A method for determining the weather boundary based on 5G service migration characteristics

Abstract

This invention discloses a method for determining the weather boundary based on 5G service migration characteristics, comprising: acquiring historical rainfall data covering 5G cells, extracting service migration characteristics and corresponding weather conditions from the data for regression analysis, and establishing a weather boundary determination model based on 5G service migration characteristics; monitoring and extracting service migration characteristics one from real-time service data of 5G cells, and inputting it into the weather boundary determination model to obtain weather condition one; extracting cells with weather boundaries in weather condition one, and determining the weather boundary between the cells based on macro base station information within the cells. The weather boundary determination method disclosed in this invention can achieve more refined weather monitoring. By incorporating dense 5G data, it can effectively improve the resolution of weather boundaries and determine them more accurately; simultaneously, based on existing 5G equipment, it does not incur additional costs.

Description

Technical Field

[0001] This invention relates to the field of weather boundary determination technology, and more specifically to a weather boundary determination method based on 5G service migration characteristics. Background Technology

[0002] Current rain and sunshine determination technologies primarily rely on rainfall monitoring equipment such as rain gauges, radar, and meteorological satellites. These technologies suffer from insufficient spatiotemporal resolution, weak spatial representativeness, and monitoring blind spots, leading to inaccurate boundary determination. Further enhancing monitoring capabilities requires additional equipment and investment. The specific reasons are as follows:

[0003] Rain gauges are mainly used to measure the total precipitation in a certain area over a period of time, with a measurement accuracy of up to 0.1 mm. However, rain gauge measurements usually only represent the precipitation characteristics of a specific area, making it difficult to accurately observe the distribution of precipitation over a large area.

[0004] Precipitation radar indirectly calculates precipitation location and intensity by measuring echo intensity, enabling observation of precipitation distribution over a relatively large area. However, the accuracy of radar-derived rainfall intensity is often affected by uncertainties in the Zr relationship and differences in precipitation type. Moreover, complex terrain and other clutter signals can also introduce significant errors into the measurement results.

[0005] For meteorological satellites, precipitation is indirectly estimated by detecting cloud structure and cloud top brightness temperature using electromagnetic waves in the visible and infrared bands. However, because these two bands of electromagnetic waves have poor penetration, they cannot directly obtain information on precipitation within clouds and on the ground, resulting in poor measurement accuracy. While spaceborne radar can acquire precipitation intensity and distribution characteristics within clouds, its spatiotemporal resolution is low, making it difficult to capture complete small- to medium-scale convective cells. Furthermore, because spaceborne radar emits a wide beam, blind spots occur near the ground, and the signal is easily interfered with by non-precipitation information during transmission, leading to significant measurement errors near the ground.

[0006] Furthermore, current methods for modeling rainy / sunny boundaries primarily rely on rain gauge and radar measurement data. The accuracy of these methods depends on rain gauge density and radar resolution. Rain gauges operate on a point-by-point basis, resulting in poor spatial representativeness. Improving the preparedness of rainy / sunny boundaries solely by providing rain gauge density is both uneconomical and inaccurate, as it extrapolates from point data to surface data. Moreover, radar spatial resolution cannot currently be further improved, and it cannot monitor actual near-surface rainfall data.

[0007] Therefore, in view of the above-mentioned defects in the existing technology, how to provide a method for determining the weather boundary without investing in new equipment and improving the accuracy of boundary determination is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0008] In view of this, based on the rapidly developing 5G services, the present invention provides a method for determining the sunny and rainy boundary based on the migration characteristics of 5G services. The method aims to analyze the changes in the migration characteristics of 5G services on rainy and sunny days in historical rainfall data covering 5G cells, and determine the sunny and rainy boundary based on the migration characteristics of 5G services, thereby facilitating accurate determination of the sunny and rainy boundary and achieving more refined sunny and rainy monitoring.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] A method for determining the weather boundary based on 5G service migration characteristics, comprising:

[0011] Historical rainfall data covering 5G cells is obtained, service migration features and corresponding weather conditions are extracted from the historical rainfall data, regression analysis is performed on the service migration features and the weather conditions, and a weather boundary determination model based on 5G service migration features is established.

[0012] Monitor and extract service migration feature 1 from the real-time service data of 5G cells, and input it into the weather boundary determination model to obtain weather condition 1.

[0013] Extract the cells with clear / rainy boundaries in the first weather condition, and determine the clear / rainy boundaries between the cells based on the macro base station information within the cells.

[0014] Preferably, the service migration features include the number of reselections and the number of switching operations.

[0015] Preferably, the service migration features also include cell type and handover type.

[0016] Preferably, the weather boundary determination model based on 5G service migration characteristics is trained and corrected using a machine learning algorithm before use.

[0017] Preferably, the weather information includes the numbers 1 and 0, where 1 represents rainfall and 0 represents no rainfall.

[0018] Preferably, based on a time series algorithm, the service migration characteristics of the historical service data of the 5G cell are predicted to obtain the predicted service migration characteristic two, and the service migration characteristic two and the service migration characteristic one are jointly input into the weather boundary determination model.

[0019] Preferably, the step of determining the weather boundary between cells based on macrocell information within the cell includes:

[0020] Obtain basic information about macro base stations within the cell;

[0021] Based on the aforementioned basic information and coverage type, calculate the theoretical coverage area of ​​the cell;

[0022] By combining MR, KPI and other data, the theoretical coverage range is corrected to obtain the actual coverage range;

[0023] Based on the actual coverage area, an overlapping coverage area is obtained, and a handover zone is obtained based on the overlapping coverage area. The handover zone is the boundary between sunny and rainy conditions between the cells.

[0024] Preferably, the boundary between sunny and rainy areas between the cells is corrected using MR big data, and the steps include:

[0025] Obtain the MR data of the cell;

[0026] Based on the user-level data in the MR data, the user-level movement trajectory is confirmed;

[0027] Cluster analysis is performed on the user-level mobile trajectories to obtain the weather boundaries after clustering;

[0028] The isosurface algorithm is used to correct the isosurface boundaries between the cells using the clustered isosurface boundaries, resulting in corrected isosurface boundaries.

[0029] Preferably, the clear / rainy boundary between the communities is combined with the clear / rainy boundary obtained from multi-source meteorological data to determine the clear / rainy boundary.

[0030] Preferably, the corrected clear / rainy boundary is combined with the clear / rainy boundary obtained from multi-source meteorological data to determine the clear / rainy boundary.

[0031] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a method for determining the weather boundary based on 5G service migration characteristics. By using the weather boundary determination model based on 5G service migration characteristics to obtain the weather conditions of the cell according to the real-time service data of the 5G cell, the weather boundary is determined based on the 5G macro base station and the 5G cell handover zone.

[0032] The method for determining the boundary between sunny and rainy conditions disclosed in this invention enables more refined monitoring of sunny and rainy conditions. By incorporating dense 5G data, it effectively improves the resolution of the boundary between sunny and rainy conditions and can determine the boundary more accurately. Furthermore, it utilizes existing 5G equipment without incurring additional costs.

[0033] Another beneficial effect of this invention includes that, based on TSF-related migration feature data and combined with MR data to assist in determining the clear / rainy boundary, the spatial resolution and accuracy of the clear / rainy boundary determination can be further improved. Simultaneously, combining the clear / rainy boundary obtained by this method with existing multi-source meteorological fusion data further improves the accuracy of existing clear / rainy boundary models.

[0034] In addition, the method for determining the boundary between sunny and rainy weather disclosed in this invention can be applied to various extreme weather conditions, and can accurately determine the boundary between extreme weather conditions when business migration changes due to weather conditions. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0036] Figure 1 This is a flowchart of the weather boundary determination method based on 5G service migration characteristics of the present invention;

[0037] Figure 2 This is a schematic diagram of the inter-cell switching zone of the present invention;

[0038] Figure 3 This invention provides a general process for determining the boundary between sunny and rainy weather. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] like Figure 1 As shown in the figure, this invention discloses a method for determining the weather boundary based on 5G service migration characteristics. The specific steps are as follows:

[0041] Historical rainfall data covering 5G cells is obtained, service migration characteristics and corresponding weather conditions are extracted from the historical rainfall data, regression analysis is performed on the service migration characteristics and weather conditions, and a weather boundary determination model based on 5G service migration characteristics is established.

[0042] Monitor and extract service migration feature 1 from real-time service data of 5G cells, input it into the weather boundary determination model, and obtain weather condition 1.

[0043] Extract cells with clear / clear boundaries in the first scenario of clear / clear weather conditions, and determine the clear / clear boundaries between cells based on macro base station information within the cells.

[0044] To make the above-described solution of the present invention more apparent and understandable, each step will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] Due to varying rainfall intensities, 5G service migration characteristics differ; specifically, light rain, moderate rain, heavy rain, torrential rain, and extremely heavy rain all produce different 5G service migration characteristics. This invention, based on existing 5G services, considers and analyzes the relationship between 5G service migration characteristics and weather conditions, thereby establishing a weather boundary determination model based on 5G service migration characteristics.

[0046] Specifically, the process involves first acquiring historical rainfall data covering 5G cells, then extracting 5G service migration characteristics and corresponding weather conditions from the historical data, and finally performing regression analysis to establish the required weather boundary determination model.

[0047] To obtain historical rainfall data covering 5G cells, it is necessary to first obtain 5G base station engineering parameter data to confirm the coverage area of ​​the cells. In this invention, the principle for determining the coverage area of ​​macro cells is as follows: the effective determination area of ​​outdoor cells uses area rainfall data, and the indoor distributed cell uses point rainfall data; then, the obtained rainfall data is organized according to time.

[0048] In one embodiment, the main analytical indicators of service migration features extracted from rainfall data include: cell reselection counts (when the UE is in idle state) and cell handover counts (when the UE is in service state) based on high temporal resolution. That is, regression analysis is performed based on the reselection count, handover count, and corresponding weather conditions. This is because reselection and handover are direct indicators of user movement; the handover and reselection data directly reflect user movement. In other words, these two indicators are most closely related to weather conditions. Furthermore, since the number of dependent variables used in this application is greater than one, multiple regression analysis is performed.

[0049] In another embodiment, cell type and handover type can also be extracted from rainfall data as analytical indicators of 5G service migration characteristics. Cell type includes outdoor macro cells and indoor distributed cells; outdoor macro cells have a large coverage area, while indoor distributed cells have a small coverage area (such as covering a building). Therefore, different cell types have different handover characteristics and different service model changes. Handover type includes handover in and handover out, used to indicate the flow of handover between cells.

[0050] To improve the accuracy of the built model, in one embodiment, a large amount of data consistent with the business migration indicators used when the model was built is acquired to form a training sample set. This set is used to train, correct, and iterate the model using machine learning algorithms before it is used. The trained weather boundary determination model can make boundary determinations more stably and accurately.

[0051] Since the model in this application uses the weather conditions corresponding to the business migration characteristics as independent variables when it is built, the output of the weather boundary determination model constructed by this invention is the weather conditions that match the input data.

[0052] In one embodiment, the weather conditions are set in the form of Rainfall Markers (RM). The specific Rainfall Markers include the numbers 1 and 0, which represent the weather conditions of two cells respectively. Here, 1 indicates a cell with rainfall and 0 indicates a cell without rainfall.

[0053] Taking cells A and B within the 5G coverage area as an example, if both cells A and B experience rainfall, the model outputs 11; if both cells A and B experience sunny weather, the model outputs 00; if one cell A experiences rainfall and the other experiences sunny weather, the model outputs 10 or 01.

[0054] If the model outputs 10 or 01, it means that the rainfall determination is based on the characteristics of 5G service migration.

[0055] Specifically, when using the weather boundary determination model disclosed in this invention, it is necessary to monitor the real-time service data of 5G cells and further extract service migration features to input into the weather boundary determination model to obtain the weather conditions of 5G cells.

[0056] In one embodiment, a sliding time window is used to perform statistical analysis on real-time monitored 5G service data, wherein the time granularity can be set to 5 minutes, 10 minutes, 30 minutes or 1 hour.

[0057] Because 5G service characteristics have certain regularities and randomness, in order to ensure the relative accuracy of the model results, this application adopts a time series algorithm to predict the service migration characteristics of historical service data of 5G cells based on time series data. The prediction results are then combined with the service migration characteristics extracted from real-time 5G service data and input into the weather boundary determination model to obtain the determination result.

[0058] As mentioned above, when the model's judgment result is 10 or 01, it means that one of the two communities is experiencing rain and the other is sunny, indicating that there is a boundary between the two communities.

[0059] Furthermore, this application uses the 5G macro base station cell-to-macro base station handover zone as the clear / rainy boundary for judgment, which features high resolution and a small total area. It primarily analyzes the effective judgment surface of the corresponding data based on the engineering parameter data, geographic information data, and base station parameters of the macro base station. For example... Figure 2 As shown, indoor distributed cell networks (DCCs) are not considered in this invention because they primarily cover indoor areas and are mainly point-based. Specifically, it is necessary to first obtain macrocell information within the cells involved in the boundary. The specific steps include:

[0060] Obtain basic information about macro base stations within the community, including latitude and longitude, transmit power, antenna elevation angle, and other data.

[0061] Based on the basic information and coverage type, calculate the theoretical coverage area of ​​the cell; where the coverage type is sector or other types of area data;

[0062] Here, the coverage area is primarily evaluated using a propagation model through simulation, including:

[0063] 1. Basic input information includes operating frequency, antenna height, surrounding geographical environment, transmission power, antenna parameters, and operating parameters of surrounding cells;

[0064] 2. Perform cell link budgeting based on the Hata-Okumuram model;

[0065] 3. Input the base station information and engineering parameters of surrounding cells; combine the link budget and coverage radius of this cell; perform coverage simulation of this cell based on high-precision map and 3D propagation model to obtain the theoretical coverage range.

[0066] In one embodiment, the theoretical coverage area is corrected by combining data such as MR (Local Range) and KPI (Key Performance Indicator) to obtain the actual coverage area. Here, MR refers to user-level latitude and longitude data, and KPI data refers to cell reselection and handover data with surrounding cells. The actual coverage area of ​​the cell is evaluated based on user-level latitude and longitude data, reselection and handover data.

[0067] Then, based on the actual coverage area, the overlapping coverage area is calculated, and the handover zone is further calculated; specifically, based on the cell handover data (including handover and handover), the latitude and longitude data of the handover service are processed according to the contour line method to obtain the handover zone.

[0068] Finally, the resulting switching zone represents the boundary between sunny and rainy areas between cells.

[0069] In one embodiment, to achieve higher resolution and accuracy in the switching band, the obtained switching band is further corrected based on MR data.

[0070] In another embodiment, to further improve the spatial resolution and accuracy of boundary determination, this application proposes to analyze user-level MR data based on the weather boundary determination model, and to cluster data according to user-level service migration, thereby further strengthening and refining the identification of 5G service migration. Specifically, this application corrects the weather boundaries between cells using MR big data. Since users in outdoor environments within the weather boundary area tend to move towards areas without rain or indoors, the correction of the weather boundaries only considers user-level data within macrocell coverage.

[0071] The correction methods include:

[0072] Obtain the weather conditions determined by the weather boundary determination model, extract the cells with weather boundaries, and obtain the MR data of the corresponding cells;

[0073] Based on user-level data in the MR data, including latitude and longitude information, confirm the user-level movement trajectory;

[0074] Then, cluster analysis is performed on user-level movement trajectories to refine the identification of user movement on rainy days, so as to obtain the clear and rainy boundaries after clustering.

[0075] Finally, the isosurface algorithm is used to correct the weather boundaries between cells using the clustered weather boundaries, resulting in the corrected weather boundaries.

[0076] The clear / rainy boundary obtained by this invention can be combined with the clear / rainy boundary obtained from multi-source meteorological data to jointly determine the clear / rainy boundary, thereby forming a more refined and accurate clear / rainy boundary.

[0077] In one embodiment, the obtained clear / rainy boundary between cells is combined with the clear / rainy boundary obtained from multi-source meteorological data; in another embodiment, the corrected clear / rainy boundary can also be combined with the clear / rainy boundary obtained from multi-source meteorological data. In this case, the overall determination process is as follows: Figure 3 ,include:

[0078] The process of determining weather conditions first requires building a weather determination model based on the characteristics of 5G service migration, and training it using historical 5G big data and rainfall data. Secondly, weather conditions are determined based on real-time monitoring and predicted service migration data to confirm the weather boundaries. Finally, the model is revised using user-level MR data to obtain more accurate weather boundaries.

[0079] The specific steps are as follows:

[0080] S1: Collect 5G cell service data in real time and perform sliding statistics based on time granularities of 5 minutes, 10 minutes, 30 minutes, and 1 hour;

[0081] S2: Combining the time series TSF algorithm, predict the TSF migration data (reselection, handover) of each cell at the current time granularity based on 5G cell service data;

[0082] S3: Input the real-time sliding business migration data and TSF migration data into the weather-assisted judgment model for calculation;

[0083] S4: If the model's determination result RM is 10 or 01, that is, between the two cells, one cell has rain and the other has sunny weather, there is a rain-sunshine boundary.

[0084] S5: Define the handover zone of cell 01 or 10 as the initial clear / rainy boundary P.

[0085] S6: Analyze user-level data for cells with RM of 10 or 01, analyze the movement of users under macrocell coverage, cluster them according to their latitude and longitude data, and form user cluster analysis weather boundary S;

[0086] S7: Based on the isosurface algorithm, the sunny and rainy boundary S is corrected by user clustering analysis to obtain the corrected sunny and rainy boundary F.

[0087] S8: The clear / rainy boundary of multi-source meteorological data is further fused and calculated by combining the clear / rainy boundary F based on the migration characteristics of 5G services, to realize the clear / rainy boundary based on multi-source meteorological data and 5G data.

[0088] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0089] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1.A sunny-rain boundary determination method based on a 5G service migration feature, characterized by, The method comprises the following steps: obtaining historical rainfall data covering 5G cells, extracting service migration features and corresponding sunny and rainy conditions from the historical rainfall data, performing regression analysis on the service migration features and the sunny and rainy conditions, and establishing a sunny and rainy boundary determination model based on 5G service migration features; the service migration features include the number of reselections, the number of handovers, cell types, and handover types; monitoring and extracting service migration features from real-time service data of 5G cells, inputting the service migration features into the sunny and rainy boundary determination model, and obtaining sunny and rainy conditions; extracting cells with sunny and rainy boundaries in the sunny and rainy conditions, determining the sunny and rainy boundaries between the cells according to macro station information in the cells; the method comprises the following steps: obtaining basic information of the macro stations in the cells; calculating the coverage range of the cells according to the basic information and coverage types; obtaining overlapping coverage areas according to the coverage range, and obtaining handover bands according to the overlapping coverage areas, wherein the handover bands are the sunny and rainy boundaries between the cells; correcting the sunny and rainy boundaries between the cells by MR big data, wherein the MR big data is user-level latitude and longitude data, and the steps comprise: obtaining MR data of the cells; confirming user-level movement tracks according to user-level data in the MR data; performing cluster analysis on the user-level movement tracks to obtain clustered sunny and rainy boundaries; correcting the sunny and rainy boundaries between the cells by using the clustered sunny and rainy boundaries through an isosurface algorithm to obtain corrected sunny and rainy boundaries. 2.The sunny and rainy boundary determination method based on 5G service migration characteristics according to claim 1, characterized in that, The sunny and rainy boundary determination model based on 5G service migration features is trained and corrected by using a machine learning algorithm before use. 3.The sunny and rainy boundary determination method based on 5G service migration characteristics according to claim 1, characterized in that, The sunny and rainy conditions include a number 1 and a number 0, wherein the number 1 represents rainfall, and the number 0 represents no rainfall. 4.The sunny and rainy boundary determination method based on 5G service migration characteristics according to claim 1, characterized in that, Based on a time series algorithm, the service migration features of the historical service data of the 5G cells are predicted to obtain predicted service migration features two, and the service migration features two and the service migration features one are input into the sunny and rainy boundary determination model. 5.The sunny and rainy boundary determination method based on 5G service migration characteristics according to claim 1, characterized in that, The sunny and rainy boundaries between the cells are combined with sunny and rainy boundaries obtained according to multi-source meteorological data to determine the sunny and rainy boundaries. 6.The sunny and rainy boundary determination method based on 5G service migration characteristics according to claim 1, characterized in that, The corrected sunny and rainy boundaries are combined with sunny and rainy boundaries obtained according to multi-source meteorological data to determine the sunny and rainy boundaries.

Images