Cell energy-saving optimization method and device, readable storage medium and program product

Abstract

The invention discloses a cell energy-saving optimization method and device, a readable storage medium and a program product. The method comprises the following steps: acquiring a plurality of characteristics influencing cell signal propagation of a target area; on the basis of the obtained characteristics influencing the signal propagation of the cells and a path loss model, predicting the signal strength and load of each opened cell in the target area after the target cell in the target area executes the closing of the energy-saving strategy; the path loss model is obtained through training by taking the characteristics influencing the signal propagation of the cell as samples and taking antenna transmitting power loss generated on a signal propagation path by a cell corresponding to the characteristics influencing the signal propagation of the cell as labels; based on the predicted signal strength and load, determining an optimization strategy of each open cell, the optimization strategy comprising adjusting antenna parameters of the corresponding open cell; and after the target cell is closed according to the energy-saving strategy, adjusting the antenna parameters of the corresponding open cell according to the optimization strategy.

Description

Technical Field

[0001] This application relates to the field of wireless communication technologies, and in particular, to a method and apparatus for optimizing cell energy saving, a readable storage medium, and a program product. Background Art

[0002] To achieve energy-saving operation of a base station, reduce operating costs, and improve resource utilization rate, certain cells can be turned off under a certain energy-saving strategy, and user terminals of the turned-off cells can be migrated to adjacent cells with strong signals. The current cell energy-saving means is to calculate the actual network signal strength of the cell, verify the actual network coverage rate of adjacent base stations based on the actual network signal strength, and determine whether the target cell goes into sleep based on the actual network coverage rate. Automatic shutdown processing can be achieved without manual operation, thereby reducing the energy consumption of the base station.

[0003] However, this method of cell energy saving through static network coverage cannot cope with load fluctuations or dynamic environmental changes after energy saving. Therefore, the signal quality of the covered network area cannot be guaranteed, and energy saving and network performance cannot be balanced. How to optimize the network area after cell energy saving while ensuring both cell energy saving and signal quality is a problem that needs to be solved currently. Summary of the Invention

[0004] The purpose of the embodiments of this application is to provide a method and apparatus for optimizing cell energy saving, a readable storage medium, and a program product, so as to solve the problem that energy saving and network performance cannot be balanced.

[0005] To solve the above technical problems, this specification is implemented as follows: In a first aspect, a method for optimizing cell energy saving is provided, including: Obtaining multiple characteristics that affect the propagation of cell signals in a target area; Based on the obtained characteristics that affect cell signal propagation and a path loss model, predicting the signal strength and load of each active cell in the target area after the target cell in the target area executes an energy-saving strategy and is turned off. The path loss model is trained with the characteristics that affect cell signal propagation as samples and the antenna transmission power loss generated by the cells corresponding to the characteristics that affect cell signal propagation on the signal propagation path as labels. The characteristics that affect cell signal propagation include a combination of multiple items such as the distance characteristic between the user terminal and the antenna in the active cell, the angle characteristic between the user terminal and the antenna in the active cell, the environmental type of the active cell, the beam width of the antenna, the frequency of the antenna, and the signal downtilt angle of the antenna; Based on the predicted signal strength and load, determining an optimization strategy for each active cell, where the optimization strategy includes adjusting the antenna parameters of the corresponding active cell; After shutting down the target cell according to the energy-saving strategy, adjust the antenna parameters of the corresponding active cells according to the optimization strategy.

[0006] Optionally, based on the obtained characteristics affecting cell signal propagation and the path loss model, predict the signal strength and load of each active cell in the target area after the target cell in the target area executes the energy-saving strategy to shut down, including: Divide the target area into multiple grids of preset units; Based on the obtained characteristics affecting cell signal propagation and the path loss model, respectively predict the signal strength of multiple grids covered by each active cell and the load of the primary serving cell corresponding to each grid after the target cell executes the energy-saving strategy to shut down. The primary serving cell is the active cell with the maximum signal strength among the multiple active cells corresponding to a grid.

[0007] Optionally, based on the obtained characteristics affecting cell signal propagation and the path loss model, respectively predict the signal strength of multiple grids covered by each active cell and the load of the primary serving cell corresponding to each grid after the target cell executes the energy-saving strategy to shut down, including: Obtain the characteristics affecting the signal propagation of multiple active cells corresponding to the target grid; Based on the obtained characteristics and the path loss model, respectively predict the antenna transmission power loss generated by multiple active cells corresponding to the target grid on the signal propagation path; Based on the antenna transmission power configured by the multiple active cells and the corresponding predicted antenna transmission power loss, determine the signal strength of the multiple active cells corresponding to the target grid after the target cell executes the energy-saving strategy to shut down; Determine the active cell with the maximum signal strength corresponding to the target grid as the primary serving cell of the target grid; Respectively predict the original user terminals of the primary serving cell before the target cell is shut down and the migrated user terminals of the primary serving cell after the target cell is shut down to determine the load of the primary serving cell after the target cell executes the energy-saving strategy to shut down.

[0008] Optionally, it further includes: Generate a signal strength map of each active cell in the target area based on the signal strength predicted by the path loss model; Based on the signal strength map, display the signal distribution of each active cell in the target area.

[0009] Optionally, determining the optimization strategy of each active cell based on the predicted signal strength and load includes: Determine the grids with weak network coverage in each active cell based on the signal strength predicted for multiple grids covered by each active cell and a preset signal strength threshold; Determine the primary serving cells with excessive load in each primary serving cell based on the load predicted for the primary serving cell of the multiple grids and a preset load threshold; For each grid with weak network coverage or an overloaded corresponding primary serving cell, determine an optimization strategy for adjusting the antenna parameters of the corresponding primary serving cell.

[0010] Optionally, the step of determining, for each grid with weak network coverage or an overloaded corresponding primary serving cell, an optimization strategy for adjusting the antenna parameters of the corresponding primary serving cell includes: For each grid with weak network coverage or an overloaded corresponding primary serving cell, determine the importance score of each grid, where the importance score is related to the user terminal density included in the grid and the importance degree of the scenario corresponding to the grid; Determine the primary serving cells corresponding to the grids with importance scores higher than a preset score threshold; Based on the combination of the transmit power, azimuth angle, and beam width of the antenna included in the antenna parameters, determine multiple optimization strategies for the primary serving cells corresponding to the grids; Select a target optimization strategy from the multiple optimization strategies for adjusting the antenna parameters of the corresponding primary serving cell, where the target optimization strategy minimizes the total importance score of the grids of each primary serving cell after the corresponding antenna parameters are adjusted.

[0011] In a second aspect, a cell energy-saving optimization device is provided, including: An acquisition module for acquiring multiple characteristics affecting the cell signal propagation in a target area; A prediction module for predicting the signal strength and load of each active cell in the target area after the target cell in the target area executes an energy-saving strategy and is turned off, based on the acquired characteristics affecting the cell signal propagation and a path loss model, where the path loss model is trained with the characteristics affecting the cell signal propagation as samples and the antenna transmit power loss generated by the cells corresponding to the characteristics affecting the cell signal propagation on the signal propagation path as labels, and the characteristics affecting the cell signal propagation include a combination of multiple items such as the distance characteristic between the user terminal and the antenna in the active cell, the angle characteristic between the user terminal and the antenna in the active cell, the environmental type of the active cell, the beam width of the antenna, the frequency of the antenna, and the signal downtilt angle of the antenna; A determination module for determining an optimization strategy for each active cell based on the predicted signal strength and load, where the optimization strategy includes adjusting the antenna parameters of the corresponding active cell; An adjustment module for adjusting the antenna parameters of the corresponding active cell according to the optimization strategy after the target cell is turned off according to the energy-saving strategy.

[0012] In a third aspect, a cell energy-saving optimization device is provided, comprising a processor and a memory, wherein the memory stores a program or instruction that can be executed on the processor, and when the program or instruction is executed by the processor, the steps of the method described in the first aspect are implemented.

[0013] In a fourth aspect, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0014] In a fifth aspect, a computer program product is provided, the computer program product comprising a non-transitory computer-readable storage medium storing a computer program, wherein the computer program is operable to cause a computer to execute the steps of the method described in the first aspect.

[0015] In an embodiment of the present application, by acquiring multiple characteristics of cell signal propagation that affect the target area; based on the acquired characteristics that affect cell signal propagation and a path loss model, predict the signal strength and load of each enabled cell in the target area after the target cell in the target area executes an energy-saving strategy to be shut down, the path loss model is trained using the characteristics that affect cell signal propagation as samples and the antenna transmission power loss generated by the cell corresponding to the characteristics that affect cell signal propagation on the signal propagation path as labels, the characteristics that affect cell signal propagation include a distance characteristic between a user terminal and an antenna in the enabled cell, an angle characteristic between a user terminal and an antenna in the enabled cell, an environment type of the enabled cell, a beam width of the antenna, a frequency of the antenna, and a combination of multiple factors including a signal downtilt angle of the antenna; based on the predicted signal strength and load, determine the optimization strategy for each enabled cell, The optimization strategy includes adjusting the antenna parameters of the corresponding turned-on cell; after the target cell is turned off according to the energy-saving strategy, the antenna parameters of the corresponding turned-on cell are adjusted according to the optimization strategy, thereby performing an energy-saving strategy preview in a simulated environment, evaluating the signal problems that may occur after the implementation of the energy-saving strategy based on the predicted signal strength and load (for example, weak network coverage, excessive load caused by user migration), and proposing corresponding optimization plans based on the energy-saving strategy preview, adjusting antenna parameters for weak coverage areas, so that signal quality problems in the implementation of the energy-saving strategy can be identified in advance and solutions can be provided. In the energy-saving mode, the antenna parameters are automatically adjusted to effectively ensure the signal quality of the target area in the energy-saving mode and the user's network experience. Through simulation and preview, the risk of implementing the energy-saving strategy is reduced, ensuring that the network maintains a stable user experience while saving energy, and ensuring a balance between energy saving and network performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings: Figure 1 It is a schematic flowchart of the cell energy-saving optimization method according to an embodiment of the present application.

[0017] Figure 2 It is a specific example flowchart of the cell energy-saving optimization method according to an embodiment of the present application.

[0018] Figure 3 It is a structural block diagram of the cell energy-saving optimization device according to the first embodiment of the present application.

[0019] Figure 4 It is a structural block diagram of the cell energy-saving optimization device according to the second embodiment of the present application. Detailed implementation manners

[0020] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application. The reference numerals in the accompanying drawings of the present application are only used to distinguish the various steps in the solution and do not limit the execution order of the various steps. The specific execution order shall be subject to the description in the specification.

[0021] To solve the problems existing in the prior art, an embodiment of the present application provides a cell energy-saving optimization method, as Figure 1 shown, including the following steps 102 to 108.

[0022] Step 102, obtain multiple characteristics affecting the cell signal propagation in the target area.

[0023] The target area is a network area that needs to perform cell energy saving and optimization, and can be selected or divided according to the geographical location area. The target area may include multiple cells for providing network communication services to user terminals. The characteristics affecting the cell signal propagation in the target area are extracted from a variety of data sources. The data sources include the basic data of network operation, such as measurement report (MR) data, measurement report original (MRO) data, performance (PM) data of base stations, and configuration data, etc., and also include user terminal data from different environments (such as cities, suburbs, and indoors). The obtained data is cleaned, standardized, and outlier processed to ensure data quality.

[0024] When collecting data, big data tools such as Kafka and Flume can be used to extract data from the network systems corresponding to different types of data sources regularly. When preprocessing data, data cleaning can be performed, that is, deleting missing values, duplicate values and abnormal data, and using mean interpolation or historical data to repair missing values; data standardization, that is, normalizing the data; data grouping, that is, grouping data by time and location to provide a basis for feature extraction.

[0025] Step 104: Based on the obtained features affecting the propagation of cell signals and the path loss model, predict the signal strength and load of each active cell in the target area after the target cell in the target area executes the energy-saving strategy and shuts down. The path loss model is trained with the features affecting the propagation of cell signals as samples and the antenna transmission power loss generated by the cells corresponding to the features affecting the propagation of cell signals on the signal propagation path as labels. The features affecting the propagation of cell signals include a combination of multiple items among the distance feature between the user terminal and the antenna in the active cell, the angle feature between the user terminal and the antenna in the active cell, the environmental type of the active cell, the beam width of the antenna, the frequency of the antenna, and the signal downtilt angle of the antenna.

[0026] In the embodiment of the present application, first, based on the features affecting the propagation of cell signals in the target area obtained in step 102 and the trained path loss model, before the target cell in the target area executes the energy-saving strategy and shuts down, predict the signal strength and load of other cells in the target area that are not shut down due to energy-saving, that is, the active cells that are working normally, if the target cell executes the corresponding energy-saving strategy and shuts down. The energy-saving strategy can be executed for one or more cells. After the corresponding cell executes the energy-saving strategy, its original user terminals will migrate to other adjacent active cells. Correspondingly, the load of the active cells may change.

[0027] The path loss model is used to output the antenna transmission power loss generated by the corresponding cell on the signal propagation path based on the input features affecting the propagation of cell signals. The path loss model is trained with the features affecting the propagation of cell signals as samples and the labels corresponding to each sample. The label is the actual antenna transmission power loss generated by the cell corresponding to the features affecting the propagation of cell signals on the signal propagation path.

[0028] Optionally, the features affecting the propagation of cell signals include a combination of multiple items among the distance feature between the user terminal and the antenna in the active cell, the angle feature between the user terminal and the antenna in the active cell, the environmental type of the active cell, the beam width of the antenna, the frequency of the antenna, and the signal downtilt angle of the antenna.

[0029] Extract various characteristics that affect signal propagation from the collected data sources. By comparing the characteristic parameters of three classic traditional statistical models, such as the Okumura-Hata model, the Cost 231-Hata model, and the SPM model, the following basic characteristics can be obtained, including: a. Distance characteristic: Based on physical knowledge, there is a significant relationship between the strength of radio signals and the propagation distance. The distance d between the antenna position A and the receiving point position B (i.e., the user terminal position) within the cell is actually determined by the relative height difference △h between A and B and the horizontal distance between them. Therefore, , △h, are all regarded as characteristics affecting the signal propagation within the cell.

[0030]

[0031]

[0032] Among them, the three-dimensional coordinates of the antenna position are ( , , ), and the three-dimensional coordinates of the receiving point position are ( , , ).

[0033] b. Angle characteristic β: That is, the angle between the signal propagation path and the line connecting the receiving point position and the antenna position. The signal emitted by the antenna has a certain concentration. The signal on the back of the antenna is usually weaker than the signal at the same angle and the same distance on the front, that is, the strength of the signal, the signal propagation path, is related to the angle between the line connecting the receiving point position and the antenna position. When other conditions are the same, the smaller the angle β between the line connecting the receiving point position and the antenna position and the signal propagation path, the stronger the signal.

[0034] c. Environment type s of the active cell: For example, residential area / college / rural area / office building, etc. Signals in open areas are usually better than those in building-dense areas. Use this environmental type information as an important characteristic for the path loss model to learn. Here, the cell scenario in the engineering parameters in the configuration data can be used as a characteristic and one-hot encoding processing can be performed.

[0035] d. Beam width t of the antenna: The beam width refers to the angular range of the signal radiated by the antenna, usually expressed in degrees. In signal propagation, the beam width has an important impact on the signal coverage area and its strength.

[0036] e. Frequency f of the antenna: That is, the center frequency of the antenna.

[0037] f, the signal downward tilt angle α of the antenna: The signal downward tilt angle is the sum of the electrical downward tilt angle and the mechanical downward tilt angle in the vertical direction of the antenna.

[0038] Furthermore, it can also include the characteristics of log transformation. By performing a log transformation on the horizontal distance between the antenna position A and the receiving point position B, denoted as ; performing a log transformation on the three-dimensional distance d, denoted as ; performing a log transformation on the center frequency of the antenna, denoted as f_log.

[0039] The above are examples of the characteristics affecting the cell signal propagation. By combining multiple of the above characteristics, corresponding samples are obtained, and combined with the actual antenna transmission power loss corresponding to the above characteristics as labels, a path loss model can be trained.

[0040] Similarly, by inputting the characteristics affecting the cell signal propagation in the target area obtained in step 102 into the path loss model, the antenna transmission power loss generated by each active cell on the signal propagation path can be predicted. Based on the antenna transmission power loss output by the path loss model, the signal strength of each active cell in the target area can be predicted.

[0041] If all the above characteristics are used as input characteristics, then (d, △h, s, t, f, α, , , f_log) 11 characteristics are used as the input characteristics of the path loss model of. .

[0042] The path loss model can integrate a feedforward neural network and a gradient boosting decision tree. The feedforward neural network and the gradient boosting decision tree are used as independent models respectively, and are trained with the above characteristics as samples. The outputs of the two independent models are then weighted and summed as the output of the final path loss model.

[0043] That is, use a deep multi-layer feedforward neural network model and design and optimize the overall architecture of the network as well as hyperparameters such as activation units, number of layers, and learning rate by yourself, and apply the backpropagation algorithm based on gradient descent to learn the feedforward neural network prediction model, and at the same time integrate with the prediction results of the gradient boosting decision tree model to establish the final combined path loss model.

[0044] By accurately simulating the signal strength in different signal propagation environments, the predictability and optimization space of signal coverage are improved. The path loss model that combines mathematics and artificial intelligence can adapt to complex environments and significantly improve the performance and generalization ability of traditional models. Through comprehensive analysis using deep learning algorithms, the adaptability of the path loss model to various environmental factors is enhanced.

[0045] The path loss model can be applied to different wireless communication standards, such as Wi-Fi, LTE, etc. In addition, input features can also introduce unstructured data such as social media and weather forecasts to enhance the adaptability of the path loss model to environmental changes. Using the real-time location data of user terminals, dynamic user behavior analysis is carried out to improve the prediction accuracy.

[0046] Based on the solution provided in the above embodiment, optionally, in step 104 above, the predicting of the signal strength and load of each active cell in the target area after the target cell in the target area executes the energy-saving strategy and shuts down, based on the obtained features affecting cell signal propagation and the path loss model, includes: dividing the target area into multiple grids of preset units; based on the obtained features affecting cell signal propagation and the path loss model, respectively predicting the signal strength of multiple grids covered by each active cell and the load of the primary serving cell corresponding to each grid after the target cell executes the energy-saving strategy and shuts down, where the primary serving cell is the active cell with the maximum signal strength among the multiple active cells corresponding to a grid.

[0047] In this embodiment, the signal strength and load of active cells are predicted in terms of grids. Specifically, the map area of the target area is divided into multiple small grids (such as 5m×5m), and after the target cell executes the energy-saving strategy and shuts down, the signal strength of each grid covered by each active cell is independently evaluated, and the load of the primary serving cell corresponding to each grid is predicted. The primary serving cell is the active cell with the maximum signal strength among the multiple active cells corresponding to a grid. Usually, one grid corresponds to one primary serving cell. The primary serving cell is also the adjacent active cell that the user terminal may migrate to after the target cell executes the energy-saving strategy and shuts down, and the load of this cell may change.

[0048] Optionally, based on the obtained characteristics affecting the signal propagation of cells and the path loss model, predict the signal strength of multiple grids covered by each active cell and the load of the primary serving cell corresponding to each grid after the target cell turns off the energy-saving policy, including: obtaining the characteristics affecting the signal propagation of multiple active cells corresponding to the target grid; based on the obtained characteristics and the path loss model, respectively predicting the antenna transmit power loss generated by the multiple active cells corresponding to the target grid on the signal propagation path; based on the antenna transmit power configured by the multiple active cells and the corresponding predicted antenna transmit power loss, determining the signal strength of the multiple active cells corresponding to the target grid after the target cell turns off the energy-saving policy; determining the active cell with the maximum signal strength corresponding to the target grid as the primary serving cell of the target grid; respectively predicting the original user terminals of the primary serving cell before the target cell is not turned off and the migrated user terminals of the primary serving cell after the target cell is turned off, so as to determine the load of the primary serving cell after the target cell turns off the energy-saving policy.

[0049] When dividing, a fixed grid size is used to cover the entire target area. Each grid obtains the relevant input characteristics of the path loss model in the grid dimension according to the characteristics affecting the signal propagation of multiple active cells corresponding to it, such as the user terminal location, the corresponding antenna location, the angular characteristics between the user terminal and the corresponding antenna, the environmental type of the grid, the beam width of the antenna, the frequency of the antenna, and the signal downtilt angle of the antenna.

[0050] The user terminal of a grid may receive the signals of multiple active cells, that is, a grid corresponds to multiple active cells at the same time. By inputting the characteristics in the grid dimension into the path loss model, the antenna transmit power loss generated by the multiple active cells corresponding to each grid on the signal propagation path can be predicted.

[0051] The signal strength of the active cell is related to the corresponding configured antenna transmit power and the corresponding predicted antenna transmit power loss. Specifically, the signal strength received by the user terminal in the target grid is shown in the following formula:

[0052] where represents the transmit power of the antenna in an active cell corresponding to the target grid, represents the loss of this active cell predicted by the path loss model on the signal propagation path.

[0053] That is, after the target cell turns off the energy-saving policy, the signal strength of the multiple active cells corresponding to the target grid can be predicted. According to the signal strength Select the primary serving cell of the target grid according to the maximum principle, and finally delimit the signal coverage area of each active cell.

[0054] Furthermore, the method further includes: generating a signal strength map of each active cell in the target area based on the signal strength predicted by the path loss model; and displaying the signal distribution of each active cell in the target area based on the signal strength map.

[0055] Generate a signal strength map of the target area using the signal strength value predicted by the path loss model, clearly showing the signal distribution and potential problems of different active cells in the target area.

[0056] Meanwhile, it is also possible to predict the load of the primary serving cell corresponding to the target grid after the target cell executes the energy-saving strategy and shuts down. This load is the sum of the load generated by the original user terminals corresponding to the primary serving cell before the target cell is shut down and the load generated by the migrated user terminals of this serving cell after the target cell is shut down.

[0057] Assume that a certain cell has a corresponding energy-saving strategy. During the energy-saving period, its users will migrate to the adjacent active cell with the strongest signal. The load generated by the original user terminals corresponding to the adjacent active cell can be obtained by collecting performance data. The load generated by the migrated user terminals that migrate to the adjacent active cell can be determined based on the primary serving cell corresponding to each grid. During the energy-saving period, the user terminals of the closed cells corresponding to each grid will migrate to the primary serving cell corresponding to this grid. In this way, the load generated by the migrated user terminals of the primary serving cell can be predicted. Based on the load of the original user terminals of the primary serving cell and the predicted load of the migrated user terminals, the load of the primary server cell after the target cell executes the energy-saving strategy and shuts down can be determined. For the active cells that are not the primary server cell, the corresponding load remains unchanged before and after the target cell executes the energy-saving strategy and shuts down.

[0058] Step 106, determine the optimization strategy for each active cell based on the predicted signal strength and load, and the optimization strategy includes adjusting the antenna parameters of the corresponding active cell.

[0059] After predicting the signal strength and load of each active cell in the target area after the target cell executes the energy-saving strategy and shuts down, an optimization strategy can be formulated for each active cell based on the predicted signal strength and load. The purpose of the optimization strategy is to ensure the quality of the signals in the network area covered by each active cell, and prevent other active cells, especially the primary serving cell where user terminal migration occurs, from having problems such as weak network coverage and / or unbalanced load due to the execution of the energy-saving strategy.

[0060] Based on the solution provided in the above embodiments, optionally, in step 106 above, the determining an optimization strategy for each active cell based on the predicted signal strength and load includes: determining grids with weak network coverage in each active cell based on the predicted signal strength of multiple grids covered by each active cell and a preset signal strength threshold; determining the main serving cells with excessive load in each main serving cell based on the predicted load of the main serving cells of the multiple grids and a preset load threshold; and for each grid with weak network coverage or an excessive load on the corresponding main serving cell, determining an optimization strategy for adjusting the antenna parameters of the corresponding main serving cell.

[0061] As described above, the path loss model can predict the signal strength of each active cell. In the dimension of grid division, the signal strength of the active cells corresponding to each grid can be obtained. Comparing the predicted signal strength with the preset signal strength threshold, if the predicted signal strength is lower than the preset signal strength threshold, it indicates that there is weak network coverage in the corresponding active cell, and the signal quality of the active cell is poor. Correspondingly, the grids where the active cells with weak network coverage are located can be determined. One grid may have one or more active cells with weak network coverage.

[0062] In addition, comparing the predicted load of the main serving cells of each grid with the preset load threshold, if the predicted load sum is lower than the preset load threshold, it indicates that after the target cell executes the energy-saving strategy and shuts down, the migrated user terminals cause the load of the main serving cell to be excessive. For example, after the implementation of the energy-saving strategy, the load of the main serving cell causes the utilization rate of the downlink physical resource block (PRB) of the main serving cell to be > 50%.

[0063] Weak network coverage or an excessive load on the corresponding main serving cell indicates that the corresponding network signal quality is low. Therefore, after the energy-saving strategy is executed, the signal quality of the active cell will be affected and needs to be optimized. In the embodiments of the present application, taking the grid as the object, an optimization strategy is formulated to adjust the antenna parameters of the corresponding main serving cell.

[0064] Optionally, for each grid with weak network coverage or excessive load on the corresponding primary serving cell, determining an optimization strategy for adjusting the antenna parameters of the corresponding primary serving cell, including: for each grid with weak network coverage or excessive load on the corresponding primary serving cell, determining the importance score of each grid, where the importance score is related to the user terminal density included in the grid and the importance degree of the scenario corresponding to the grid; determining the primary serving cells corresponding to the grids with importance scores higher than a preset score threshold; based on the combination of the transmit power, azimuth angle, and beam width of the antenna included in the antenna parameters, determining multiple optimization strategies for the primary serving cells corresponding to the grids; and selecting a target optimization strategy from the multiple optimization strategies for adjusting the antenna parameters of the corresponding primary serving cell, where the target optimization strategy minimizes the total grid importance score of each primary serving cell after the corresponding antenna parameters are adjusted.

[0065] In this embodiment, for each grid with weak network coverage or excessive load on the corresponding primary serving cell, first, an importance score is performed, and then an antenna parameter adjustment scheme is formulated for the grids with higher importance scores.

[0066] To evaluate the importance of each grid, the following indicators are defined: a. Score of user terminal density : Assume that the user terminal distribution of the grid corresponding to the primary serving cell is balanced. The user terminal density r of each grid = (the original number of user terminals of the primary serving cell of the grid + the number of migrated user terminals of the primary serving cell of the grid) ÷ the total number of grids covered by the primary serving cell. The same primary serving cell may cover multiple different grids. For example, the scores of different user terminal densities are divided according to the following criteria : r ≥ 5: score 3; 3 ≤ r < 5: score = 2; 1 ≤ r < 3: score 1; r < 1: score 0.

[0067] b. Grid label o: o = 0 indicates that the grid has good coverage and the load of the primary serving cell is appropriate, and no antenna parameter adjustment is required for this cell or the surrounding cells; o = 1 indicates that the network coverage of this grid is weak, or the current load of the primary serving cell of this grid is excessive / unbalanced.

[0068] c. Grid importance : High historical complaints, important sensitive scenarios such as VIP areas: score 3; Tidal peak scenarios such as CBD areas and school areas during working hours: Score 2; Tidal valley scenarios such as CBD areas and school areas during off - working hours: Score 1.

[0069] Based on the above - mentioned metrics, comprehensively evaluate the importance scores of each grid The calculation formula is as follows:

[0070] Where γ and δ are weight coefficients and can be adjusted according to the actual situation.

[0071] After calculating the importance scores of each grid, if the importance score is higher than the preset score threshold, it indicates that the grid is a relatively important grid. Then, prioritize optimizing the important grids and adjust the antenna parameters of the main serving cell corresponding to the important grids.

[0072] Thus, high - precision regional (grid) optimization can be achieved to ensure that key areas are preferentially guaranteed during energy - saving periods. By dynamically evaluating the importance of each grid, resources and optimization strategies can be intelligently adjusted.

[0073] When generating the antenna parameter adjustment plan, the selectable adjustable parameters include, for example: a. Antenna transmit power adjustment, aiming to expand the signal coverage range and intensity by increasing the antenna transmit power within the adjustable range. Specifically, for grids with higher importance scores, increase the antenna transmit power of the corresponding main serving cell. And use the path loss model to simulate the signal strength value of the main serving cell after antenna parameter adjustment to ensure that the coverage standard is met after adjustment.

[0074] The selectable adjustable parameters can also include: b. Antenna azimuth angle adjustment, aiming to improve the signal coverage of grids with higher importance scores. Specifically, adjust the antenna azimuth angle according to the comprehensive importance score of the grid so that it points to the area of the grid with a higher importance score. Before the tidal peak period (such as during commuting hours), adjust the azimuth angle in advance to meet the signal requirements of users.

[0075] The selectable adjustable parameters can also include: c. Beam width adjustment, aiming to optimize signal coverage according to the density of user terminals and the importance score of the grid. Specifically, for grids with higher importance scores, moderately expand the beam width to enhance signal coverage, especially during peak user periods. For areas with lower importance scores, narrow the beam width to improve signal strength and coverage quality and avoid resource waste. By providing a flexible parameter adjustment plan, ensure signal coverage and load balancing and optimize the user experience.

[0076] By determining the importance grid that needs to be optimized, it is possible to determine the cells for which the antenna parameters corresponding to the grid can be adjusted. For multiple cells, the embodiments of the present application adopt a multi-scheme evaluation method to generate optimization strategies for different antenna adjustment parameter schemes corresponding to each cell respectively.

[0077] For example, for the transmit power: determine the adjustable range (considering the difference between the maximum adjustable space of the antenna and the current transmit power comprehensively), and select several key points (such as one point every 5 dBm). Azimuth angle: set the adjustable range (for example, -3° to 3°), and it can be selected to adjust once every 1°. Beam width: set the adjustable range (for example, -5° to 5°), and select different beam width adjustment values (such as -1°, 1°, -2°, 2°, etc.).

[0078] Use a combination algorithm to generate all possible combination schemes of transmit power, azimuth angle, and beam width based on the above antenna parameter adjustment methods.

[0079] For example, if there are 5 choices for the transmit power, 6 choices for the azimuth angle, and 10 choices for the beam width, the number of combination schemes that can be finally adjusted for one cell is 5×6×10 = 300. If there are 100 cells in the target area that need to adjust the antenna parameters, then 100×300 = 30000 combination schemes can be obtained. Independently adjust the antenna parameters of the corresponding cells based on each combination scheme respectively, and after the antenna parameters are adjusted, repeat the above steps 102 to step 106, re-predict the signal strength and load of the corresponding active cells, determine the grids with weak network coverage or excessive load of the corresponding main serving cells, and calculate the importance scores of these grids. Sum up the importance scores recalculated for the 100 cells after adjustment based on the combination schemes. The combination scheme corresponding to the 100 active cells with the smallest sum of the importance scores is the final optimization strategy.

[0080] The above simulates the adjustment of the antenna transmit power, azimuth angle, and beam width. The relevant adjustments will bring about changes in the network coverage of the corresponding cells. Re-predict the signal strength value of the grid according to the path loss model. Weak coverage and user terminal migration are closely related to the signal strength. The importance scores of each grid can be recalculated according to the prediction results. Select the combination scheme in the set of combination schemes that minimizes the total sum of the importance score values of the corresponding grids in the target area as the scheme corresponding to the final optimization strategy.

[0081] The minimum sum of the importance score values of the grids indicates that after adjusting the antenna parameters of the main serving cell corresponding to the grid according to the corresponding optimization strategy, the problems of weak network coverage of the grid and excessive load of the main serving cell are optimized and solved. There is no need to adjust the antenna parameters of the main serving cell corresponding to the grid. The network coverage of each grid is strong and the load of the main serving cell is appropriate. After the cells in the target area perform energy saving, the overall network signal quality of other enabled cells reaches the optimal level.

[0082] Step 108, after turning off the target cell according to the energy saving strategy, adjust the antenna parameters of the corresponding enabled cells according to the optimization strategy.

[0083] The energy saving strategy has been determined in advance. After simulating the energy saving strategy through steps 102 to 106 and determining the corresponding optimization plan, the energy saving strategy can be directly issued, then the corresponding cell is turned off according to the energy saving strategy, and then the antenna parameters of other enabled cells are adjusted according to the optimization strategy.

[0084] The following combines Figure 2 , and describes an example of the cell energy saving optimization method of the present application. As Figure 2 shown, it includes the following steps: Step 202, data collection and preprocessing; Step 204, perform feature engineering based on the collected data to obtain multiple features affecting the signal propagation of the cells in the target area; Step 206, input the obtained features into the path loss model; Step 208, obtain the energy saving strategy; Step 210, rasterize the map of the target area; Step 212, the path loss model combines with the energy saving strategy, outputs the signal strength of each grid, and generates a signal strength map; Step 214, based on the signal strength, identify whether there are problems of weak coverage or excessive load of the corresponding main serving cell in each grid. If so, enter step 220, otherwise enter step 216; Step 216, issue the energy saving strategy to turn off the corresponding cell for energy saving; Step 218, obtain the scene information and user terminal density information of each grid; Step 220, determine the importance score of each grid; Step 222, generate an antenna parameter adjustment plan for the corresponding cell based on the importance score; Step 224, based on the antenna parameters of the adjustment plan, update the input features of the path loss model and re-predict the corresponding signal strength value RSRP, signal strength map, and calculate the sum of the importance scores of the grids in the target area as the importance score value of the target area; Step 226: After determining the adjustment plan corresponding to the minimum importance score value of the target area, issue an energy-saving strategy.

[0085] In the embodiment of the present application, by obtaining multiple characteristics affecting the cell signal propagation in the target area; based on the obtained characteristics affecting the cell signal propagation and the path loss model, predicting the signal strength and load of each active cell in the target area after the target cell in the target area executes the energy-saving strategy and shuts down, the path loss model is trained with the characteristics affecting the cell signal propagation as samples and the antenna transmission power loss generated by the cells corresponding to the characteristics affecting the cell signal propagation on the signal propagation path as labels, and the characteristics affecting the cell signal propagation include a combination of multiple items among the distance characteristic between the user terminal and the antenna in the active cell, the angle characteristic between the user terminal and the antenna in the active cell, the environmental type of the active cell, the beam width of the antenna, the frequency of the antenna, and the signal downtilt angle of the antenna; based on the predicted signal strength and load, determining the optimization strategy for each active cell, the optimization strategy including adjusting the antenna parameters of the corresponding active cell; after shutting down the target cell according to the energy-saving strategy, adjusting the antenna parameters of the corresponding active cell according to the optimization strategy, thereby pre-playing the energy-saving strategy in a simulation environment, evaluating possible signal problems (such as weak network coverage and excessive load caused by user migration) after the implementation of the energy-saving strategy based on the predicted signal strength and load, and proposing corresponding optimization plans based on the pre-play of the energy-saving strategy, adjusting the antenna parameters for weak coverage areas, so as to identify signal quality problems in the implementation of the energy-saving strategy in advance and provide solutions, automatically adjusting the antenna parameters in the energy-saving mode, effectively ensuring the signal quality in the target area in the energy-saving mode and the network experience of users, reducing the implementation risk of the energy-saving strategy through simulation and pre-play, ensuring a stable user experience while the network is energy-saving, and ensuring the balance between energy-saving and network performance.

[0086] Optionally, as Figure 3 shown, the embodiment of the present application further provides a cell energy-saving optimization device 1000, including: An acquisition module 1200, configured to acquire multiple characteristics affecting the cell signal propagation in the target area; A prediction module 1400, configured to predict the signal strengths and loads of the enabled cells in the target area after the target cell in the target area turns off its energy-saving strategy, based on the obtained characteristics affecting the propagation of the cell signal and the path loss model. The path loss model is trained with the characteristics affecting the propagation of the cell signal as samples and the antenna transmission power loss generated by the cell corresponding to the characteristics affecting the propagation of the cell signal on the signal propagation path as labels. The characteristics affecting the propagation of the cell signal include a combination of multiple items among the distance characteristic between the user terminal and the antenna in the enabled cell, the angle characteristic between the user terminal and the antenna in the enabled cell, the environmental type of the enabled cell, the beam width of the antenna, the frequency of the antenna, and the signal downtilt angle of the antenna; A determination module 1600, configured to determine the optimization strategy for each enabled cell based on the predicted signal strength and load. The optimization strategy includes adjusting the antenna parameters of the corresponding enabled cell; An adjustment module 1800, configured to adjust the antenna parameters of the corresponding enabled cell according to the optimization strategy after turning off the target cell according to the energy-saving strategy.

[0087] The cell energy-saving optimization device provided in the embodiments of this specification can implement Figures 1 to 2 each process implemented in the method embodiments. To avoid repetition, details are not described here again.

[0088] Optionally, as Figure 4 shown, the embodiments of this application further provide a cell energy-saving optimization device 2000, including a processor 2400 and a memory 2200. A program or instruction that can run on the processor 2400 is stored on the memory 2200. When the program or instruction is executed by the processor 2400, it implements each step of the above-mentioned cell energy-saving optimization method embodiment and can achieve the same technical effect. To avoid repetition, details are not described here again.

[0089] The embodiments of this application further provide a readable storage medium. A program or instruction is stored on the readable storage medium. When the program or instruction is executed by a processor, it implements each process of any of the above-mentioned cell energy-saving optimization method embodiments and can achieve the same technical effect. To avoid repetition, details are not described here again. Among them, the readable storage medium includes computer-readable storage media, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs, etc.

[0090] The embodiment of the present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program. The computer program is operable to cause a computer to execute each process of any of the above-described embodiments of the cell energy-saving optimization method and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0091] It should be noted that in this document, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the presence of another identical element in the process, method, article or device including that element.

[0092] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc) and includes several instructions for causing a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of the present application.

[0093] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them belong to the protection scope of the present application.

Claims

1. A method for optimizing cell energy conservation, characterized in that, Including: Obtaining multiple characteristics that affect the propagation of cell signals in a target area; Based on the obtained characteristics affecting cell signal propagation and a path loss model, predicting the signal strength and load of each active cell in the target area after the target cell in the target area executes an energy-saving policy and shuts down. The path loss model is trained with the characteristics affecting cell signal propagation as samples and the antenna transmission power loss generated by the corresponding cell of the characteristics affecting cell signal propagation on the signal propagation path as labels. The characteristics affecting cell signal propagation include a combination of multiple items such as the distance characteristic between the user terminal and the antenna in the active cell, the angle characteristic between the user terminal and the antenna in the active cell, the environmental type of the active cell, the beam width of the antenna, the frequency of the antenna, and the signal downtilt angle of the antenna; Based on the predicted signal strength and load, determining an optimization strategy for each active cell, where the optimization strategy includes adjusting the antenna parameters of the corresponding active cell; After shutting down the target cell according to the energy-saving policy, adjusting the antenna parameters of the corresponding active cell according to the optimization strategy.

2. The method according to claim 1, characterized in that The predicting, based on the obtained characteristics affecting cell signal propagation and the path loss model, the signal strength and load of each active cell in the target area after the target cell in the target area executes an energy-saving policy and shuts down, includes: Dividing the target area into multiple grids of preset units; Based on the obtained characteristics affecting cell signal propagation and the path loss model, respectively predicting the signal strength of multiple grids covered by each active cell and the load of the primary serving cell corresponding to each grid after the target cell executes the energy-saving policy and shuts down. The primary serving cell is the active cell with the maximum signal strength among the multiple active cells corresponding to a grid.

3. The method according to claim 2, wherein The predicting, based on the obtained characteristics affecting cell signal propagation and the path loss model, respectively the signal strength of multiple grids covered by each active cell and the load of the primary serving cell corresponding to each grid after the target cell executes the energy-saving policy and shuts down, includes: Obtaining the characteristics that affect the propagation of signals of multiple active cells corresponding to a target grid; Based on the obtained characteristics and the path loss model, respectively predicting the antenna transmission power loss generated by the multiple active cells corresponding to the target grid on the signal propagation path; Based on the antenna transmission power configured for the multiple active cells and the corresponding predicted antenna transmission power loss, determining the signal strength of the multiple active cells corresponding to the target grid after the target cell executes the energy-saving policy and shuts down; Determining the active cell with the maximum signal strength corresponding to the target grid as the primary serving cell of the target grid; Respectively predicting the original user terminals of the primary serving cell before the target cell is shut down and the migrated user terminals of the primary serving cell after the target cell is shut down to determine the load of the primary serving cell after the target cell executes the energy-saving policy and shuts down.

4. The method according to any one of claims 1 to 3, characterized in that Further including: Generating a signal strength map of each active cell in the target area based on the signal strength predicted by the path loss model; Based on the signal strength map, displaying the signal distribution of each active cell in the target area.

5. The method according to claim 3, wherein Determining an optimization strategy for each active cell based on predicted signal strength and load, including: Determining grids with weak network coverage in each active cell based on the predicted signal strength of multiple grids covered by each active cell and a preset signal strength threshold; Determining main serving cells with excessive load in each main serving cell based on the predicted load of the main serving cells of the multiple grids and a preset load threshold; For each grid with weak network coverage or an overloaded corresponding main serving cell, determining an optimization strategy for adjusting the antenna parameters of the corresponding main serving cell.

6. The method according to claim 5, characterized in that, The determining, for each grid with weak network coverage or an overloaded corresponding main serving cell, an optimization strategy for adjusting the antenna parameters of the corresponding main serving cell, includes: For each grid with weak network coverage or an overloaded corresponding main serving cell, determining an importance score for each grid, where the importance score is related to the user terminal density included in the grid and the importance degree of the scenario corresponding to the grid; Determining the main serving cells corresponding to the grids with importance scores higher than a preset score threshold; Based on combinations of the transmit power, azimuth angle, and beam width of the antennas included in the antenna parameters, determining multiple optimization strategies for the main serving cells corresponding to each grid; Selecting a target optimization strategy from the multiple optimization strategies for adjusting the antenna parameters of the corresponding main serving cell, where the target optimization strategy minimizes the total importance score of the grids of each main serving cell after the corresponding antenna parameters are adjusted.

7. An energy-saving optimization device for a community, characterized in that, Including: An acquisition module, configured to acquire multiple characteristics affecting the cell signal propagation in a target area; A prediction module, configured to predict the signal strength and load of each active cell in the target area after an energy-saving strategy is executed to turn off a target cell in the target area, based on the acquired characteristics affecting cell signal propagation and a path loss model, where the path loss model is trained with the characteristics affecting cell signal propagation as samples and the antenna transmit power loss generated by the cells corresponding to the characteristics affecting cell signal propagation on the signal propagation path as labels, and the characteristics affecting cell signal propagation include a combination of multiple items such as the distance characteristic between a user terminal and an antenna in an active cell, the angle characteristic between the user terminal and the antenna in the active cell, the environmental type of the active cell, the beam width of the antenna, the frequency of the antenna, and the signal downtilt angle of the antenna; A determination module, configured to determine an optimization strategy for each active cell based on the predicted signal strength and load, where the optimization strategy includes adjusting the antenna parameters of the corresponding active cell; An adjustment module, configured to adjust the antenna parameters of the corresponding active cell according to the optimization strategy after turning off the target cell according to the energy-saving strategy.

8. A device for optimizing energy conservation in a community, characterized in that, Including a processor and a memory, where the memory stores a program or instruction that can run on the processor, and when the program or instruction is executed by the processor, the steps of the method according to any one of claims 1-6 are implemented.

9. A readable storage medium, characterized in that, A program or instruction is stored on the readable storage medium, and when the program or instruction is executed by a processor, the steps of the method according to any one of claims 1-6 are implemented.

10. A computer program product, characterized in that, The computer program product includes a non-transitory computer-readable storage medium storing a computer program, the computer program being operable to cause a computer to perform the steps of the method according to any one of claims 1-6.

Images