Off-service cell pre-judgment method and device and medium

Abstract

The invention provides an out-of-service cell pre-judgment method and device and a medium, and relates to the technical field of communication, and the method comprises the steps: obtaining cell out-of-service pre-judgment parameters, which comprise the number of resident users of a cell, the number of complaints of the cell and the configuration of adjacent cells of the cell; pre-judging and analyzing an out-of-service cell according to the cell out-of-service pre-judgment parameter, comprising the following steps of: in response to the situation that the decreasing amplitude of the resident user number of the cell is greater than a first threshold value, determining the cell as a preliminary pre-judged out-of-service cell, and obtaining an out-of-service influence weighted score of the preliminary pre-judged out-of-service cell according to the resident user number of the cell, the complaint number of the cell and the configuration of the adjacent cell of the cell; and according to the out-of-service influence weighted score, formulating a troubleshooting processing scheme for preliminarily pre-judging the out-of-service cell. According to the invention, the out-of-service cell is pre-judged and analyzed through the cell out-of-service pre-judgment parameter, and the possible out-of-service condition of the cell is timely checked and processed according to the pre-judgment analysis result, so that the out-of-service cell is pre-judged in advance.

Description

Technical Field

[0001] The present disclosure at least relates to the field of communication technology, and in particular to a method for predicting a decommissioned cell, a device for predicting a decommissioned cell, and a computer-readable storage medium. Background Art

[0002] The decommissioning of a community has a great impact on network quality and user perception. If a community decommissioning occurs, the operation and maintenance personnel must be contacted as soon as possible for repairs to ensure normal communication for users.

[0003] The existing statistical and analytical methods for decommissioned cells in the industry can only identify cells that have already been decommissioned, but cannot predict and analyze decommissioned cells, or follow up in advance on cells that may be decommissioned, resulting in a decline in the service quality and user experience of cell communications. Summary of the invention

[0004] The technical problem to be solved by the present disclosure is to provide a method for predicting a de-serviced cell, a device for predicting a de-serviced cell, and a computer-readable storage medium in view of the above-mentioned shortcomings, so as to solve the problem of how to predict a de-serviced cell.

[0005] In a first aspect, the present disclosure provides a method for predicting a decommissioned cell, the method comprising:

[0006] Obtain the predicted parameters for cell decommissioning, including the number of permanent users in the cell, the number of complaints in the cell, and the configuration of the cell's neighboring areas;

[0007] Predict and analyze the decommissioned cells based on the cell decommissioning prediction parameters, including:

[0008] In response to a decrease in the number of resident users in the cell being greater than a first threshold, determining the cell as a preliminarily predicted de-serviced cell,

[0009] According to the number of permanent users in the cell, the number of complaints in the cell and the configuration of the cell's neighboring areas, obtain the weighted score of the impact of the initial prediction of the decommissioning of the cell;

[0010] Based on the weighted scores of the impact of decommissioning, formulate preliminary screening and treatment plans for communities that are expected to be decommissioned.

[0011] Furthermore, the cell out-of-service prediction parameters are obtained, including:

[0012] Obtain the number of resident users in the cell yesterday and the N days before yesterday from the core network;

[0013] Obtain the number of complaints about the physical microgrid in the community from the customer complaint platform;

[0014] Obtain the cell antenna configuration, the cell's neighboring cells, and the neighboring cell's decommissioning information from the wireless network management device.

[0015] Furthermore, the cell antenna configuration, the neighboring cells of the cell, and the decommissioning information of the neighboring cells are obtained from the wireless network management device, specifically including:

[0016] Obtain the antenna height, downtilt angle, and vertical beam width of the cell from the wireless network management device;

[0017] Calculate the maximum coverage distance of this cell Dmax (m) = antenna height × Tan (90° - antenna downtilt angle + antenna vertical pattern beam width ÷ 2), the maximum value of Dmax (m) is 1000m;

[0018] Obtain the longitude and latitude of each base station from the wireless network management device, and calculate the neighboring base stations within 1.2Dmax(m) of the cell based on the longitude and latitude;

[0019] The neighboring cells of the neighboring base station toward the base station of the current cell and the de-service information of the neighboring cells are obtained from the wireless network management device.

[0020] Further, in response to the decrease in the number of resident users in the cell being greater than a first threshold, determining the cell as a preliminarily predicted de-serviced cell specifically includes:

[0021] Calculate the average number of resident users in the cell in the previous N days before yesterday;

[0022] Calculate the change ratio between the number of resident users in the cell yesterday and the average number;

[0023] In response to the absolute value of the change ratio being greater than a first threshold, the cell is determined as a preliminarily predicted de-service cell.

[0024] Furthermore, according to the number of resident users in the cell, the number of complaints in the cell and the configuration of the neighboring cells, the weighted score of the decommissioning impact of the cell that is initially predicted to be decommissioned is obtained, including:

[0025] Obtain the user number coefficient of the cell according to the number of permanent users in the cell;

[0026] Obtain the user complaint coefficient of the community according to the number of community complaints;

[0027] Obtain the cell neighboring area decommissioning coefficient according to the cell neighboring area configuration;

[0028] According to the cell user number coefficient, cell user complaint coefficient and cell neighboring cell decommissioning coefficient, a weighted score of the decommissioning impact of the cell that is initially predicted to be decommissioned is obtained.

[0029] Further, wherein:

[0030] Cell user number coefficient T yh =1+(T Nyh ÷M), T NyhIt is the average number of permanent users in a cell in the N days before yesterday, when the cell was initially predicted to be out of service. M is a benchmark number of permanent users in the cell.

[0031] Community user complaint coefficient K ts =1+(K Nts ÷N), K Nts It is the sum of the number of complaints about a community that is initially predicted to be decommissioned in the previous N days before yesterday;

[0032] Neighboring cell outage coefficient K n =1+N tf , N tf is the number of de-serviced cells in the neighboring cells of a cell that is initially predicted to be de-serviced;

[0033] Weighted score of the impact of withdrawal w =T yh ×K ts ×K n .

[0034] Furthermore, according to the weighted scores of the impact of decommissioning, a preliminary investigation and treatment plan for the communities that are expected to be decommissioned is formulated, including:

[0035] According to the weighted scores of their respective out-of-service impacts, the preliminarily predicted out-of-service communities in the region are ranked;

[0036] According to the order of the preliminary predicted out-of-service communities, the community out-of-service investigation and processing tasks are pushed to the operation and maintenance personnel.

[0037] Furthermore, the method further comprises:

[0038] Compare and record the number of resident users and wireless network performance indicators of each cell that is initially predicted to be decommissioned before and after the investigation and processing.

[0039] In a second aspect, the present disclosure provides a device for predicting a decommissioned cell, the device comprising:

[0040] The network monitoring module is used to obtain the prediction parameters of cell decommissioning, including the number of resident users in the cell, the number of cell complaints, and the configuration of the cell's neighboring areas;

[0041] The intelligent analysis module is connected to the network monitoring module and is used to predict and analyze the decommissioned cells according to the cell decommissioning prediction parameters, including:

[0042] a preliminary prediction unit, configured to determine the cell as a preliminarily predicted decommissioned cell in response to a decrease in the number of resident users in the cell being greater than a first threshold,

[0043] The impact analysis unit is connected to the preliminary prediction unit and is used to obtain a weighted score of the decommissioning impact of the preliminarily predicted decommissioning cell according to the number of resident users in the cell, the number of complaints in the cell, and the configuration of the neighboring cells of the cell;

[0044] The investigation and processing module is connected to the intelligent analysis module and is used to formulate an investigation and processing plan for the preliminary predicted decommissioning community based on the weighted score of the decommissioning impact.

[0045] In a third aspect, the present disclosure provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the method for predicting a decommissioned cell as described above is implemented.

[0046] The present disclosure provides a decommissioned cell prediction method, a decommissioned cell prediction device and a computer-readable storage medium, which predict and analyze decommissioned cells through cell decommissioning prediction parameters, including preliminarily predicting decommissioned cells according to a decrease in the number of resident users in the cell, and analyzing the impact of cell decommissioning according to the number of resident users in the cell, the number of cell complaints and the configuration of neighboring cells, taking the preliminarily predicted decommissioning impact of the decommissioned cell as a prediction analysis result, and timely troubleshooting and handling possible cell decommissioning situations according to the prediction analysis result, so as to achieve early prediction of decommissioned cells. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a flow chart of a method for predicting a decommissioned cell according to an embodiment of the present disclosure;

[0048] Figure 2 It is a structural schematic diagram of a device for predicting a decommissioned cell according to an embodiment of the present disclosure;

[0049] Figure 3 It is a structural schematic diagram of another device for predicting a decommissioned cell according to an embodiment of the present disclosure;

[0050] Figure 4 is a flow chart of a method for determining a decommissioned cell according to an embodiment of the present disclosure;

[0051] Figure 5 is a schematic diagram of a cell coverage distance calculation according to an embodiment of the present disclosure;

[0052] Figure 6 is a schematic diagram of a neighboring cell of a cell according to an embodiment of the present disclosure;

[0053] Figure 7 is a schematic diagram of a geographical microgrid according to an embodiment of the present disclosure;

[0054] Figure 8 It is a schematic diagram of predicting neighboring cells around a de-serviced cell according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0055] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the embodiments of the present disclosure will be further described in detail below with reference to the accompanying drawings.

[0056] It should be understood that the specific embodiments and drawings described herein are only used to explain the present disclosure rather than to limit the present disclosure.

[0057] It can be understood that, in the absence of conflict, the various embodiments of the present disclosure and the various features in the embodiments can be combined with each other.

[0058] It can be understood that, for the convenience of description, the drawings of the present disclosure only show parts related to the present disclosure, while parts irrelevant to the present disclosure are not shown in the drawings.

[0059] It can be understood that each module or unit involved in the embodiments of the present disclosure may correspond to only one entity structure, or may be composed of multiple entity structures, or multiple modules or units may be integrated into one entity structure.

[0060] It will be understood that, in the absence of conflict, the functions and steps marked in the flowcharts and block diagrams of the present disclosure may occur in an order different from that marked in the drawings.

[0061] It is understood that the flowcharts and block diagrams of the present disclosure illustrate the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to the various embodiments of the present disclosure. Each box in the flowchart or block diagram may represent a module, unit, program segment, or code, which contains executable instructions for implementing the specified functions. Moreover, each box or combination of boxes in the block diagram and flowchart may be implemented by a hardware-based device that implements the specified functions, or may be implemented by a combination of hardware and computer instructions.

[0062] It can be understood that the modules and units involved in the embodiments of the present disclosure can be implemented by software or hardware, for example, the modules and units can be located in a processor.

[0063] Embodiment 1:

[0064] like Figure 1 As shown, the present disclosure provides a method for predicting a decommissioned cell, the method comprising:

[0065] S1. Obtaining the predicted parameters of cell decommissioning, including the number of permanent users in the cell, the number of complaints in the cell, and the configuration of the cell's neighboring areas;

[0066] S2. Predict and analyze the decommissioned cells according to the cell decommissioning prediction parameters, including:

[0067] In response to a decrease in the number of resident users in the cell being greater than a first threshold, determining the cell as a preliminarily predicted de-serviced cell,

[0068] According to the number of permanent users in the cell, the number of complaints in the cell and the configuration of the cell's neighboring areas, obtain the weighted score of the impact of the initial prediction of the decommissioning of the cell;

[0069] S3. Based on the weighted scores of the impact of decommissioning, formulate preliminary screening and handling plans for communities that are expected to be decommissioned.

[0070] In this embodiment, the cell decommissioning prediction parameters are used to predict and analyze the decommissioning cells, including preliminarily predicting the decommissioning cells according to the decrease in the number of resident users in the cells, and analyzing the impact of the cell decommissioning according to the number of resident users in the cells, the number of cell complaints, and the configuration of the cell neighboring areas. The decommissioning impact of the preliminarily predicted decommissioning cells is used as the prediction analysis result, and the possible cell decommissioning situations are promptly checked and handled according to the prediction analysis result, so as to realize the early prediction of the decommissioning cells. The method can be applied to Figure 2 The device for predicting the decommissioning of a cell shown in the figure comprises a network monitoring module for executing step S1, an intelligent analysis module for executing step S2 and a troubleshooting module for executing step S3, wherein the intelligent analysis module comprises a preliminary prediction unit and an impact analysis unit.

[0071] Specifically, with the rapid development of 5G (fifth generation mobile communication technology) networks, various new services have put forward higher requirements for network quality, network reliability, and fault handling and recovery time. In daily network maintenance and network optimization, the decommissioning of cells has a great impact on network quality and user perception. For large cities, 5G networks usually have tens of thousands of cells. Every day, due to network failures, network adjustments and other factors, a considerable number of 5G decommissioned / zero service cells will be generated.

[0072] In the present disclosure, the predicted decommissioning of cells includes three situations: cell failure exiting service, cell zero service, and cell service sharply reduced. For the convenience of description, the decommissioning / zero service cell is referred to as the decommissioning cell. The complete decommissioning of a cell means that the cell has no communication signal and cannot perform normal signal transmission and reception. It is generally caused by transmission failure, carrier frequency failure, antenna feed failure, power outage or high temperature. The cell zero service means that the cell has a signal but cannot absorb the service. The cell is not completely decommissioned. It may be caused by cell parameter configuration errors, base station / cell software abnormalities and other problems. The sharp decline in cell service is due to the decline in cell equipment performance or antenna feed system problems, resulting in an abnormal and sharp decline in service volume. In all the above situations, it is necessary to contact the operation and maintenance personnel in time for repair to ensure the normal communication of users.

[0073] In view of this, this embodiment provides Figure 3The 5G network pre-judgment and decommissioning cell analysis and investigation device shown is a decommissioning cell pre-judgment device, which is a "network monitoring module", "data management module", "intelligent analysis module" and "investigation and processing module" are set through computer programming. Among them:

[0074] The "network monitoring module" monitors and collects statistics on the "number of permanent users in the cell" of the 5G cell from the 5G core network as an evaluation indicator for decommissioned / zero-service cells; collects statistics on the business volume, neighboring cell configuration, etc. of the 5G cell from the 5G wireless network management, which can include cell wireless performance monitoring, base station and cell alarm monitoring, etc.; collects statistics on the number of complaints about the physical micro-grids in the cell's location from the customer complaint platform; and provides this to the "intelligent analysis module" for weighted calculation of the impact of each 5G predicted decommissioned cell, so as to prioritize them.

[0075] The "data management module" is used for database storage and call of relevant data, which can include storage and call of wireless records of all cells in the entire network, configuration tables of neighboring cells in the entire network, opening / expansion records, site removal / cell closure records, wireless optimization data, historical decommissioning processing records, experience databases, etc.

[0076] The "intelligent analysis module" realizes the predicted decommissioning identification of 5G cells, including the analysis, calculation and output of "preliminary identification of cells to be decommissioned", "cell user quantity coefficient", "micro-grid user complaint coefficient" and "neighboring cell decommissioning coefficient". Finally, the "weighted score calculation of cells to be decommissioned" is performed based on the above coefficients.

[0077] The "troubleshooting and processing module" is used to prioritize the processing of 5G predicted decommissioned cells, evaluate processing effects, record historical fault processing, and maintain the experience database.

[0078] More specifically, a method for judging cell decommissioning that can be used for reference and comparison is as follows: Figure 4 As shown, it mainly determines whether the cell is a zero-service cell by counting the cell traffic volume, and then determines whether corresponding troubleshooting is required to restore the service. The main process includes:

[0079] Step 1: Extract the network traffic data of all 5G cells in the network from the day before yesterday (hereinafter referred to as the first 5 days) and yesterday from the statistics of each network management.

[0080] Step 2: Compare the business volume of all 5G cells in the entire network yesterday and in the previous five days. If the cell had business in the previous five days (every day) but had no business yesterday, it is judged as a newly decommissioned cell.

[0081] Step 3: Output the list of newly added decommissioned communities.

[0082] Step 4: Filter out the decommissioned cells that do not need to be processed according to the whitelist cell summary table. Note: Decommissioned cells that are demolished, reduced in capacity, or locked for observation according to the network planning professional plan are called whitelist cells and do not need to be followed up.

[0083] Step 5: Analyze each out-of-service / zero-service cell. The reasons for out-of-service / zero-service are usually as follows:

[0084] 1) Cell failure;

[0085] 2) The cell network management data is incorrectly configured or deleted;

[0086] 3) Cell wireless optimization parameters / strategies are incorrectly configured;

[0087] For the above issues, submit them to operation and maintenance, construction, and network optimization related professionals for follow-up processing based on the specific circumstances.

[0088] In addition, newly discovered decommissioned / zero-service cells that meet the whitelist conditions and do not require follow-up are analyzed and added to the whitelist cell list.

[0089] Step 6: Operation and maintenance, construction, network optimization and other professionals follow up on the processing of decommissioned / zero-service communities and provide feedback upon completion.

[0090] Step 7: Network optimization professionals follow up on the service recovery status of each decommissioned / zero-service cell. If the service has been restored, a closed loop is achieved.

[0091] Step 9: The cells that have not been closed-loop decommissioned / zero-service cells will continue to be included in the list of long-term decommissioned / zero-service cells for follow-up (marked as long-term decommissioned / zero-service cells).

[0092] like Figure 4 The statistical analysis, troubleshooting method and process of 5G decommissioned / zero-service cells shown have the following shortcomings:

[0093] 1. There are many steps in the analysis, investigation, processing, feedback and verification of 5G decommissioning / zero-service cells, and the processing time is long.

[0094] 2. Most of the links are handled manually, and report statistics and analysis require a lot of time and resources of network optimization engineers.

[0095] 3. There is a lack of a directly accessible database, and it is necessary to manually compile statistics for current and historical network data and generate reports for comparative analysis.

[0096] 4. The analysis process relies heavily on the work experience of network optimization engineers. There are many factors that need to be analyzed and checked, which takes a lot of time and effort, and it is difficult to analyze and process in large batches.

[0097] 5. It is difficult to quickly conduct analysis and follow-up investigations based on real-time changes in the network.

[0098] 6. After the closure of the decommissioned / zero-service cell loop, there is a lack of database tools to record the processing process and conduct subsequent indicator tracking and evaluation, and it is impossible to effectively form an experience database for subsequent network optimization and troubleshooting reference when the decommissioned / zero-service cell recurs.

[0099] This embodiment is aimed at Figure 4 The shortcomings of the method shown above focus on solving the following problems:

[0100] 1. Use the "maximum number of resident users in the community per day" (hereinafter referred to as: the number of resident users in the community) indicator instead of the "community business volume" indicator as the evaluation indicator for decommissioned / zero-business communities. Through computer program tools, realize the "pre-judgment" function of decommissioned communities and improve processing efficiency (hereinafter referred to as "pre-judgment of decommissioned communities" analysis and processing).

[0101] 2. Set up a network monitoring module, establish a connection with the platform / server related to the core network and wireless network, directly obtain statistical data, summarize and filter them, and obtain a list of 5G predicted decommissioned cells.

[0102] 3. Set up a database module to realize database storage of real-time and historical data of wireless networks and core networks for rapid call and analysis in subsequent analysis processes.

[0103] 4. Set up an intelligent analysis module to replace most of the manual statistics, analysis, background investigation, and effect evaluation work in the process of predicting and decommissioning community analysis and investigation.

[0104] 5. The weighted algorithm of "weighted score of predicted decommissioned cells" is adopted, which combines wireless network practical experience with computer intelligent algorithms to calculate the weighted score of the impact degree of each predicted decommissioned cell and rank them, so as to guide the accurate investigation of predicted decommissioned cells, greatly improve the investigation efficiency, and save a lot of manpower and material costs.

[0105] 6. The investigation and processing status of the predicted decommissioned cells, the historical / current / subsequent network indicators of the cells, user perception indicators and other data are recorded in the database to form an experience database for reference in subsequent network optimization and analysis and processing of predicted decommissioned cells.

[0106] In one implementation, in step S1, obtaining a cell out-of-service prediction parameter specifically includes:

[0107] Obtain the number of resident users in the cell yesterday and the N days before yesterday from the core network;

[0108] Obtain the number of complaints about the physical microgrid in the community from the customer complaint platform;

[0109] Obtain the cell antenna configuration, the cell's neighboring cells, and the neighboring cell's decommissioning information from the wireless network management device.

[0110] In this embodiment, the cell decommissioning prediction parameters are one of the important technical improvements of the present disclosure. The "network monitoring module" establishes connections with relevant servers and platforms such as the core network, wireless network, and customer complaint information platform, and regularly obtains indicators such as the number of resident users in the cell, realizing network monitoring and data acquisition functions, that is, obtaining 5G decommissioning cell prediction parameters.

[0111] Through the "Data Management Module", Figure 3 The network indicators and related data required by the device shown are managed in a database for instant calling and analysis by other program modules (including the "network monitoring module").

[0112] The "Data Management Module" realizes the automatic storage and call of network indicator statistics and related data, including:

[0113] Statistics on data obtained by the "network monitoring module": mainly including the number of resident users in the 5G cell, the main wireless indicators of the cell, the number of geographical micro-grid points for user complaints, etc. The main wireless indicators of the cell refer to the common wireless indicators such as the cell business volume, wireless connection rate, wireless switching success rate, and wireless drop rate, which are used as references for network optimization engineers to evaluate the wireless quality of the cell;

[0114] Relevant data used for 5G predicted decommissioning cell analysis and subsequent optimization needs: such as 5G full-network cell wireless ledger, 5G full-network neighboring cell configuration table, new site opening records, cell carrier capacity expansion / reduction data records, etc.

[0115] Table 1 below is an example of statistics of the number of resident users in a cell obtained using the above method:

[0116] Table 1 Example of statistics on the number of permanent users in a community

[0117]

[0118] In one implementation, obtaining the cell antenna configuration, the neighboring cells of the cell, and the decommissioning information of the neighboring cells from the wireless network management device specifically includes:

[0119] Obtain the antenna height, downtilt angle, and vertical beam width of the cell from the wireless network management device;

[0120] Calculate the maximum coverage distance of this cell Dmax (m) = antenna height × Tan (90° - antenna downtilt angle + antenna vertical pattern beam width ÷ 2), the maximum value of Dmax (m) is 1000m;

[0121] Obtain the longitude and latitude of each base station from the wireless network management device, and calculate the neighboring base stations within 1.2Dmax(m) of the cell based on the longitude and latitude;

[0122] The neighboring cells of the neighboring base station toward the base station of the current cell and the de-service information of the neighboring cells are obtained from the wireless network management device.

[0123] In this embodiment, to analyze and determine the "neighboring cell out-of-service coefficient" of the 5G predicted out-of-service cell, it is necessary to first determine which neighboring cells the cell has. The method for determining the neighboring cells is as follows: Figure 5 As shown, according to the hanging height, downtilt angle and antenna vertical pattern beam width (3dB) of the base station cell antenna, the cell coverage distance Dmax (m) is calculated as follows: antenna hanging height × Tan (90-antenna downtilt angle + antenna vertical pattern beam width ÷ 2). The units of 90, antenna downtilt angle and antenna vertical pattern beam width in the formula are all "angle".

[0124] When Dmax(m)>1000m, Dmax(m) is set to 1000m. The distance between 4G sites in urban and suburban Guangzhou is basically within 1000m. If it exceeds 1000m, the overlapping coverage area between adjacent sites is usually very small. For special scenarios, the Dmax(m) value can be adjusted appropriately according to the specific situation.

[0125] According to the distribution of 5G network cells in Guangzhou, within the 1.2Dmax(m) distance of the decommissioned cell, the first layer of neighboring cells around the decommissioned cell are usually covered, that is, the neighboring cells with a certain overlapping coverage area and the closest relationship. For the convenience of description, the neighboring cells within the 1.2Dmax(m) distance of the decommissioned cell are referred to as "neighboring cells" for short.

[0126] like Figure 6 The following is a simple schematic diagram. The S1 cell of base station A has a predicted decommissioned cell. The neighboring cells within a distance of 1.2Dmax(m) around the A-S1 cell of base station A are "neighboring cells" ( Figure 6 The gray filled neighboring cells within the dotted line range). Note: Figure 6 This is just a simple definition diagram. The actual network neighbor configuration is much more complicated.

[0127] In one implementation, in step S2, in response to the decrease in the number of resident users in the cell being greater than a first threshold, determining the cell as a preliminarily predicted decommissioned cell specifically includes:

[0128] Calculate the average number of resident users in the cell in the previous N days before yesterday;

[0129] Calculate the change ratio between the number of resident users in the cell yesterday and the average number;

[0130] In response to the absolute value of the change ratio being greater than a first threshold, the cell is determined as a preliminarily predicted de-service cell.

[0131] In this embodiment, if Figure 3 The device for predicting and analyzing a 5G predicted decommissioned cell shown in the figure realizes the analysis and output of the 5G predicted decommissioned cell through an intelligent analysis module, and is suitable for the prediction, analysis, calculation and output of the 5G predicted decommissioned cell.

[0132] For example, the criteria for determining 5G service decommissioning cells are set as follows: the number of "permanent resident users in the cell" was more than 10 in the previous five days, and the decrease in the number of "permanent resident users in the cell" yesterday was ≥ 80%.

[0133] Note: The decrease in the number of permanent users in the community in the previous 5 days and the decrease in the number of permanent users in the community yesterday can be adjusted according to actual needs. 10 is a benchmark number for the number of permanent users in the community, which can be represented by M.

[0134] The "Intelligent Analysis Module" Figure 3 In the database (data management module) shown, various indicators of the local 5G network cells are called to output the "5G predicted decommissioning cell list".

[0135] An example of outputting 5G predicted decommissioning cells is as follows:

[0136] The statistical data of the maximum number of daily resident users in the cell from July 24 to July 29 were retrieved from the database. After screening according to the "5G Predicted Decommissioning Cell Determination Criteria" and filtering according to the "White List Cells", a preliminary list of 5G predicted decommissioning cells that need to be followed up was obtained, as shown in Table 2 below.

[0137] Table 2: Example of a preliminary list of 5G predicted decommissioned cells

[0138]

[0139] In one implementation, in step S2, according to the number of resident users in the cell, the number of complaints in the cell, and the configuration of the neighboring cells of the cell, a weighted score of the decommissioning impact of the cell preliminarily predicted to be decommissioned is obtained, specifically including:

[0140] Obtain the user number coefficient of the cell according to the number of permanent users in the cell;

[0141] Obtain the user complaint coefficient of the community according to the number of community complaints;

[0142] Obtain the cell neighboring area decommissioning coefficient according to the cell neighboring area configuration;

[0143] According to the cell user number coefficient, cell user complaint coefficient and cell neighboring cell decommissioning coefficient, a weighted score of the decommissioning impact of the cell that is initially predicted to be decommissioned is obtained.

[0144] In this embodiment, the number of resident users in the cell, the number of complaints about the cell, and the configuration of neighboring cells are comprehensively considered to take into account the possible impact of the cell decommissioning, which is used as a scientific basis for judging the urgency of the investigation and treatment of the cell decommissioning. The above three aspects are all converted into numerical values ​​to facilitate judgment by the computer program.

[0145] In one embodiment, wherein:

[0146] Cell user number coefficient T yh =1+(T Nyh ÷M), T Nyh It is the average number of permanent users in a cell in the N days before yesterday, when the cell was initially predicted to be out of service. M is a benchmark number of permanent users in the cell.

[0147] Community user complaint coefficient K ts =1+(K Nts ÷N), K Nts It is the sum of the number of complaints about a community that is initially predicted to be decommissioned in the previous N days before yesterday;

[0148] Neighboring cell outage coefficient K n =1+N tf , N tf is the number of de-serviced cells in the neighboring cells of a cell that is initially predicted to be de-serviced;

[0149] Weighted score of the impact of withdrawal w =T yh ×K ts ×K n .

[0150] In this embodiment, an example of calculating the weighted score of the impact of cell decommissioning is as follows:

[0151] 1. Analyze and calculate the "cell user quantity coefficient" and "micro-grid user complaint coefficient" of the 5G predicted decommissioning cell:

[0152] In order to refine the local 5G network optimization work in Guangzhou and the processing of 5G predicted decommissioning cells, a geographical micro-grid management method is adopted to divide the Guangzhou area into more than 30,000 micro-grid areas of about 460*460 meters, and analyze the micro-grid user complaint points. According to the latitude and longitude of the user complaint location, the user complaint can be assigned to the corresponding micro-grid. The more complaints a micro-grid has, the more network quality problems there may be in the micro-grid, and the greater the impact on user perception. It is necessary to prioritize wireless network troubleshooting or network optimization of the micro-grid related coverage cells.

[0153] For example, Figure 7As shown in the figure, Guangzhou-H-Lejiao North Relocation-8455845-2-1-ONR, according to the longitude and latitude of the community, the microgrid is "Guangzhou Haizhu District_6906". Match the microgrid where each 5G predicted decommissioning community is located, and count the number of complaints in the microgrid in the past 5 days as shown in Table 3 below:

[0154] Table 3 Example of statistics on the number of complaints handled by geographical microgrids in residential areas

[0155]

[0156] According to the information in Table 3, the next step is to analyze the impact of the community's predicted service withdrawal based on the average number of resident users in the community in the previous 5 days and the total number of complaints filed by users in the micro-grid where the community is located in the previous 5 days.

[0157] The larger the number of permanent users in a cell, the greater the impact of cell decommissioning on network users. Therefore, the “cell user quantity coefficient” T is used. yh , which is used to calculate the weighted score of "predicted decommissioning community" later.

[0158] T yh Definition: Cell user quantity coefficient T yh =1+(T 5yh ÷10).

[0159] Among them, T 5yh It is the average number of permanent users in the community in the previous 5 days. Usually, the number of permanent users in a community may span a large range. In order to avoid excessive fluctuations in the coefficient, it is divided by 10 for adjustment.

[0160] T yh Calculation example:

[0161] According to the statistical indicators from July 24 to July 28, the average number of permanent users in Guangzhou-H-Lejiaobei Relocation-8455845-2-1-ONR community in the first five days was 76. yh =1+(T 5yh ÷10)=1+(76÷10)=8.6;

[0162] The average number of permanent users in Guangzhou-H-Haizhu Bulinji-8456119-3-1-ONR community in the first five days was 134. yh =1+(T 5yh ÷10)=1+(134÷10)=14.4.

[0163] There may be one or more cells in a micro-grid, or there may be no cells in the micro-grid, and the network coverage is provided by cells located outside the micro-grid.

[0164] The more complaints a microgrid receives, the worse the quality of the cell network (insufficient coverage or insufficient capacity) is. The greater the impact of the cell's service withdrawal on network users, so the "microgrid user complaint coefficient" K is increased. ts , which is used to subsequently calculate the weighted score of the impact of the "predicted decommissioning cell" on the network.

[0165] K ts Definition: Microgrid user complaint coefficient K ts =1+(K 5ts ÷5).

[0166] Among them, K 5ts It is the total number of complaints handled by microgrids in the community in the past five days, divided by 5 to get the average daily number of complaints handled by microgrids in the community.

[0167] K ts Calculation Example:

[0168] According to the statistical indicators from July 24 to July 28, Guangzhou-H-Lijiao North Relocation-8455845-2-1-ONR community received 5 complaints from microgrids in the first 5 days.

[0169] K ts =1+(K 5ts ÷5)=1+(5÷5)=2;

[0170] Guangzhou-H-Haizhu Bulinji-8456119-3-1-ONR community microgrid management complaints in the first 5 days 4, K ts =1+(K 5ts ÷5)=1+(4÷5)=1.8.

[0171] 2. Analyze and determine the "neighboring cell out-of-service coefficient" of the 5G predicted out-of-service cell:

[0172] If the base station where the decommissioned cell is located or a neighboring cell within 1.2Dmax(m) is decommissioned, it will cause greater network quality degradation in this area. In this case, increase the "neighboring cell decommissioning coefficient" K n , which is used to calculate the weighted score of "predicted decommissioning community" later.

[0173] K n Definition: Neighboring cell decommissioning coefficient K n =1+N tf .

[0174] Note: 1 in the above formula represents the predicted decommissioning cell, which is 1. tf , defined as the total number of de-serviced cells in the neighboring cells of the predicted de-serviced cell.

[0175] K nCalculation Example:

[0176] The wireless engineering parameters of Guangzhou-H-Lijiao North Relocation-8455845-2-1-ONR are shown in Table 4 below:

[0177] Table 4 Examples of engineering parameters for a cell

[0178]

[0179] Dmax (m) = antenna height × Tan (90-antenna downtilt angle + antenna vertical pattern beam width ÷ 2) = 21 × Tan (90-10+14÷2) = 400.7 (meters)

[0180] 1.2Dmax(m)=1.2×400.7=480.8(m), then the neighboring cells within 480.7 meters around Guangzhou-H-Leijiao North Relocation-8455845-2-1-ONR community are "neighboring cells", and the geographical distribution is as follows Figure 8 shown.

[0181] 1.2 times Dmax(m) is a general scenario in dense urban areas. It is set based on the base station spacing in a circle near the cell. This value can be adjusted flexibly. When the base station spacing is large, it can be adjusted to values ​​such as 1.5 times or 2 times. The neighboring cells mentioned by wireless network optimization professionals usually refer to the nearest neighboring cells or the neighboring cells with a relatively close distance and frequent wireless handovers.

[0182] From the database "Network Neighborhood Relationship Table", it is found that the "Guangzhou-H-Lejiao North Relocation-8455845-2-1-ONR" community has 37 neighboring communities, see Table 5 below for details:

[0183] Table 5 Examples of neighboring cells around a predicted decommissioned cell

[0184]

[0185] The database contains the longitude and latitude data of each cell in the basic record of the entire network. The longitude and latitude data of each cell can be used to calculate the surface distance between the predicted decommissioned cell source cell and its neighboring cells. The calculation formula is as follows:

[0186]

[0187] Note: In the above formula, latitude is defined as latitude and longitude is defined as longitude. a = latitude1-latitude2, which is the difference between the latitudes of two points; b = longitude1-longitude2, which is the difference between the longitudes of two points; 6378.137 (KM) is the radius of the earth; S is in meters. The longitude and latitude in the above formula need to be converted to "radians" before entering the formula for calculation.

[0188] As shown in Table 5, among the neighboring areas of the retired community Guangzhou-H-Lejiao North Relocation-8455845-2-1-ONR, there are 22 neighboring areas within 1.2Dmax(m)=480.8 meters, and these 22 neighboring areas are its "neighboring areas".

[0189] Among these 22 "neighboring communities", analysis and statistics show that Guangzhou-H-Lijiao Park-8457482-2-1-ONR and Guangzhou-H-Lijiao Park Ertai-8457482-2-3-ONR are currently in the state of community decommissioning.

[0190] K n =1+N tf =K n =1+2=3

[0191] 3. Analysis of the impact of 5G predicted decommissioning of service cells and calculation of weighted scores:

[0192] The weighted scoring method is used to define the "weighted score for predicting the impact of the decommissioning community" S w , evaluate the impact of the pre-arranged and pre-judged decommissioned cells on 5G network users, according to S w The weighted scores are used to sort the predicted de-serviced cells and determine the processing priority.

[0193] S w =T yh ×K ts ×K n , where S w is the weighted score of the impact of predicting the decommissioning of the community; T yh , K ts , K n The definition of is detailed above.

[0194] Example: Calculating the "weighted score of the impact degree of the predicted decommissioning community" S of Guangzhou-H-Lejiao North Relocation-8455845-2-1-ONR w :

[0195] S w =T yh ×K ts ×K n

[0196] =[1+(T 5yh ÷10)]×[1+(K 5ts ÷5)]×(1+N tf )

[0197] =[1+(76÷10)]×[1+(3÷5)]×(1+2)=8.6×1.6×3=41.28 (points)

[0198] In one implementation, S3, according to the weighted score of the out-of-service impact, a preliminary investigation and processing plan for the out-of-service cell is formulated, which specifically includes:

[0199] According to the weighted scores of their respective out-of-service impacts, the preliminarily predicted out-of-service communities in the region are ranked;

[0200] According to the order of the preliminary predicted out-of-service communities, the community out-of-service investigation and processing tasks are pushed to the operation and maintenance personnel.

[0201] In this embodiment, after the predicted de-serviced cells are obtained, the predicted de-serviced cell list is output in order. The "intelligent analysis module" adopts the above algorithm to analyze and check each predicted de-serviced cell in turn, outputs the predicted de-serviced cell list, and sorts them according to the size of the weighted score S. The results are shown in Table 6.

[0202] Table 6 Example of predicted decommissioned cell list (including investigation priority order)

[0203]

[0204] According to Table 6 above, the predicted de-serviced cells are checked and processed, and the output is a list of predicted de-serviced cells. Figure 3 The human-machine communication interface of the device shown pushes the operation and maintenance professionals to arrange on-site investigation and processing, and continuously performs statistical analysis of indicators through the "intelligent analysis module" to verify and evaluate the effectiveness of the investigation and resolution. For example, within one week after the elimination process, the number of resident users and major wireless indicators of the cells that were previously predicted to be decommissioned return to normal, and the closed loop is completed. The list of predicted decommissioned cells can be quickly sorted and output according to their importance, and the decommissioned cells that are ranked high and have a great impact on the network can be quickly processed.

[0205] Example: Guangzhou-H-Lijiao North Relocation-8455845-2-1-ONR predicted decommissioning cell. After professional maintenance personnel replaced the faulty radio frequency unit (RRU), the predicted decommissioning cell resumed normal operation. After a week of observation, the number of resident users and wireless indicators in the cell were normal, completing the closed loop.

[0206] In one embodiment, the method further comprises:

[0207] Compare and record the number of resident users and wireless network performance indicators of each cell that is initially predicted to be decommissioned before and after the investigation and processing.

[0208] In this embodiment, the "data management module" is also responsible for establishing an experience database for analyzing and troubleshooting predicted out-of-service communities. The discovery, analysis, and troubleshooting processes of each predicted out-of-service community are classified and saved in the form of a ledger to establish an experience database for analyzing and troubleshooting predicted out-of-service communities, providing an experience output reference for subsequent analysis and troubleshooting work.

[0209] This embodiment 1 proposes an analysis and algorithm for 5G predicted decommissioning cells, and provides a prediction analysis device for 5G predicted decommissioning cells. By setting a network monitoring module, a database module, and an intelligent analysis module, a technical method for realizing 5G predicted decommissioning cells from network monitoring to efficient prediction analysis and list sorting output is implemented, which is applicable to the prediction, analysis calculation and output of 5G predicted decommissioning cells. The main technical innovations are as follows:

[0210] 1) The "maximum number of resident users in the cell" indicator counted from the 5G core network is used as an evaluation indicator for decommissioned / zero-service cells, and the "prediction" function of decommissioned cells is realized through computer program tools.

[0211] 2) Use the weighted algorithm of "weighted score of predicted decommissioned cells" to calculate the weighted score and ranking of the impact of each predicted decommissioned cell. Among them, for the neighboring cells of the predicted decommissioned cell, the "neighboring cell decommissioning coefficient" K is calculated. n , calculate one of the weighted score coefficients of this "predicted decommissioning community".

[0212] 3) Before 5G generates zero traffic, the cell decommissioning status is predicted through the sudden change in the number of resident users in the cell, and based on the intelligent analysis of various monitoring data, the priority ranking of the impact of the 5G predicted decommissioning cells on users is output. The impact coefficient on users is increased to distinguish the importance and urgency of the problem, and a list of 5G predicted decommissioning problem cells and their processing priority is output, making the processing of 5G decommissioning cells more efficient.

[0213] Embodiment 2:

[0214] like Figure 2 and 3 As shown, the present disclosure provides a device for predicting a decommissioned cell, the device comprising:

[0215] The network monitoring module is used to obtain the prediction parameters of cell decommissioning, including the number of resident users in the cell, the number of cell complaints, and the configuration of the cell's neighboring areas;

[0216] The intelligent analysis module is connected to the network monitoring module and is used to predict and analyze the decommissioned cells according to the cell decommissioning prediction parameters, including:

[0217] a preliminary prediction unit, configured to determine the cell as a preliminarily predicted decommissioned cell in response to a decrease in the number of resident users in the cell being greater than a first threshold,

[0218] The impact analysis unit is connected to the preliminary prediction unit and is used to obtain a weighted score of the decommissioning impact of the preliminarily predicted decommissioning cell according to the number of resident users in the cell, the number of complaints in the cell, and the configuration of the neighboring cells of the cell;

[0219] The investigation and processing module is connected to the intelligent analysis module and is used to formulate an investigation and processing plan for the preliminary predicted decommissioning community based on the weighted score of the decommissioning impact.

[0220] In one implementation, the network monitoring module specifically includes:

[0221] A user quantity monitoring unit is used to obtain the number of resident users in the cell yesterday and the N days before yesterday from the core network;

[0222] A complaint quantity monitoring unit is used to obtain the number of complaints about the physical microgrid in the cell from the customer complaint platform;

[0223] The neighboring cell monitoring unit is used to obtain the cell antenna configuration, the neighboring cells of the cell and the de-service information of the neighboring cells from the wireless network management equipment.

[0224] In one implementation, the neighboring cell monitoring unit specifically includes:

[0225] Antenna configuration monitoring unit, used to obtain the height, downtilt angle, and vertical beam width of the antenna in the cell from the wireless network management device;

[0226] The coverage distance calculation unit is connected to the antenna configuration monitoring unit and is used to calculate the maximum coverage distance of the cell Dmax (m) = antenna hanging height × Tan (90°-antenna downtilt angle + antenna vertical pattern beam width ÷ 2), and the maximum value of Dmax (m) is 1000m;

[0227] A neighbor calculation unit is connected to the coverage distance calculation unit and is used to obtain the longitude and latitude of each base station from the wireless network management device, and obtain the neighbor base stations within 1.2Dmax(m) of the cell according to the longitude and latitude;

[0228] The neighboring cell configuration monitoring unit is connected to the neighboring calculation unit and is used to obtain the neighboring cells of the neighboring base station toward the base station of the cell and the de-service information of the neighboring cells from the wireless network management device.

[0229] In one implementation, the preliminary prediction unit specifically includes:

[0230] An average calculation unit, used to calculate the average number of resident users in the cell in the previous N days before yesterday;

[0231] A change calculation unit, connected to the average calculation unit, for calculating a change ratio between the number of resident users in the cell yesterday and the average number;

[0232] The comparison and determination unit is connected to the change calculation unit, and is used to determine the cell as a preliminarily predicted de-service cell in response to the absolute value of the change ratio being greater than a first threshold.

[0233] In one implementation, the impact analysis unit specifically includes:

[0234] A first coefficient unit, used to obtain a user number coefficient of a cell according to the number of resident users in the cell;

[0235] The second coefficient unit is used to obtain the user complaint coefficient of the cell according to the number of complaints in the cell;

[0236] The third coefficient unit is used to obtain the cell neighboring area decommissioning coefficient according to the cell neighboring area configuration;

[0237] The weighted score unit is connected to the first, second and third coefficient units, and is used to obtain the weighted score of the decommissioning impact of the preliminarily predicted decommissioned cell according to the cell user number coefficient, the cell user complaint coefficient and the cell neighboring cell decommissioning coefficient.

[0238] In one embodiment, wherein:

[0239] The first coefficient unit calculates the cell user number coefficient T yh =1+(T Nyh ÷M), T Nyh It is the average number of permanent users in a cell in the N days before yesterday, when the cell was initially predicted to be out of service. M is a benchmark number of permanent users in the cell.

[0240] The second coefficient unit calculates the cell user complaint coefficient K ts =1+(K Nts ÷N), K Nts It is the sum of the number of complaints about a community that is initially predicted to be decommissioned in the previous N days before yesterday;

[0241] The third coefficient unit calculates the cell neighboring area decommissioning coefficient K n =1+N tf , N tf is the number of de-serviced cells in the neighboring cells of a cell that is initially predicted to be de-serviced;

[0242] Weighted score unit, calculates the weighted score S of the impact of decommissioning w =T yh ×K ts ×K n .

[0243] In one implementation, the troubleshooting module specifically includes:

[0244] A sorting unit is used to sort the cells that are initially predicted to be decommissioned in the area according to the weighted scores of their respective decommissioning impacts;

[0245] The push unit is connected to the sorting unit and is used to push the cell decommissioning investigation and processing tasks to the operation and maintenance personnel according to the arrangement order of the cells preliminarily predicted to be decommissioned.

[0246] In one embodiment, the device further comprises:

[0247] The data management module is connected to the network monitoring module, the intelligent analysis module and the troubleshooting module to compare and record the number of resident users and wireless network performance indicators of each preliminarily predicted decommissioned cell before and after the troubleshooting.

[0248] Embodiment 3:

[0249] Embodiment 3 of the present disclosure provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the method for predicting a decommissioned cell as described in Embodiment 1 or the device for predicting a decommissioned cell as described in Embodiment 2 is implemented.

[0250] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program elements or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0251] In addition, the present disclosure may also provide a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor runs the computer program stored in the memory, the processor executes the decommissioned cell prediction method as described in Example 1. The computer device may be the decommissioned cell prediction device as described in Example 2.

[0252] The memory is connected to the processor, the memory may be a flash memory or a read-only memory or other memory, and the processor may be a central processing unit or a single-chip microcomputer.

[0253] Embodiments 1-3 of the present disclosure provide a method for predicting a decommissioned cell, an apparatus for predicting a decommissioned cell, and a computer-readable storage medium, which predict and analyze decommissioned cells through cell decommissioning prediction parameters, including preliminarily predicting a decommissioned cell based on a decrease in the number of resident users in the cell, and analyzing the impact of cell decommissioning based on the number of resident users in the cell, the number of cell complaints, and the configuration of neighboring cells in the cell, using the preliminarily predicted decommissioning impact of the decommissioned cell as a prediction analysis result, and timely troubleshooting and handling of possible cell decommissioning situations based on the prediction analysis result, thereby realizing early prediction of decommissioned cells.

[0254] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present disclosure, but the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and substance of the present disclosure, and these modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. A method for predicting a decommissioned cell, characterized in that: The method comprises: Obtain the predicted parameters for cell decommissioning, including the number of permanent users in the cell, the number of complaints in the cell, and the configuration of the cell's neighboring areas; Predict and analyze the decommissioned cells based on the cell decommissioning prediction parameters, including: In response to a decrease in the number of resident users in the cell being greater than a first threshold, determining the cell as a preliminarily predicted de-serviced cell, According to the number of permanent users in the cell, the number of complaints in the cell and the configuration of the cell's neighboring areas, obtain the weighted score of the impact of the initial prediction of the decommissioning of the cell; Based on the weighted scores of the impact of decommissioning, formulate preliminary screening and treatment plans for communities that are expected to be decommissioned.

2. The method according to claim 1, characterized in that Get the cell out-of-service prediction parameters, including: Obtain the number of resident users in the cell yesterday and the N days before yesterday from the core network; Obtain the number of complaints about the physical microgrid in the community from the customer complaint platform; Obtain the cell antenna configuration, the cell's neighboring cells, and the neighboring cell's decommissioning information from the wireless network management device.

3. The method according to claim 2, characterized in that Obtain the cell antenna configuration, neighboring cells of the cell, and the decommissioning information of the neighboring cells from the wireless network management device, including: Obtain the antenna height, downtilt angle, and vertical beam width of the cell from the wireless network management device; Calculate the maximum coverage distance of this cell Dmax (m) = antenna height × Tan (90° - antenna downtilt angle + antenna vertical pattern beam width ÷ 2), the maximum value of Dmax (m) is 1000m; Obtain the longitude and latitude of each base station from the wireless network management device, and calculate the neighboring base stations within 1.2Dmax(m) of the cell based on the longitude and latitude; The neighboring cells of the neighboring base station toward the base station of the current cell and the de-service information of the neighboring cells are obtained from the wireless network management device.

4. The method according to claim 2, characterized in that: In response to a decrease in the number of resident users in the cell being greater than a first threshold, determining the cell as a preliminarily predicted de-serviced cell, specifically including: Calculate the average number of resident users in the cell in the previous N days before yesterday; Calculate the change ratio between the number of resident users in the cell yesterday and the average number; In response to the absolute value of the change ratio being greater than a first threshold, the cell is determined as a preliminarily predicted de-service cell.

5. The method according to any one of claims 1 to 4, characterized in that: According to the number of resident users in the cell, the number of complaints in the cell, and the configuration of the cell's neighboring areas, the weighted score of the service withdrawal impact of the cell to be initially predicted to be withdrawn is obtained, including: Obtain the user number coefficient of the cell according to the number of permanent users in the cell; Obtain the user complaint coefficient of the community according to the number of community complaints; Obtain the cell neighboring area decommissioning coefficient according to the cell neighboring area configuration; According to the cell user number coefficient, cell user complaint coefficient and cell neighboring cell decommissioning coefficient, a weighted score of the decommissioning impact of the cell that is initially predicted to be decommissioned is obtained.

6. The method according to claim 5, characterized in that in: Cell user number coefficient T yh =1+(T Nyh ÷M), T Nyh It is the average number of permanent users in a cell in the N days before yesterday, when the cell was initially predicted to be out of service. M is a benchmark number of permanent users in the cell. Community user complaint coefficient K ts =1+(K Nts ÷N), K Nts It is the sum of the number of complaints about a community that is initially predicted to be decommissioned in the previous N days before yesterday; Neighboring cell outage coefficient K n =1+N tf , N tf is the number of de-serviced cells in the neighboring cells of a cell that is initially predicted to be de-serviced; Weighted score of the impact of withdrawal w =T yh ×K ts ×K n .

7. The method according to claim 1, characterized in that According to the weighted scores of the impact of decommissioning, a preliminary investigation and treatment plan for the communities that are expected to be decommissioned is formulated, including: According to the weighted scores of their respective out-of-service impacts, the preliminarily predicted out-of-service communities in the region are ranked; According to the order of the preliminary predicted out-of-service communities, the community out-of-service investigation and processing tasks are pushed to the operation and maintenance personnel.

8. The method according to claim 7, characterized in that The method further comprises: Compare and record the number of resident users and wireless network performance indicators of each cell that is initially predicted to be decommissioned before and after the investigation and processing.

9. A device for predicting a decommissioned cell, characterized in that: The device comprises: The network monitoring module is used to obtain the prediction parameters of cell decommissioning, including the number of resident users in the cell, the number of cell complaints, and the configuration of the cell's neighboring areas; The intelligent analysis module is connected to the network monitoring module and is used to predict and analyze the decommissioned cells according to the cell decommissioning prediction parameters, including: a preliminary prediction unit, configured to determine the cell as a preliminarily predicted decommissioned cell in response to a decrease in the number of resident users in the cell being greater than a first threshold, The impact analysis unit is connected to the preliminary prediction unit and is used to obtain a weighted score of the decommissioning impact of the preliminarily predicted decommissioning cell according to the number of resident users in the cell, the number of complaints in the cell, and the configuration of the neighboring cells of the cell; The investigation and processing module is connected to the intelligent analysis module and is used to formulate an investigation and processing plan for the preliminary predicted decommissioning community based on the weighted score of the decommissioning impact.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the method for predicting a de-serviced cell as described in any one of claims 1 to 8 is implemented.

Images