Mobile communication network upgrade evolution method and device of base station, equipment and medium

Abstract

The invention provides a mobile communication network upgrade evolution method and device of a base station, equipment and a medium, and relates to the technical field of communication. The method comprises the following steps: acquiring a space-time diagram of a 4G base station, wherein the space-time diagram is used for describing a correlation dependency relationship between the 4G base station and other 4G base stations or 5G base stations except the 4G base station in space; obtaining service feature data of each 4G base station included in the space-time diagram; predicting service feature data of the 4G base station in a future target time period based on a space-time diagram neural network configured for the 4G base station, the space-time diagram and the service feature data of the 4G base station; determining a to-be-evolved base station in all 4G base stations included in the space-time diagram based on service feature data of each 4G base station in a future target time period; and carrying out mobile communication network upgrade evolution on the base station to be evolved. According to the method and the device, which existing 4G base stations can be directly upgraded and evolved into 5G base stations can be accurately determined, and the accuracy of upgrading and evolving the base stations is improved.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to a method, device, equipment and medium for upgrading and evolving a mobile communication network of a base station. Background Art

[0002] With the advancement of mobile communication technology and the continuous upgrading of user terminals, mobile communication networks have gradually evolved from 4G to 5G. The load of 4G base stations has continued to decrease, while the load of 5G base stations has continued to increase. In actual scenarios, among all 4G base stations, some 4G base stations can be directly upgraded to 5G base stations, thereby reducing the load of 5G base stations, but it is necessary to accurately determine which 4G base stations are upgradeable.

[0003] At present, the main way to determine whether a 4G base station is upgradeable is to conduct business analysis on the 4G base station alone and determine whether it is upgradeable based on the mobile communication services it carries.

[0004] However, this method cannot currently accurately determine whether the 4G base station is upgradeable, and there is a problem of low determination accuracy. Summary of the invention

[0005] The embodiments of the present invention provide a method, device, equipment and medium for upgrading and evolving a mobile communication network of a base station, which can more accurately grasp which 4G base stations among the existing 4G base stations can be upgraded and evolved, improve accuracy, and thus upgrade these 4G base stations to 5G base stations in a targeted manner.

[0006] In a first aspect, an embodiment of the present invention provides a method for upgrading and evolving a mobile communication network of a base station, the method comprising:

[0007] Obtain a space-time diagram of the 4G base station, where the space-time diagram is used to describe a spatial association dependency relationship between the 4G base station and other 4G base stations or 5G base stations other than the 4G base station;

[0008] Obtaining service characteristic data of each 4G base station included in the space-time graph;

[0009] Based on the spatiotemporal graph neural network configured for the 4G base station, the spatiotemporal graph and the service characteristic data of the 4G base station, predicting the service characteristic data of the 4G base station in a future target time period;

[0010] Based on the service characteristic data of each 4G base station in the future target time period, determining the base station to be evolved from all 4G base stations included in the space-time graph;

[0011] Perform mobile communication network upgrade evolution on the base station to be evolved.

[0012] In a second aspect, an embodiment of the present invention provides a mobile communication network upgrade evolution device of a base station, the device comprising:

[0013] A space-time graph acquisition module, used to acquire a space-time graph of a 4G base station, wherein the space-time graph is used to describe a spatial correlation dependency relationship between the 4G base station and other 4G base stations or 5G base stations other than the 4G base station;

[0014] A data acquisition module, used to acquire service characteristic data of each 4G base station included in the space-time diagram;

[0015] A feature prediction module, configured to predict the service feature data of the 4G base station within a future target time period based on the spatiotemporal graph neural network configured for the 4G base station, the spatiotemporal graph and the service feature data of the 4G base station;

[0016] A base station determination module, configured to determine a base station to be evolved from among all 4G base stations included in the space-time graph based on service characteristic data of each 4G base station in a future target time period;

[0017] The upgrade and evolution module is used to upgrade and evolve the mobile communication network of the base station to be evolved.

[0018] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a processor, and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the method as described above.

[0019] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which computer program instructions are stored, and the computer program instructions implement the above-mentioned method when executed by a processor.

[0020] The mobile communication network upgrade and evolution method, device, equipment and medium of the base station provided in the embodiments of the present invention use a space-time graph to describe the spatial position association relationship between base stations, and use the space-time graph as the input of the space-time graph neural network to predict the service feature data of the 4G base station in the future target time period. Based on this, the changes in the time and space of the 4G base station's service can be analyzed, so as to more accurately determine whether the 4G base station can be upgraded to a 5G base station, thereby improving the accuracy of the upgrade and evolution of the mobile communication network. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the following is a brief introduction to the drawings required for use in the embodiment of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 A flowchart of a method for upgrading and evolving a mobile communication network provided in an embodiment of the present application;

[0023] Figure 2 A schematic diagram of the spatiotemporal graph neural network model architecture provided in an embodiment of the present application;

[0024] Figure 3 A 4G base station upgrade evolution determination flow chart provided in an embodiment of the present application.

[0025] Figure 4 A schematic diagram of a 4G base station upgrade evolution sequence determination process provided in an embodiment of the present application;

[0026] Figure 5 A flow chart for constructing a time-space graph provided in an embodiment of the present application;

[0027] Figure 6 An overall flow chart of the mobile communication network upgrade evolution method provided in the embodiment of the present application;

[0028] Figure 7 Base station associated data processing flow chart provided for the embodiment of the present application

[0029] Figure 8 A schematic diagram of the structure of a mobile communication network upgrade and evolution device provided in an embodiment of the present application;

[0030] Fig. 9 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The features and exemplary embodiments of various aspects of the present invention will be described in detail below. In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present invention, rather than to limit the present invention. For those skilled in the art, the present invention can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present invention by illustrating examples of the present invention.

[0032] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0033] With the advancement of mobile communication technology and the continuous upgrading of user terminals, mobile communication networks have gradually evolved from 4G to 5G, which has led to a continuous decrease in the load of 4G base station equipment and a continuous increase in the load of 5G base station equipment. In actual network operations, there are a large number of 4G base station equipment that can be directly upgraded to 5G base station equipment. This direct evolution and upgrade method can greatly improve the deployment efficiency of 5G networks and reduce construction costs. It is an optimal evolution and upgrade method. However, this method will cause major changes in the network structure, requiring a detailed assessment of the network's business conditions and determining the evolution and upgrade order of each base station. Therefore, there is a great deal of difficulty in evaluation and analysis.

[0034] In practical applications, the business conditions of mobile communications not only change with time, but are also closely related to spatial locations. Therefore, the analysis of business conditions requires in-depth mining of spatiotemporal correlation data. However, existing methods cannot grasp both time and space at the same time. For example, the time series prediction of services based on the time dimension of a cell or base station can accurately obtain the future business volume and business trends of the cell, but ignores the spatial correlation relationship; another example is the business analysis method based on the scenario dimension. Although it reflects the business changes in space, it cannot be better combined with time. Specifically speaking, the existing methods have the following shortcomings: ① The existing methods evaluate 4G base stations and 5G base stations separately, without integrating the inherent correlation between the two, and there is a one-sided analysis; ② The existing methods analyze the services separately for time or space, and do not realize spatiotemporal correlation analysis, which affects the accuracy of the results; ③ The existing methods cannot extract the quantitative relationship of spatiotemporal dependence, so it is difficult to provide spatiotemporal quantitative support for the evolution and upgrade of 4G base stations and 5G base stations.

[0035] In view of the above situation, it is necessary to find a method that can efficiently mine spatiotemporal correlation data to drive the evolution and upgrade of current 4G base stations to 5G base stations. To this end, the inventors of this application have creatively invented a mobile communication network upgrade and evolution plan through research. First, a method for evaluating the correlation between 4G base stations and 5G base stations for network evolution and upgrade is developed to realize the integrated evaluation of 4G base stations and 5G base stations; secondly, a method for mining spatiotemporal correlation business data of 4G base stations and 5G base stations using a spatiotemporal graph neural network STGNN is invented to capture the geographical mobility patterns of site services over time and realize spatiotemporal correlation analysis; finally, a method for judging the evolution and upgrade of 4G base stations and 5G base stations using the spatiotemporal dependency data output by STGNN is invented to improve the overall analysis accuracy.

[0036] The technical solution of the present application is described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0037] Figure 1 A flowchart of a mobile communication network upgrade and evolution method provided in an embodiment of the present application.

[0038] like Figure 1 As shown, the method may include the following steps:

[0039] Step S11: Obtain a spatiotemporal diagram of a 4G base station.

[0040] Among them, the space-time diagram is used to describe the spatial correlation and dependency between a 4G base station and other 4G base stations or 5G base stations other than the 4G base station.

[0041] The 4G base station in this embodiment may refer to all existing 4G base stations in a fixed area in an actual scenario. In addition, illustratively, the 4G base station in this embodiment may also refer to those 4G base stations that can support upgrades in hardware. For example, in a fixed area, there are base station A, base station B, and base station C. According to the hardware support, if base station A is found to be unable to support the upgrade, then base station A is screened out. If the remaining base stations B and C have hardware that supports the upgrade, their time-space diagrams can be obtained, and then subsequent steps S12 to S15 can be executed. In this way, a batch of 4G base stations can be preliminarily screened out through the hardware support, reducing the total sample size, thereby improving the upgrade and evolution efficiency of the mobile communication network.

[0042] In this embodiment, after the 4G base station is determined, a base station that is associated with the location of the 4G base station can be searched to construct a space-time graph.

[0043] Exemplarily, the neighboring base station of the 4G base station can be regarded as a base station associated with the 4G base station. The neighboring base station may be of the same standard as the 4G base station, or may be different. For example, the neighboring base station may be a 4G base station or a 5G base station.

[0044] In this embodiment, the construction of the space-time graph is to obtain the spatial correlation dependency between 4G base stations, between 5G site base stations, and between 4G base stations and 5G base stations. The correlation dependency will serve as the input of the subsequent space-time graph neural network model.

[0045] For example, the space-time graph G can be represented as G B =(V B , E B , A B ). Where V is the vertex of the graph, which consists of base stations, E represents all edge sets, A represents the adjacency matrix, and B represents the base stations.

[0046] Step S12: Obtain the service characteristic data of each 4G base station included in the space-time diagram.

[0047] In this embodiment, after the space-time graph is constructed, since the vertices of the space-time graph represent base stations, all 4G base stations are found based on the vertex set (these vertices may represent 4G base stations or 5G base stations). Then, other base stations that have an associated dependency relationship with the 4G base station can be found through the space-time graph (these base stations include 4G base stations and / or 5G base stations). 4G base stations are found among these other base stations, and then their service feature data are obtained.

[0048] In this embodiment, the service characteristic data may refer to at least one of the number of cell users, service flow, and wireless utilization rate of the base station at a historical moment. By using the number of cell users, service flow, and wireless utilization rate, and integrating the time-space graph, the service change trend of the base station can be more comprehensively analyzed, thereby more accurately predicting the service characteristic data of the base station at a future moment.

[0049] For example, in some implementations, when acquiring service feature data, historical service data within a certain historical period before the current time may be collected, such as more than 2 years in the past, so that long-term service changes can be grasped. In addition, the time granularity of the service feature data can be set. For example, if the time granularity is 1 hour, the service feature data may specifically include the cell name, base station name, number of users, service flow, and wireless utilization rate within each hour.

[0050] In addition, after collecting the service feature data, the service feature data can be further processed. Specifically, according to the evolution and upgrade evaluation cycle set by the system, the statistical indicators of different cells can be aggregated based on the base station and time period. The service feature data after aggregation processing is more reliable and can improve the accuracy of grasping the service changes of the base station.

[0051] The specific aggregation feature fields include the statistical time interval, base station name, average number of users, average service flow and average wireless utilization. The aggregation method is to first sum and aggregate the data of the cell to the base station, and then average and aggregate according to the time granularity.

[0052] The average aggregation method may include but is not limited to extracting 6 busy hours per day, 15 busy hours per day, the busiest hour per day, etc. If the upgrade evaluation cycle is 1 day, the sampling point features within 24 hours are aggregated to form a statistical sample; if the upgrade evaluation cycle is 1 week, the sampling point features within 24*7 hours are aggregated to form a statistical sample.

[0053] Step S13: Based on the spatiotemporal graph neural network configured for the 4G base station, the spatiotemporal graph and the service feature data of the 4G base station, predict the service feature data of the 4G base station in a future target time period.

[0054] In this embodiment, each 4G base station can be configured with a corresponding spatio-temporal graph neural network (STGNN), which can be used to process spatio-temporal data, such as videos, time series, etc. It combines the ideas of graph neural networks and spatio-temporal modeling, and demonstrates powerful capabilities in processing spatio-temporal data. Usually in spatio-temporal data, there are complex spatio-temporal relationships between data points. Under this spatio-temporal relationship, the spatio-temporal graph neural network better captures the dependencies and evolution laws between data, learns the spatio-temporal graph structure from the data, and performs information transmission and feature extraction on this structure, thereby predicting the service feature data of the 4G base station in the future target time period.

[0055] Exemplarily, the spatio-temporal graph neural network (STGNN) may include a spatial graph neural network layer, a time series model, a spatio-temporal attention mechanism layer, and an output layer in sequence. The spatial graph neural network layer is used to capture the spatial relationship between 4G base stations, 5G base stations, and 4G base stations and 5G base stations through the learning of the spatio-temporal graph; the time series model may be a long short-term memory network (LSTM) and a gated recurrent unit (GRU) layer, which analyzes the evolution of the characteristic variables of nodes and edges in the spatio-temporal graph data at different time points, so that the STGNN can predict the future spatio-temporal state; the spatio-temporal attention mechanism layer (such as a transformer layer) can directly capture the long-range time dependency in the sequence, improve the model's expressiveness and prediction accuracy. The output layer can output the prediction results.

[0056] Among them, the spatial graph neural network layer, the time series model, the spatiotemporal attention mechanism layer and the output layer are connected in sequence. The spatial relationship between 4G base stations, between 5G base stations, and between 4G base stations and 5G base stations is determined through the learning of the spatiotemporal graph by the spatial graph neural network layer; then, based on the spatial relationship, the time series model is used to learn the changes of the characteristic variables represented by the vertices and edges in the spatiotemporal graph at different time points; then, based on the changes, the spatiotemporal attention mechanism layer is used to capture the associated dependencies between base stations at different time points; finally, based on the associated dependencies, the output layer outputs the prediction results.

[0057] For example, Figure 2 A schematic diagram of the spatiotemporal graph neural network architecture provided in the embodiment of the present application is shown in FIG. Figure 2 As shown in Figure 1, the fusion layer can be used to merge features or information from different sources into a unified representation. The dynamic graph convolution layer can dynamically update node features according to changes in the graph structure. The graph structure estimation layer can be used to estimate or infer the structural information of the graph, such as the connection relationship between nodes. Long short-term memory networks can be used to handle long-term dependency problems in sequence data. Spectral clustering is used to construct a similarity matrix of data and use its feature vectors for clustering. Memory matrices can be used in reinforcement learning or other tasks that require memory mechanisms.

[0058] Since there are complex spatiotemporal relationships between data points in spatiotemporal data, by utilizing STGNN, we can better capture the dependencies and evolution laws between data under this spatiotemporal relationship, learn the spatiotemporal graph structure from the data, and perform information transmission and feature extraction on this structure to improve the accuracy of the prediction and obtain accurate business feature data of 4G base stations in the future target time period.

[0059] Step S14: Based on the service characteristic data of each 4G base station in the future target time period, the base station to be evolved is determined from all 4G base stations included in the space-time graph.

[0060] In this embodiment, a determination condition may be configured, and when the service characteristic data within the future target time period meets the determination condition, the 4G base station is determined to be a base station to be evolved.

[0061] Exemplarily, in order to improve the accuracy of the determination and determine the upgrade time sequence of each base station, further determination refinement can be made. Specifically, the future target time period is divided into N time periods, and then it is determined whether the service feature data in each time period meets the determination conditions. If the service feature data in a certain time period meets the determination conditions, then this time period can be used as the upgrade evolution period of the 4G base station, thereby determining the upgrade evolution period of each 4G base station, and performing the evolution upgrade in sequence according to the time sequence of the upgrade evolution period.

[0062] The determination conditions may be configured according to actual conditions.

[0063] Step S15: Perform mobile communication network upgrade evolution on the base station to be evolved.

[0064] In this embodiment, the base stations to be evolved refer to those 4G base stations that can be directly upgraded and evolved. The mobile communication network is upgraded and evolved through the base stations to be evolved, so that the 4G base stations are directly upgraded to 5G base stations, which improves the deployment efficiency of the 5G network and reduces the construction cost.

[0065] The embodiment of the present application uses a space-time graph to describe the relationship between base stations, and then uses a space-time graph neural network to capture the changes in base station services over time, predict the service feature data of the base station in the future, and then determine whether the base station can be upgraded to a 5G base station based on the future service feature data. This method of integrating space and time to analyze the service feature data of the base station can more accurately determine whether the 4G site can be upgraded and evolved, and by upgrading these 4G base stations, the deployment efficiency of the 5G network can be improved while reducing construction costs.

[0066] The following describes in detail how to obtain the service feature data of a 4G base station within a future target time period through some embodiments.

[0067] In some embodiments, a statistical convergence period can be set in combination with actual conditions, and then the service feature data obtained in the above step S12 can be divided according to the statistical convergence period to form service feature data of T statistical convergence periods. Then, the service feature data and the space-time graph of the T statistical convergence periods are used as the input of the space-time graph neural network to obtain the service feature data of each base station in the next T' upgrade evaluation period. Wherein, T' is a positive integer.

[0068] In this embodiment, illustratively, if the service characteristic data acquired in the above step S12 is the service characteristic data of the past two years, assuming that the statistical aggregation period is set to 1 month, a total of service characteristic data of 24 statistical aggregation periods can be obtained.

[0069] In this embodiment, a spatiotemporal graph neural network STGNN modeling and prediction is performed for each 4G base station (for example, the 4G base station mentioned above that can support the upgrade in hardware), and the implementation process is as follows:

[0070]

[0071] In the above formula, D is input to the spatiotemporal graph neural network STGNN 1 , ..., D T is the service characteristic data of T statistical aggregation periods obtained by the aggregation processing of each base station in the space-time diagram in the previous step S12, D T represents the service feature data of the Tth statistical aggregation period. G is the spatiotemporal graph data. The output of the spatiotemporal graph neural network STGNN is the spatiotemporal feature data of each base station within the spatiotemporal graph in the next T' upgrade evaluation period, D T+T' is the service characteristic data of the T+T'th upgrade evaluation cycle in the future target time period, where T and T' are both positive integers.

[0072] Among them, the service feature data of the 4G base station in the future target time period predicted by the spatiotemporal graph neural network STGNN may refer to the service feature data of T' upgrade evaluation cycles in the above formula.

[0073] In this embodiment, the durations of the statistics aggregation period and the upgrade evaluation period can be configured based on actual conditions.

[0074] The embodiment of the present application utilizes the space-time graph neural network STGNN to capture the geographical mobility patterns of site services over time and realize space-time correlation analysis. In addition, the method of judging the evolution and upgrade of 4G base stations using the space-time dependency data output by the space-time graph neural network STGNN can further improve the overall analysis accuracy.

[0075] Furthermore, in some implementations, after obtaining the service feature data of T' upgrade evaluation cycles, the service feature data of each upgrade evaluation cycle can be matched and compared with the set upgrade conditions to determine whether the 4G base station meets the upgrade conditions. If so, the 4G base station is determined to be a base station to be evolved.

[0076] Exemplarily, assuming that the value of T' is 10, that is, the spatiotemporal graph neural network STGNN outputs the business feature data for the next 10 upgrade evaluation cycles, then a target upgrade evaluation cycle can be selected from these 10 upgrade evaluation cycles, and then it is determined whether the business feature data of the target upgrade evaluation cycle meets the set upgrade conditions. If so, the 4G base station is determined to be a base station to be evolved.

[0077] In other embodiments, the determination conditions may be set to be more complex. For example, it may be first determined whether the service feature data of the first and second consecutive upgrade evaluation cycles in the 10 upgrade evaluation cycles meet the service volume requirements. If the first service volume requirements are met, then the most recent upgrade evaluation cycle (e.g., the third upgrade evaluation cycle) is selected to determine whether the service volume of the neighboring 4G base stations of the base station meets the second service volume requirements in the most recent upgrade evaluation cycle. If so, the 4G base station is determined to be a base station to be evolved.

[0078] The embodiment of the present application can refine the service characteristic data by predicting the service characteristic data of the future T' upgrade evaluation cycles. By using the service characteristic data of each upgrade evaluation cycle in the T' upgrade evaluation cycles, it can more accurately determine whether the 4G base station meets the upgrade conditions, thereby improving the accuracy of the determination.

[0079] Furthermore, in some implementations, the upgrade condition may refer to one of the following conditions:

[0080] (1) In N consecutive upgrade evaluation cycles, the service characteristic data of each upgrade evaluation cycle is less than the lower threshold value, where N is a positive integer.

[0081] (2) During the most recent upgrade evaluation cycle, the service characteristic data of the adjacent 4G base station is less than the upper threshold value.

[0082] The service feature data of adjacent 4G base stations are weighted based on the distance weights between vertices in the space-time graph. The service feature data of the 4G base station in the most recent upgrade evaluation cycle are weighted and superimposed on the adjacent 4G base stations. The most recent upgrade evaluation cycle is the upgrade evaluation cycle closest to the N consecutive upgrade evaluation cycles.

[0083] (3) During the most recent upgrade evaluation cycle, the signal strength of adjacent 4G base stations is greater than a target value, where the target value is the ratio of the set number to the total number of adjacent 4G base stations.

[0084] Exemplarily, it is possible to determine whether the service feature data of the T' upgrade evaluation cycles satisfies the above conditions (1)-(3) in sequence, thereby determining whether the 4G base station is a base station to be evolved. Exemplarily, Figure 3 The 4G base station upgrade evolution determination flow chart provided in the embodiment of the present application is as follows: Figure 3 As shown, the specific steps are as follows:

[0085] Step S31: First, the business volume (i.e., characteristics such as the number of users, business traffic, and wireless utilization) of the 4G base station is detected to determine whether the business volume of the 4G base station in each upgrade evaluation cycle is continuously lower than the preset low business decision threshold (corresponding to the lower threshold value above). If so, proceed to the next step.

[0086] Step S32: Determine the most recent upgrade evaluation cycle that satisfies step S1, and superimpose the business volume to the neighboring 4G base stations around the 4G base station, and further determine whether the above-mentioned neighboring 4G base stations have the above-mentioned business volume indicators exceeding the preset high business threshold (corresponding to the upper limit value of the threshold above). If not, proceed to the next step.

[0087] When the traffic is added to the neighboring 4G base stations around the 4G base station, a weight coefficient can be calculated, and then the traffic is added to the neighboring 4G base stations around the 4G base station according to the proportion of the weight coefficient. The calculation method of the weight coefficient is specifically implemented as follows:

[0088] In the space-time graph, E is the set of all edges, and is the connection between the vertices of the adjacency matrix. The distance weight calculation formula of the connection is:

[0089] Z+W+k*(r-min(d,r)) / r(2)

[0090] In the above formula, Z is the ratio of neighboring sampling points of the same-standard system, W is the ratio of neighboring sampling points of different-standard systems, k is a coefficient, exemplarily ranging from 0.05 to 0.15, r is the effective coverage distance between base stations, and d is the physical distance between the two base stations.

[0091] For example, assume that there are two base stations C and D in the vertex set, and Z is the Z calculated when C is the serving base station. Ci and Z obtained by D as the serving base station Di The mean of is set to 0.12; W is the W calculated by C as the serving base station Ci and W obtained by D as the serving base station DiThe mean of is set to 0.13, and k is set to 0.1; r is the sum of the effective coverage distance of base station C and the effective coverage distance of base station D. Through frequency band evaluation, it is assumed that they are 1km and 1.2km respectively, and the physical distance between the two base stations is 0.35km. The weight value is calculated to be 0.12+0.13+0.1*(2.2-0.35) / 2.2=0.334.

[0092] Step S33: Calculate the ratio of the coverage satisfaction rate of the measurement report data (Measurement Report, MR), the minimized drive test data (Minimization of Drive Tests, MDT), and the neighboring cell sampling point measurement of the 4G neighboring cell within the most recent upgrade evaluation period T, that is, the number of sampling points where the 4G neighboring cell signal strength is greater than the preset threshold (that is, the set number in the above (3)) / the total number of 4G neighboring cell sampling points (that is, the total number of adjacent 4G base stations in the above (3)). If the ratio is greater than the preset threshold, it is considered that the 4G base station can be upgraded to a 5G base station in the most recent upgrade evaluation period, otherwise the upgrade conditions are not met.

[0093] The embodiment of the present application can improve the accuracy of the determination of the 4G base station upgrade by configuring multiple conditions to determine in sequence whether the 4G base station has a recent upgrade evaluation cycle, and accurately screen out which 4G base stations can be upgraded to 5G base stations. At the same time, by dividing the T' upgrade evaluation cycle and finding the most recent upgrade evaluation cycle, it is equivalent to giving a detailed indication of when the 4G base station can be upgraded, providing guidance for subsequent 4G base station upgrade operations.

[0094] Further, in some implementations, after completing the analysis of all 4G base stations, the time and order of the 4G base stations in the area to be analyzed that can be upgraded to 5G can be determined according to the time sequence of the most recent upgrade evaluation cycle. Figure 4 A schematic diagram of a 4G base station upgrade evolution sequence determination process provided in an embodiment of the present application is shown in FIG. Figure 4 As shown in the above Figure 3 On this basis, the following steps are also included:

[0095] Step S41: comparing the most recent upgrade evaluation period of each base station to be evolved, and determining the time sequence between each most recent upgrade evaluation period;

[0096] Step S42: Determine the upgrade evolution order of each base station to be evolved based on the time sequence;

[0097] Step S43: Based on the upgrade evolution sequence, the mobile communication network upgrade evolution is performed on each base station to be evolved in turn.

[0098] In this embodiment, when the most recent upgrade evaluation period of a certain 4G base station meets the above upgrade conditions, the 4G base station will be regarded as a base station to be evolved. Therefore, each base station to be evolved will have a most recent upgrade evaluation period.

[0099] Exemplarily, assume that there are base stations to be evolved 1, 2 and 3, where the most recent upgrade evaluation cycle of base station 1 is the first upgrade evaluation cycle, the most recent upgrade evaluation cycle of base station 2 is the second upgrade evaluation cycle, and the most recent upgrade evaluation cycle of base station 3 is the third upgrade evaluation cycle. Then, in chronological order, base station 1 is upgraded first, followed by base station 2, and finally base station 3 is upgraded.

[0100] The embodiments of the present application can give the time sequence of 4G base station evolution and upgrade, provide strong data support for the subsequent 5G base station planning, and achieve the goal of reducing costs and increasing efficiency.

[0101] The construction process of the space-time graph is introduced below through some embodiments.

[0102] In some embodiments, the construction of the space-time graph is to obtain the spatial correlation dependency between 4G base stations, 5G base stations, and 4G base stations and 5G base stations, and the correlation relationship will be used as the input of the space-time graph neural network model. Among them, the space-time graph specifically includes vertices and edges connecting each vertex, and the vertex can refer to a base station.

[0103] The space-time graph G is represented by G B =(V B , E B , A B ), where V is the vertex of the graph, consisting of base stations, E represents all edge sets, A represents the adjacency matrix, and B represents base station B.

[0104] Figure 5 The flow chart of constructing the space-time graph provided in the embodiment of the present application is as follows: Figure 5 As shown, the method may specifically include the following steps:

[0105] Step S51: Acquire a set of other base stations within a preset range of the 4G base station as a set of vertices of the space-time graph.

[0106] Step S52: Determine the distance weights between the vertices of the adjacency matrix according to the proportion of neighboring cells of the same standard system as the 4G base station, the proportion of neighboring cells of different standard systems as the 4G base station, the effective coverage distance between base stations and the physical distance between base stations.

[0107] Step S53: construct a spatiotemporal graph based on the vertex set and the distance weights between vertices.

[0108] Specifically, the construction of the space-time graph may include the following steps:

[0109] ①Screen base stations that support 4G upgrade to 5G.

[0110] Exemplarily, 4G base stations that can be upgraded can be screened out based on hardware support, and then space-time graphs are constructed for the above 4G base stations one by one, and the space-time graph neural network is modeled one by one.

[0111] ②The scope of the space-time map of the site to be analyzed is determined.

[0112] In this embodiment, the scope of the space-time graph, that is, the vertex set V mentioned above, is to determine the base stations included in the set. Therefore, the scope of the space-time graph for the upgradeable 4G base station B is the base station set with a strong correlation therewith.

[0113] Among them, the proportion of neighboring base stations of the corresponding standard of 4G base stations Z can be obtained i , the switching ratio of neighboring base stations to neighboring base stations of the corresponding standard of 4G base stations W i , the distance between the 4G base station and other 4G base stations or 5G base stations; then based on the neighboring base station ratio Z i , neighboring base station switching ratio W i and distance, and obtain target base stations that meet the conditions from other 4G base stations or 5G base stations to form a base station set.

[0114] Specifically, you can configure the following conditions to filter out the vertex set V:

[0115] {V B |Z i Neighboring base stations greater than the threshold OR W i The neighboring base station OR and V B

[0116] Base stations whose distance is less than the threshold}(3)

[0117] In the above formula, Z i and W i It needs to be calculated through associated data. For details, please refer to the following Figure 6 An introduction to the processing of linked data in .

[0118] Figure 6 The overall flow chart of the mobile communication network upgrade evolution method provided in the embodiment of the present application is as follows: Figure 6 As shown, the method may specifically include the following steps:

[0119] Step S61: 4G base station and 5G base station fusion related data collection and processing.

[0120] Step S62: Creating a spatiotemporal graph structure for fusion analysis.

[0121] Step S63: Collection and processing of time-space associated business characteristic data.

[0122] Step S64: spatiotemporal graph neural network modeling and prediction.

[0123] Step S65: Output the 4G upgrade to 5G evolution plan.

[0124] In this embodiment, before constructing the space-time graph, it is necessary to first collect related data, that is, to construct the space-time graph by fusing the related data.

[0125] Among them, the related data mainly include measurement report data, minimized drive test data, switching data and physical location data between base stations. By integrating the above related data, the spatiotemporal graph relationship between 4G base stations, 5G air battles, and between 4G base stations and 5G base stations is constructed.

[0126] Among them, the measurement report data and minimized drive test data mainly include sampling point time, service base station name, service cell name, service cell signal strength, service cell signal-to-noise ratio, neighboring base station name, neighboring cell name, and neighboring cell signal strength. The switching data mainly includes statistical time, service base station name, service cell name, neighboring base station name, neighboring cell name, and number of switching. The physical location data between sites mainly includes the collection of base station engineering parameter data, including base station name, longitude, latitude, antenna height, frequency band, etc.

[0127] Furthermore, after the associated data is collected, further processing is required. The data processing is mainly aimed at the temporal and spatial aggregation of measurement report data, minimized road test data, and switching data, and further calculations are performed to obtain features that can be used to construct a 45G fusion time-space graph, that is, to aggregate the base station-level data for the past N days.

[0128] Among them, the processing is mainly to obtain the proportion of neighboring base stations of the corresponding standard of 4G base stations Z i , the switching ratio of neighboring base stations to neighboring base stations of the corresponding standard of 4G base stations W i , which specifically includes the following steps:

[0129] Step 1: Obtain the association data between the 4G base station and other 4G base stations or 5G base stations;

[0130] Step 2: Based on the associated data, obtain the total number of neighboring base stations of each standard of the 4G base station Y j , Total number of neighboring base stations X i , the total number of times the 4G base station switches to neighboring base stations of various standards U j , the number of times the 4G base station switches to the neighboring base station V i ;

[0131] Step 3: According to the total number of sampling points Y of neighboring base stations of each standard j, the total number of neighboring base station sampling points X i , determine the ratio Z of neighboring base stations of the corresponding standard of the 4G base station i ;

[0132] Step 4: According to the total number of times U that the 4G base station switches to neighboring base stations of each standard j , the number of times the 4G base station switches to the neighboring base station V i , determine the switching ratio of neighboring base stations to neighboring base stations corresponding to the 4G base station standard W i .

[0133] Specifically, Figure 7 The base station associated data processing flow chart provided in the embodiment of the present application is as follows: Figure 7 As shown, it includes the following steps:

[0134] Step S71: Measurement report data and minimization of drive test data are aggregated and calculated.

[0135] First, remove the abnormal sampling point data, the samples whose service cell signal strength, service cell signal-to-noise ratio and neighboring cell signal strength are all lower than the threshold values; further gather the base station level indicator data for the last N days, including the total number of sampling points Y of neighboring base stations of each standard of the service base station j (such as 4G neighboring base station sampling points, 5G neighboring base station sampling points), service base station name, neighboring base station name, total number of neighboring base station sampling points X i (aggregate the sampling point entries of the neighboring base station measured by the occupied service base station signal); further calculate the proportion Z of the sampling points of the neighboring base station to the sampling points of the corresponding standard of the serving base station i , that is, by X i / Y j (When Y j When Z is 0 i Take 0).

[0136] Step S72: Switch data aggregation calculation.

[0137] Among them, the base station level indicator data of the last N days are gathered, including the total number of switches from the service base station to the neighboring base stations of each standard Uj, the service base station name, the neighboring base station name, and the number of switches from the service base station to the neighboring base station Vi; the switching ratio of the neighboring base station to the corresponding standard of the service base station Wi is further calculated, that is, Vi / Uj (when Uj is 0, Wi is 0).

[0138] ③ Calculation of distance weights between vertices of the adjacency matrix. For details, please refer to the above formula (2).

[0139] The embodiment of the present application uses the construction of a spatiotemporal graph to evaluate the relationship between 4G base stations and 5G base stations for network evolution and upgrade, and conducts a fusion evaluation of the inherent relationship between 4G base stations and 5G base stations, solving the problem of one-sidedness in the existing 4G base station upgrade evaluation method, which analyzes and judges 4G base stations and 5G base stations separately. At the same time, it can deeply explore the spatiotemporal dependency relationship between 4G base stations and 5G base stations in the region, analyze and upgrade sites in a refined manner, ensure user perception, reduce operation and maintenance costs, and further improve 5G deployment efficiency.

[0140] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0141] Figure 8 A schematic diagram of the structure of a mobile communication network upgrade and evolution device provided in an embodiment of the present application, such as Figure 8 As shown, the mobile communication network upgrade evolution device 800 includes a spatiotemporal graph acquisition module 810 , a data acquisition module 820 , a feature prediction module 830 , a base station determination module 840 and an upgrade evolution module 850 .

[0142] Among them, the spatiotemporal graph acquisition module is used to obtain the spatiotemporal graph of the 4G base station, and the spatiotemporal graph is used to describe the spatial correlation and dependency between the 4G base station and other 4G base stations or 5G base stations outside the 4G base station. The data acquisition module is used to obtain the service feature data of each 4G base station contained in the spatiotemporal graph. The feature prediction module is used to predict the service feature data of the 4G base station in the future target time period based on the spatiotemporal graph neural network configured for the 4G base station, the spatiotemporal graph and the service feature data of the 4G base station. The base station determination module is used to determine the base station to be evolved among all 4G base stations contained in the spatiotemporal graph based on the service feature data of each 4G base station in the future target time period; the upgrade evolution module is used to upgrade and evolve the mobile communication network of the base station to be evolved.

[0143] In one example, the spatiotemporal graph neural network includes: a spatial graph neural network layer, a time series model, a spatiotemporal attention mechanism layer and an output layer, wherein the feature prediction module can be specifically used to: determine the spatial relationship between 4G base stations, between 5G base stations, and between 4G base stations and 5G base stations through the learning of the spatiotemporal graph by the spatial graph neural network layer; then based on the spatial relationship, learn the changes of the feature variables represented by the vertices and edges in the spatiotemporal graph at different time points through the time series model; then based on the changes, capture the associated dependencies between base stations at different time points through the spatiotemporal attention mechanism layer; finally, based on the associated dependencies, output the service feature data of the 4G base station in the future target time period through the output layer as the prediction result.

[0144] In one example, the service characteristic data includes at least one of the number of cell users, service traffic, and wireless utilization.

[0145] In an example, the feature prediction module can be specifically used to: divide the service feature data of each base station into service feature data of T statistical convergence periods according to the set statistical convergence period, where T is a positive integer; and then use the service feature data of T statistical convergence periods and the space-time graph as the input of the space-time graph neural network to obtain the service feature data of each base station in the future T' upgrade evaluation periods, where T' is a positive integer.

[0146] In an example, the base station determination module can be specifically used to: determine whether the 4G base station meets the set upgrade conditions based on the service feature data of the 4G base station in the next T' upgrade evaluation cycles; when the 4G base station meets the upgrade conditions, determine the 4G base station as a base station to be evolved.

[0147] In one example, the upgrade condition includes at least one of the following:

[0148] (1) In N consecutive upgrade evaluation cycles, the service characteristic data of each upgrade evaluation cycle is less than the lower threshold value, where N is a positive integer;

[0149] (2) In the most recent upgrade evaluation cycle, the service feature data of the adjacent 4G base station is less than the upper threshold value. The service feature data of the adjacent 4G base station is weighted based on the distance weight between vertices in the space-time graph, and is superimposed on the service feature data of the 4G base station in the most recent upgrade evaluation cycle. The most recent upgrade evaluation cycle is the upgrade evaluation cycle closest to the continuous N upgrade evaluation cycles.

[0150] (3) During the most recent upgrade evaluation cycle, the signal strength of adjacent 4G base stations is greater than a target value, where the target value is the ratio of the set number to the total number of adjacent 4G base stations.

[0151] In an example, the upgrade and evolution module can be specifically used to: compare the most recent upgrade evaluation cycle of each base station to be evolved, and determine the time sequence between each most recent upgrade evaluation cycle; then determine the upgrade and evolution order of each base station to be evolved based on the time sequence; finally, based on the upgrade and evolution order, perform mobile communication network upgrade and evolution on each base station to be evolved in turn.

[0152] In one example, the space-time graph acquisition module can be specifically used to: obtain a set of base stations within a preset range of a 4G base station as a set of vertices of the space-time graph; then determine the distance weights between vertices in the space-time graph based on the proportion of neighboring areas with the same system standard as the 4G base station, the proportion of neighboring areas with different systems standard as the 4G base station, the effective coverage distance between base stations, and the physical distance between base stations; finally, construct a space-time graph based on the vertex set and the distance weights between vertices.

[0153] In an example, the space-time graph acquisition module can be specifically used to: obtain the ratio of neighboring base stations Zi of the corresponding standard of the 4G base station, the ratio of neighboring base stations Wi of the corresponding standard of the 4G base station, and the distance between the 4G base station and other 4G base stations or 5G base stations; then based on the ratio of neighboring base stations Zi, the ratio of neighboring base stations Wi and the distance, obtain target base stations that meet the conditions from other 4G base stations or 5G base stations to form a base station set.

[0154] In an example, the spatiotemporal graph acquisition module can be specifically used to: obtain the association data between a 4G base station and other 4G base stations or 5G base stations; then based on the association data, obtain the total number of neighboring base stations of each standard of the 4G base station Yj, the total number of neighboring base stations Xi, the total number of times the 4G base station switches to neighboring base stations of each standard Uj, and the number of times the 4G base station switches to the neighboring base station Vi; then, according to the total number of sampling points of neighboring base stations of each standard Yj and the total number of sampling points of neighboring base stations Xi, determine the proportion of neighboring base stations of the corresponding standard of the 4G base station Zi; finally, according to the total number of times the 4G base station switches to neighboring base stations of each standard Uj and the number of times the 4G base station switches to the neighboring base station Vi, determine the proportion of neighboring base station switching Wi of the corresponding standard of the 4G base station.

[0155] The device provided in the embodiment of the present application can be used to execute the method in the above-mentioned embodiment. Its implementation principle and technical effect are similar and will not be repeated here.

[0156] Fig. 9 The hardware structure diagram of the electronic device provided in the embodiment of the present application is shown in FIG. Fig. 9 As shown, the electronic device 900 may include a processor 901 and a memory 902 storing computer program instructions.

[0157] Specifically, the processor 901 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiment of the present invention.

[0158] The memory 902 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 902 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In one example, the memory 902 may include a removable or non-removable (or fixed) medium, or the memory 902 is a non-volatile solid-state memory. The memory 902 may be inside or outside the integrated gateway disaster recovery device.

[0159] In one example, the memory 902 may be a read-only memory (ROM). In one example, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or a flash memory, or a combination of two or more of these.

[0160] The memory 902 may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage media device, an optical storage media device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0161] The processor 901 implements the mobile communication network upgrade and evolution method in the above-mentioned embodiment by reading and executing the computer program instructions stored in the memory 902 .

[0162] In one example, the electronic device may further include a communication interface 903 and a bus 904. Fig. 9 As shown, the processor 901, the memory 902, and the communication interface 903 are connected via a bus 904 and communicate with each other.

[0163] The communication interface 903 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiment of the present invention.

[0164] Bus 904 includes hardware, software or both, and the components of online data flow billing equipment are coupled to each other. For example, but not limitation, the bus may include Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), Hyper Transport (HT) interconnection, Industry Standard Architecture (ISA) bus, InfiniBand interconnection, Low Pin Count (LPC) bus, Memory bus, Micro Channel Architecture (MCA) bus, Peripheral Component Interconnect (PCI) bus, PCI-Express (PCI-X) bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 904 may include one or more buses. Although the embodiments of the present invention describe and illustrate specific buses, the present invention considers any suitable bus or interconnection.

[0165] In addition, in combination with the method for upgrading and evolving a mobile communication network in the above embodiments, the embodiments of the present invention may provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any one of the methods for upgrading and evolving a mobile communication network in the above embodiments is implemented.

[0166] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed, implements any one of the methods for upgrading and evolving a mobile communication network in the above embodiments.

[0167] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0168] The functional blocks shown in the above structural block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), appropriate firmware, plug-in, function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or communication link by a data signal carried in a carrier. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, read-only memory (Read-Only Memory, ROM), flash memory, erasable read-only memory (Erasable ReadOnly Memory, EROM), floppy disks, compact disc read-only memory (Compact Disc Read-Only Memory, CD-ROM), optical discs, hard disks, optical fiber media, radio frequency (Radio Frequency, RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0169] It should also be noted that the exemplary embodiments mentioned in the present invention describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiments, or in a different order from the embodiments, or several steps can be performed simultaneously.

[0170] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0171] The above is only a specific implementation of the present invention. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be covered within the protection scope of the present invention.

Claims

1. A method for upgrading and evolving a mobile communication network of a base station, characterized in that: include: Obtain a space-time diagram of the 4G base station, where the space-time diagram is used to describe a spatial association dependency relationship between the 4G base station and other 4G base stations or 5G base stations other than the 4G base station; Obtaining service characteristic data of each 4G base station included in the space-time graph; Based on the spatiotemporal graph neural network configured for the 4G base station, the spatiotemporal graph and the service characteristic data of the 4G base station, predicting the service characteristic data of the 4G base station in a future target time period; Based on the service characteristic data of each 4G base station in the future target time period, determining the base station to be evolved from all 4G base stations included in the space-time graph; Perform mobile communication network upgrade evolution on the base station to be evolved.

2. The method according to claim 1, characterized in that The spatiotemporal graph neural network includes: a spatial graph neural network layer, a time series model, a spatiotemporal attention mechanism layer and an output layer; The predicting the service characteristic data of the 4G base station in a future target time period based on the spatiotemporal graph neural network configured for the 4G base station, the spatiotemporal graph and the service characteristic data of the 4G base station includes: Determine the spatial relationship between 4G base stations, between 5G base stations, and between 4G base stations and 5G base stations by learning the space-time graph by the spatial graph neural network layer; Based on the spatial relationship, the time series model is used to learn the changes of the characteristic variables represented by the vertices and edges in the space-time graph at different time points; Based on the changes, the spatiotemporal attention mechanism layer is used to capture the correlation dependencies between base stations at different time points; Based on the associated dependency, the service feature data of the 4G base station in the future target time period is output through the output layer.

3. The method according to claim 1, characterized in that: The service characteristic data includes at least one of the number of cell users, service flow and wireless utilization rate.

4. The method according to any one of claims 1 to 3, characterized in that The predicting the service characteristic data of the 4G base station in a future target time period based on the space-time graph neural network configured for the 4G base station, the space-time graph, and the service characteristic data of the 4G base station includes: According to the set statistical convergence period, the service characteristic data of each base station is divided to obtain service characteristic data of T statistical convergence periods, where T is a positive integer; The service feature data of T statistical convergence periods and the space-time graph are used as inputs of the space-time graph neural network to obtain the service feature data of each base station in the next T' upgrade evaluation periods, where T' is a positive integer.

5. The method according to claim 4, characterized in that The determining the base station to be evolved from all 4G base stations included in the space-time graph based on the service characteristic data of each 4G base station in the future target time period includes: Determine whether the 4G base station meets the set upgrade conditions based on the service feature data of the 4G base station in the next T' upgrade evaluation cycles; When the 4G base station meets the upgrade condition, the 4G base station is determined to be a base station to be evolved.

6. The method according to claim 5, characterized in that The upgrade conditions include at least one of the following: In N consecutive upgrade evaluation cycles, the service characteristic data of each upgrade evaluation cycle is less than the lower threshold value, where N is a positive integer; In the most recent upgrade evaluation cycle, the service feature data of the adjacent 4G base station is less than the threshold upper limit value, and the service feature data of the adjacent 4G base station is weighted based on the distance weight between vertices in the space-time graph, and is superimposed on the service feature data of the 4G base station in the most recent upgrade evaluation cycle. The most recent upgrade evaluation cycle is the upgrade evaluation cycle closest to the continuous N upgrade evaluation cycles; During the most recent upgrade evaluation cycle, the signal strength of the adjacent 4G base stations is greater than a target value, and the target value is the ratio of a set number to the total number of adjacent 4G base stations.

7. The method according to claim 6, characterized in that The performing mobile communication network upgrade and evolution on the base station to be evolved includes: Compare the most recent upgrade evaluation cycle of each base station to be evolved, and determine the time sequence between each most recent upgrade evaluation cycle; Based on the time sequence, determine the upgrade evolution order of each base station to be evolved; Based on the upgrade evolution sequence, the mobile communication network upgrade evolution is performed on each base station to be evolved in turn.

8. The method according to claim 1, characterized in that The obtaining of the spatiotemporal diagram of the 4G base station includes: Acquire a set of base stations within a preset range of the 4G base station as a set of vertices of a space-time graph; Determine the distance weights between each vertex in the space-time graph according to the ratio of neighboring cells of the same standard system as the 4G base station, the ratio of neighboring cells of different standard systems as the 4G base station, the effective coverage distance between base stations and the physical distance between base stations; The space-time graph is constructed according to the vertex set and the distance weights between vertices.

9. The method according to claim 8, characterized in that The acquiring a base station set within a preset range of the 4G base station includes: Get the ratio Z of neighboring base stations to the neighboring base stations of the corresponding standard of the 4G base station i , the switching ratio W of the neighboring base stations to the neighboring base stations of the standard corresponding to the 4G base station i , the distance between the 4G base station and other 4G base stations or 5G base stations; Based on the neighboring base station ratio Z i , neighboring base station switching ratio W i and distance, obtain the target base station that meets the conditions from the other 4G base stations or 5G base stations to constitute the base station set.

10. The method according to claim 9, characterized in that The obtaining of the ratio of neighboring base stations to neighboring base stations of a standard corresponding to the 4G base station and the ratio of neighboring base stations switched to neighboring base stations of a standard corresponding to the 4G base station includes: Obtaining association data between the 4G base station and other 4G base stations or 5G base stations; Based on the association data, the total number Y of neighboring base stations of each standard of the 4G base station is obtained. j , Total number of neighboring base stations X i , the total number of times the 4G base station switches to neighboring base stations of various standards U j , the number of times the 4G base station switches to a neighboring base station V i ; According to the total number of sampling points Y of neighboring base stations of each standard j , the total number of neighboring base station sampling points X i , determine the ratio Z of neighboring base stations of the corresponding standard of the 4G base station i ; According to the total number of times U that the 4G base station switches to neighboring base stations of various standards j , the number of times the 4G base station switches to a neighboring base station V i , determine the switching ratio W of the neighboring base stations of the neighboring base stations corresponding to the standard of the 4G base station i .

11. A mobile communication network upgrade and evolution device for a base station, characterized in that: include: A space-time graph acquisition module, used to acquire a space-time graph of a 4G base station, wherein the space-time graph is used to describe a spatial correlation dependency relationship between the 4G base station and other 4G base stations or 5G base stations other than the 4G base station; A data acquisition module, used to acquire service characteristic data of each 4G base station included in the space-time diagram; A feature prediction module, configured to predict the service feature data of the 4G base station within a future target time period based on the spatiotemporal graph neural network configured for the 4G base station, the spatiotemporal graph and the service feature data of the 4G base station; A base station determination module, configured to determine a base station to be evolved from among all 4G base stations included in the space-time graph based on service characteristic data of each 4G base station in a future target time period; The upgrade and evolution module is used to upgrade and evolve the mobile communication network of the base station to be evolved.

12. An electronic device, characterized in that: The electronic device comprises: a processor and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the method according to any one of claims 1 to 10.

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

Images