is connected to the third generation unit 113, the third generation unit 113 is connected to the fourth generation unit 114, and the fourth generation unit 114 is connected to the first selection unit 115.

[0165] The first generation unit 111 connects the MDT measurement data table and the DPI acquisition data table to generate a second dataset. The second generation unit 112 filters the MDT measurement data table to generate a third dataset. The third generation unit 113 connects the third dataset and the Wi-Fi information data table to generate a fourth dataset. The fourth generation unit 114 merges the generated first and fourth datasets to generate a fifth dataset. The first selection unit 115 selects first feature data from the fifth dataset according to a preset selection rule.

[0166] In this embodiment of the invention, the first generation unit 111 is specifically used to query the second data corresponding to the first IMSI from the DPI collection data table; connect the first data corresponding to the first IMSI and the second data corresponding to the first IMSI to generate a second dataset; or, query the first data corresponding to the second IMSI from the MDT collection data table; connect the first data corresponding to the second IMSI and the second data corresponding to the second IMSI to generate a second dataset.

[0167] In this embodiment of the invention, the second generation unit 112 is specifically used to select data corresponding to the cell identification code of the indoor distributed antenna system (DAS) from the MDT measurement data according to the cell identification code of the indoor DAS; and to generate a third dataset based on the data corresponding to the cell identification code of the indoor DAS and the set invalid data.

[0168] In this embodiment of the invention, the third generation unit 113 is specifically configured to: query the fourth data corresponding to the first MAC address from the Wi-Fi information data table; concatenate the third data corresponding to the first MAC address and the fourth data corresponding to the first MAC address to generate first connection data; generate a fourth dataset based on the first MAC address and the first connection data; or, query the third data corresponding to the second MAC address from the third dataset; concatenate the third data corresponding to the second MAC address and the fourth data corresponding to the second MAC address to generate second connection data; and generate a fourth dataset based on the second MAC address and the second connection data.

[0169] In this embodiment of the invention, the device further includes a fourth generation module 16, which is connected to the output module 12. The fourth generation module 16 is used to perform random initialization processing on the first feature data to generate a first initial centroid.

[0170] In this embodiment of the invention, the third generation module 14 includes a fifth generation unit 141, a second selection unit 142, and a sixth generation unit 143. The fifth generation unit 141 is connected to the second selection unit 142, and the second selection unit 142 is connected to the sixth generation unit 143.

[0171] The fifth generation unit 141 is used to generate line anomaly indicators for the sample data based on the generated first dataset and the first target centroid. The second selection unit 142 is used to select the minimum and maximum values ​​from the line anomaly indicators of the sample data, and use the minimum and maximum values ​​as the two endpoints of the anomaly interval. The sixth generation unit 143 is used to generate the anomaly interval based on the two endpoints of the anomaly interval.

[0172] In this embodiment of the invention, the fifth generation unit 141 is specifically used to calculate the first target centroid according to the area formula to generate the first centroid area; divide the first centroid area according to the generated first dataset to determine the first normal centroid area and the first abnormal centroid area; select the smallest centroid area from the first normal centroid area and use the smallest centroid area as the floor standard reference surface of the indoor distributed antenna system; calculate the floor standard reference surface and the first abnormal centroid area according to the standard ratio formula to generate the line anomaly index of the sample data.

[0173] In this embodiment of the invention, the device further includes a fifth generation module 17, which is connected to the third generation module 14. The fifth generation module 17 is used to determine whether the antenna line loss of each floor covered by the indoor distributed antenna system (DAS) is less than or equal to a set threshold; if the antenna line loss of a floor is determined to be less than or equal to the set threshold, the floor is marked according to the intact enumeration value; if the antenna line loss of a floor is determined to be greater than the set threshold, the floor is marked according to the incomplete enumeration value; a first dataset is generated according to the cell identification code of the indoor DAS and the line loss enumeration values ​​of each floor covered by the indoor DAS, wherein the line loss enumeration values ​​of each floor covered by the indoor DAS include intact enumeration values ​​and / or incomplete enumeration values.

[0174] In this embodiment of the invention, the judgment module 15 includes a seventh generation unit 1501, a first judgment unit 1502, a first determination unit 1503, and a second determination unit 1504. The seventh generation unit 1501 is connected to the first judgment unit 1502, and the first judgment unit 1502 is connected to the first determination unit 1503 and the second determination unit 1504.

[0175] The seventh generation unit 1501 is used to generate line anomaly indicators for the target floor based on the generated sixth dataset and the generated second target centroid. The first judgment unit 1502 is used to determine whether the line anomaly indicators for the target floor are within the anomaly range. The first determination unit 1503 is used to determine that the feeder line of the target floor is not abnormal if the first judgment unit 1502 determines that the line anomaly indicators for the target floor are within the anomaly range. The second determination unit 1504 is used to determine that the feeder line of the target floor is abnormal if the first judgment unit 1502 determines that the line anomaly indicators for the target floor are not within the anomaly range.

[0176] In this embodiment of the invention, the judgment module 15 further includes an eighth generation unit 1505, an output unit 1506, and a ninth generation unit 1507. The eighth generation unit 155 is connected to the output unit 1506, the output unit 1506 is connected to the ninth generation unit 1507, and the ninth generation unit 1507 is connected to the seventh generation unit 1501. The eighth generation unit 1505 is used to extract features from the collected communication data of the target floor to generate second feature data. The output unit 1506 is used to input the second feature data and the generated second initial centroid into the centroid training model and output the second clustering result. The ninth generation unit 1507 is used to perform dimensionality reduction processing on the second clustering result to generate the second target centroid.

[0177] In this embodiment of the invention, the judgment module 15 further includes a tenth generation unit 1508, which is connected to the output unit 1506. The tenth generation unit 1508 is used to perform random initialization processing on the second feature data to generate a second initial centroid.

[0178] In this embodiment of the invention, the seventh generation unit 1501 is specifically used to calculate the second target centroid according to the area formula to generate the second centroid area; divide the second centroid area according to the generated sixth dataset to determine the second normal centroid area and the second abnormal centroid area; select the smallest centroid area from the second normal centroid area and use the smallest centroid area as the floor standard reference surface of the indoor distributed antenna system; calculate the floor standard reference surface and the second abnormal centroid area according to the standard ratio formula to generate the line anomaly index of the target floor.

[0179] In this embodiment of the invention, the judgment module 15 further includes a second judgment unit 1509, a first marking unit 1510, a second marking unit 1511 and an eleventh generation unit 1512, wherein the second judgment unit 1509 is connected to the seventh generation unit 1501.

[0180] In the technical solution provided by this invention, the line anomaly determination device extracts features from the sample data collected from each floor to generate first feature data; inputs the first feature data and the generated first initial centroid into a centroid training model to output a first clustering result; performs dimensionality reduction processing on the first clustering result to generate a first target centroid; generates anomaly intervals based on the generated first dataset and the first target centroid; and determines whether there is an anomaly in the feeder line of the target floor based on the collected communication data of the target floor and the anomaly intervals. In the technical solution provided by this invention, computer equipment can quickly determine whether there is an anomaly in the feeder line of the target floor based on the generated anomaly intervals, saving testing costs, reducing testing difficulty, and improving testing efficiency.

[0181] This invention provides a computer-readable storage medium including a stored program, wherein the program, when running, controls the device containing the computer-readable storage medium to execute the above-described method for determining line anomalies.

[0182] Figure 10 The schematic diagram of a computer device provided in an embodiment of the present invention includes: the computer device 3 of this embodiment includes: a processor 31, a memory 32, and a computer program 33 stored in the memory 32 and executable on the processor 31. When the computer program 33 is executed by the processor 31, it implements the circuit anomaly determination method in the embodiment. To avoid repetition, it will not be described in detail here.

[0183] Computer device 3 includes, but is not limited to, processor 31 and memory 32. Those skilled in the art will understand that... Figure 10This is merely an example of computer device 3 and does not constitute a limitation on computer device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, a network device may also include input / output devices, network access devices, buses, etc.

[0184] The processor 31 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0185] The memory 32 can be an internal storage unit of the computer device 3, such as a hard disk or RAM of the computer device 3. The memory 32 can also be an external storage device of the computer device 3, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or FlashCard equipped on the computer device 3. Furthermore, the memory 32 can include both internal and external storage units of the computer device 3. The memory 32 is used to store computer programs and other programs and data required by network devices. The memory 32 can also be used to temporarily store data that has been output or will be output.

[0186] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0187] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0188] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of the invention pertain.

[0189] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0190] In the embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0191] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for determining line anomalies, characterized in that, The method includes: Feature extraction is performed on the sample data collected from each floor to generate the first feature data; Input the first feature data and the generated first initial centroid into the centroid training model, and output the first clustering result; The first clustering result is subjected to dimensionality reduction processing to generate the first target centroid; Anomaly intervals are generated based on the first generated dataset and the first target centroid. Based on the collected communication data of the target floor and the abnormal interval, it is determined whether there is an abnormality in the feeder line of the target floor. The target floor includes the floor selected from at least one floor. The step of generating anomaly intervals based on the generated first dataset and the first target centroid includes: Based on the generated first dataset and the first target centroid, generate line anomaly indicators for sample data; The minimum and maximum values ​​are selected from the line anomaly indicators of the sample data, and the minimum and maximum values ​​are used as the two endpoints of the anomaly interval. An abnormal interval is generated based on the two endpoints of the abnormal interval; The line anomaly index generated based on the generated first dataset and the first target centroid includes: The area of ​​the first centroid is generated by calculating the first centroid area according to the area formula. The area of ​​the first centroid is divided according to the generated first dataset to determine the area of ​​the first normal centroid and the area of ​​the first abnormal centroid. The smallest centroid area is selected from the first normal centroid area, and the smallest centroid area is used as the floor standard reference surface of the indoor distribution cell. The standard reference surface of the floor and the area of ​​the first abnormal centroid are calculated according to the standard ratio formula to generate the line anomaly index of the sample data.

2. The method according to claim 1, characterized in that, The sample data includes RFID line loss data tables for indoor distributed antenna systems (DAS) cells, minimum drive test (MDT) measurement data tables for indoor DAS cells, deep packet inspection (DPI) data collection tables, and Wi-Fi information data tables. The step of extracting features from the collected sample data from each floor to generate first feature data includes: Connect the MDT measurement data table and the DPI acquisition data table to generate a second dataset; The MDT measurement data table is filtered to generate a third dataset; The third dataset and the Wi-Fi information data table are connected to generate the fourth dataset; The first and fourth datasets are merged to generate the fifth dataset. The first feature data is selected from the fifth dataset according to the preset selection rules.

3. The method according to claim 2, characterized in that, The RFID line loss data table includes the antenna line loss of each floor covered by the indoor distributed antenna system (DAS) and the cell identification code of the indoor DAS. Before merging the generated first and fourth datasets to generate the fifth dataset, the process further includes: Determine whether the antenna line loss of each floor covered by the indoor distributed antenna system is less than or equal to a set threshold. If it is determined that the antenna line loss of a floor is less than or equal to a set threshold, then the floor is marked according to the intact enumeration value; If it is determined that the antenna loss of a floor is greater than a set threshold, the floor is marked according to the defective enumeration value. The first dataset is generated based on the cell identification code of the indoor distributed antenna system (DAS) cell and the line loss enumeration values ​​of each floor covered by the indoor DAS cell. The line loss enumeration values ​​of each floor covered by the indoor DAS cell include intact enumeration values ​​and / or incomplete enumeration values.

4. The method according to claim 1, characterized in that, The step of determining whether there is an anomaly in the feeder line of the target floor based on the collected communication data of the target floor and the abnormal interval includes: Based on the generated sixth dataset and the generated second target centroid, generate the line anomaly index for the target floor; Determine whether the abnormal line indicators of the target floor are within the abnormal range; If the abnormality index of the line on the target floor is determined to be within the abnormal range, then it is determined that there is no abnormality in the feeder line of the target floor. If it is determined that the abnormality index of the line on the target floor is not within the abnormal range, then it is determined that there is an abnormality in the feeder line of the target floor.

5. The method according to claim 4, characterized in that, The step of generating line anomaly indicators for the target floor based on the generated sixth dataset and the generated second target centroid includes: The area of ​​the second centroid is calculated based on the area formula to generate the area of ​​the second centroid. The area of ​​the second centroid is divided according to the generated sixth dataset to determine the area of ​​the second normal centroid and the area of ​​the second abnormal centroid. The smallest centroid area is selected from the second normal centroid area, and the smallest centroid area is used as the floor standard reference surface of the indoor distribution cell. The standard reference surface of the floor and the area of ​​the second abnormal centroid are calculated according to the standard ratio formula to generate the line anomaly index of the target floor.

6. A device for determining line anomalies, characterized in that, The device includes: The first generation module is used to extract features from the sample data collected from each floor and generate the first feature data. The output module is used to input the first feature data and the generated first initial centroid into the centroid training model and output the first clustering result; The second generation module is used to perform dimensionality reduction processing on the first clustering result to generate the first target centroid; The third generation module generates anomaly intervals based on the generated first dataset and the first target centroid. The judgment module determines whether there is an anomaly in the feeder line of the target floor based on the collected communication data of the target floor and the abnormal interval. The target floor includes a floor selected from at least one floor. The third generation module includes a fifth generation unit, a second selection unit, and a sixth generation unit; The fifth generation unit is used to generate line anomaly indicators for sample data based on the generated first dataset and the first target centroid. The second selection unit is used to select the minimum and maximum values ​​from the line anomaly indicators of the sample data, and to use the minimum and maximum values ​​as the two endpoints of the anomaly interval; The sixth generation unit is used to generate an abnormal interval based on the two endpoints of the abnormal interval. The fifth generation unit is specifically used to calculate the first target centroid according to the area formula to generate the first centroid area; divide the first centroid area according to the generated first dataset to determine the first normal centroid area and the first abnormal centroid area; select the smallest centroid area from the first normal centroid area and use the smallest centroid area as the floor standard reference surface of the indoor distributed antenna system; calculate the floor standard reference surface and the first abnormal centroid area according to the standard ratio formula to generate the line anomaly index of the sample data.

7. A computer device, characterized in that, include: One or more processors; Memory; And one or more computer programs, wherein the one or more computer programs are stored in the memory, the one or more computer programs including instructions that, when executed by the device, cause the device to perform the method of any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method of any one of claims 1 to 5.

Images