Business simulation generation method, device and equipment based on dynamic communication relationship and storage medium

Abstract

The invention discloses a service simulation generation method based on a dynamic communication relation, and the method comprises the steps: building user service matrixes in different time periods based on service data collected by a network management system, and enabling the user service matrixes to comprise a service size matrix, a service generation frequency matrix and a communication relation matrix; based on the user service matrix, obtaining a logic relation matrix between users, and generating a user service generation feature model; constructing a model training set based on the collected communication service data, and training the user service generation feature model; and in the simulation platform, based on the actually acquired data, the communication service is simulated and generated through the trained service generation feature model. The invention further discloses a service simulation generation device based on the dynamic communication relation, corresponding equipment and a storage medium. According to the invention, the simulation service similar to the real communication service feature rule can be generated, and the accuracy and credibility of service simulation are improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and more specifically, to a service simulation generation method, apparatus, device, and storage medium based on dynamic communication relationships. Background Technology

[0002] In communication network performance simulation verification, the generation of simulated communication services is a crucial step in ensuring the accuracy and reliability of simulation results. Existing methods for generating simulated communication services mainly rely on traditional statistical analysis methods, such as the Poisson flow model or the exponential function model. These methods are based on historical statistical data, constructing parameter models through analytical methods, and then generating user services in the simulation platform. However, these traditional methods have significant drawbacks: First, their descriptions of communication service parameters and characteristics are coarse-grained, making it difficult to capture the fine-grained features of real communication services, such as the dynamic communication relationships between users, the impact of time series changes, and spatial relationships; second, due to the insufficient precision in describing service characteristics, the user services generated in the simulation differ significantly from the real service characteristics, resulting in low credibility of the simulation results and affecting the accuracy of communication network performance simulation verification.

[0003] Furthermore, traditional methods fail to fully consider the impact of dynamic communication relationships on service generation, and their mining of time-series features is relatively simplistic, failing to accurately reflect the changing patterns of communication services at different time stages. For example, peak and off-peak periods of service generation are often closely related to factors such as user behavior and network load, which traditional methods struggle to capture. Therefore, there is an urgent need for a service simulation generation method that can more accurately describe the characteristics of communication services and improve the reliability of simulation results. Summary of the Invention

[0004] In view of at least one defect or improvement need of the prior art, the present invention provides a service simulation generation method, apparatus, device and storage medium based on dynamic communication relationship, which can solve at least one of the problems existing in the background art.

[0005] To achieve the above objectives, according to a first aspect of the present invention, a service simulation generation method based on dynamic communication relationships is provided, the method comprising: Based on the service data collected by the network management system, a user service matrix for different time periods is established. The user service matrix includes a service size matrix, a service occurrence frequency matrix, and a communication relationship matrix. Based on the user business matrix, obtain the logical relationship matrix between users and generate a user business generation feature model. Based on the collected communication service data, a model training set is constructed, and the user service generation feature model is trained. In the simulation platform, communication services are simulated and generated based on real-world data and a trained service generation feature model.

[0006] Furthermore, in the aforementioned service simulation generation method based on dynamic communication relationships, the establishment of user service matrices for different time periods specifically includes: Establish a service size matrix based on service size data collected by the network management system; A communication frequency matrix is ​​established based on the service generation frequency data collected by the network management system; A communication relationship matrix is ​​established based on user ID data collected by the network management system.

[0007] Furthermore, in the above-mentioned service simulation generation method based on dynamic connectivity, the generation of user service generation feature model includes generating a user service feature parameter learning framework based on connectivity. The framework includes a spatial convolution module based on connectivity to learn and obtain spatial features of user services, and a sequence convolution module based on the strength of dynamic connectivity to obtain temporal features of user services under different connectivity strengths.

[0008] Furthermore, in the aforementioned service simulation generation method based on dynamic connectivity, the spatial convolution module based on connectivity learns and acquires the spatial features of user services, specifically including: Construct a fusion matrix based on the user business size matrix and logical relationship matrix; Spatial features of user services are obtained by extracting spatial features from the fusion matrix using a graph convolutional neural network.

[0009] Furthermore, in the aforementioned service simulation generation method based on dynamic connectivity relationships, the sequence convolution module based on the strength of dynamic connectivity relationships obtains user service time features under different connectivity relationship strengths, specifically including: The time series of user services are divided based on different communication strengths; By using sequence convolution operation functions, we can learn the time characteristics of user services under different connectivity strengths.

[0010] Furthermore, in the above-mentioned service simulation generation method based on dynamic communication relationships, obtaining the logical relationship matrix between users includes establishing a logical relationship matrix based on the communication relationships between users, wherein the logical relationship matrix is ​​used to describe the strength of the communication relationships between users.

[0011] According to a second aspect of the present invention, a service simulation generation apparatus based on dynamic connectivity is also provided, comprising: The matrix generation module is used to establish a user service matrix for different time periods based on the service data collected by the network management system. The user service matrix includes a service size matrix, a service generation frequency matrix, and a communication relationship matrix. The model generation module is used to obtain the logical relationship matrix between users based on the user business matrix and generate a user business generation feature model. The model training module is used to construct a model training set based on the collected communication service data and train the user service generation feature model. The service generation module is used to simulate and generate communication services in the simulation platform based on real-collected data and through a trained service generation feature model.

[0012] According to a third aspect of the present invention, a service simulation generation device based on dynamic communication relationships is also provided, which includes at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit performs the steps of any of the methods described above.

[0013] According to a fourth aspect of the present invention, a storage medium is also provided, which stores a computer program executable by a service simulation generation device based on dynamic connectivity relationships, wherein when the computer program is run on the service simulation generation device based on dynamic connectivity relationships, the service simulation generation device based on dynamic connectivity relationships performs the steps of any of the methods described above.

[0014] According to a fifth aspect of the present invention, a computer program product is also provided, comprising a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of any of the methods described above.

[0015] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects: The present invention provides a service simulation generation method based on dynamic connectivity relationships. By utilizing a user service matrix and a connectivity matrix, combined with a graph convolutional neural network and a sequence convolution module, this method can generate simulated services that closely resemble the characteristics and patterns of real communication services. This significantly improves the accuracy and reliability of service simulation, allows for a more detailed description of the spatiotemporal characteristics of service generation, adapts to dynamic changes in different communication scenarios, and enhances the model's generalization ability and simulation efficiency. Furthermore, this method supports multi-scenario simulation verification, reducing testing costs and optimizing communication network performance simulation verification. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating a service simulation generation method based on dynamic connectivity relationships provided in this application embodiment; Figure 2 A schematic diagram of a user service feature parameter learning framework based on spatiotemporal features provided in this application embodiment; Figure 3 A schematic diagram of a sequence learning framework based on connectivity strength provided in an embodiment of this application; Figure 4 This is a schematic diagram of a feature fusion method with different connectivity strengths provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0019] The terms "first," "second," "third," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0020] Figure 1 A flowchart illustrating a service simulation generation method based on dynamic connectivity relationships provided in this application embodiment is shown below. Figure 1 As shown in the figure, an embodiment of this application provides a service simulation generation method based on dynamic communication relationships, which includes the following steps: S101 establishes a user service matrix for different time periods based on the service data collected by the network management system. The user service matrix includes a service size matrix, a service generation frequency matrix, and a communication relationship matrix. S102 Based on the user business matrix, obtain the logical relationship matrix between users and generate a user business generation feature model; S103 Based on the collected communication service data, construct a model training set and train the user service generation feature model; In the simulation platform, S104 simulates the generation of communication services based on real-world data and a trained service generation feature model.

[0021] Specifically, service data is collected from the network management system of the communication network. This data includes, but is not limited to, information such as communication time between users, sender address, receiver address, and service size. The collected data is then labeled with annotations including communication relationships between users and service characteristics for subsequent processing.

[0022] Based on the collected data, user service matrices for different time periods are constructed, including a service size matrix, a service occurrence frequency matrix, and a communication relationship matrix. The service size matrix represents the volume of communication services between users in different time periods, the service occurrence frequency matrix represents the number of communications between users in different time periods, and the communication relationship matrix represents the strength of communication relationships between users.

[0023] Based on the aforementioned user business matrix, the logical relationship matrix between users is obtained through the following steps, and a user business generation feature model is generated: The user business matrix and the connectivity matrix are fused, combining the business size matrix, the business generation frequency matrix, and the connectivity matrix to form a comprehensive matrix. The comprehensive matrix is ​​then normalized to construct an adjacency matrix and a degree matrix. A graph convolutional neural network is used to process the adjacency matrix, extracting spatial feature parameters of user businesses to generate the user business generation feature model.

[0024] A large amount of historical communication service data is collected from the network management system, and the data is labeled with time, user relationships, and service characteristics to form a model training set. The training set includes input data, such as a service size matrix, a service generation frequency matrix, and a communication relationship matrix, and corresponding output data, such as service generation pattern parameters.

[0025] The user business generation feature model is trained using the constructed training set. During training, the model parameters are optimized through backpropagation of the loss function to ensure that the model can accurately learn the generation rules of user businesses. After training is complete, the model weights and parameters are saved.

[0026] In the simulation platform, a trained user service generation feature model is used to simulate and generate communication services based on collected data, such as communication relationships and service characteristics between users. The specific steps are as follows: Input the collected data: Input the collected user communication relationships and business characteristics into the trained model.

[0027] Generating simulated services: The model generates simulated communication services based on the input data, including features such as service size and number of communications.

[0028] Simulation verification: Verify in the simulation platform whether the generated simulated business is similar to the characteristics and patterns of real business, and ensure the accuracy and credibility of the simulation results.

[0029] The service simulation generation method based on dynamic connectivity relationships provided in this application, by utilizing a user service matrix and a connectivity matrix, combined with a graph convolutional neural network and a sequence convolution module, can generate simulated services that closely resemble the characteristics and patterns of real communication services. This significantly improves the accuracy and reliability of service simulation, allows for a more detailed description of the spatiotemporal characteristics of service generation, adapts to dynamic changes in different communication scenarios, and enhances the model's generalization ability and simulation efficiency. Furthermore, this method supports multi-scenario simulation verification, reducing testing costs and optimizing communication network performance simulation verification.

[0030] Optionally, the service simulation generation method based on dynamic communication relationships provided in this application embodiment, specifically includes the following steps in establishing a user service matrix for different time periods: Establish a service size matrix based on service size data collected by the network management system; A communication frequency matrix is ​​established based on the service generation frequency data collected by the network management system; A communication relationship matrix is ​​established based on user ID data collected by the network management system.

[0031] Specifically, the data collected by the network management system is labeled, as shown in Table 1: Table 1. Examples of data collected by the network management system

[0032] Collect communication service data from the network management system, including service size, number of service occurrences, and user ID.

[0033] Based on the collected service size data, a service size matrix is ​​established to represent the end-to-end service size. By statistically analyzing the collected data at different stages, the number of service requests sent between users within that period can be determined. Furthermore, based on the logical relationship between the sender and receiver, a [database structure / database] can be established. Time period ( The user service generation matrix is ​​shown in the following formula:

[0034] in, express Time of the first The size of the business sent from one user to the first user.

[0035] Based on the frequency of business transactions, a communication frequency matrix is ​​established to represent the number of end-to-end communication transactions between users. This can be analyzed according to a time series. Statistical analysis was conducted to obtain... to The number of communications between users within a time period is calculated, and a frequency matrix is ​​constructed as shown in the following formula:

[0036] in, express to Within the time period The number of times a user sends a service to the first user.

[0037] Based on user ID data, a connectivity matrix is ​​established to represent the strength of connectivity between user ends.

[0038] The business simulation generation problem can be transformed into how to generate problems based on... and In the simulation space, end-to-end user services are simulated and generated, that is...

[0039] in, It is the user service simulation generation matrix in the simulation space.

[0040] Considering that the simulated entity models in the simulation space are mainly generated based on the parameters of communication service generation rules, and with the support of the simulation engine, simulated services are generated. In the above formula... It should be an end-to-end business generation pattern parameter matrix, mainly including the mean and variance of business size and the mean and variance of the number of business transactions generated in different time periods, which can be represented as:

[0041]

[0042]

[0043]

[0044] in, These are the mean matrix, standard deviation matrix, mean matrix of generation frequency, and standard deviation matrix of generation frequency of the end-to-end communication service obtained through learning, respectively.

[0045] Optionally, the service simulation generation method based on dynamic connectivity provided in this application embodiment includes generating a user service generation feature model, which includes generating a user service feature parameter learning framework based on connectivity. The framework includes a spatial convolution module based on connectivity to learn and obtain spatial features of user services, and a sequence convolution module based on the strength of dynamic connectivity to obtain temporal features of user services under different connectivity strengths.

[0046] Specifically, based on the user service generation model architecture, it can be seen that user service feature parameters are related to both the communication relationships between users and the collected communication service sequences. Therefore, the generation of user service feature parameters has obvious spatiotemporal characteristics. For this reason, a spatiotemporal learning framework for user service feature parameters based on communication relationships can be established, such as... Figure 2 As shown, the framework includes a data layer, a dataset, a spatial convolution module, a sequence convolution module, and a feature parameter generation part, including feature parameters for the size of communication services and feature parameters for the number of times communication services are generated.

[0047] The design incorporates a spatial convolution module based on connectivity relationships to learn the business volume generation feature patterns of spatial features. This includes mapping and sampling sample data based on connectivity relationships of different strengths, and designing a sequence convolution module based on connectivity relationships of different strengths to improve the accuracy of business volume generation parameter learning.

[0048] Optionally, the service simulation generation method based on dynamic connectivity provided in this application embodiment, wherein the spatial convolution module based on connectivity learns and acquires the spatial features of user services, specifically includes: Construct a fusion matrix based on the user business size matrix and logical relationship matrix; Spatial features of user services are obtained by extracting spatial features from the fusion matrix using a graph convolutional neural network.

[0049] Specifically, in the real world, communication between users is initiated based on actual needs, such as communication relationships. Therefore, a communication relationship matrix is ​​established, marking the mapping relationship between the strength of communication relationships between users and their corresponding user services. The communication relationship matrix is ​​shown in the following formula:

[0050] Considering that matrices and graphs can be transformed into descriptions, we take the sender and receiver as vertices in the graph, communication relationships as edges, and the edge weights as the strength of the communication relationships, respectively. , , , The value is [0,1] and satisfies 0. .

[0051] The user service size matrix or user service frequency matrix can be merged with the user connection relationship matrix to form a fusion matrix based on user services and connection relationships, as shown in the following formula:

[0052] Then to Normalization is performed to construct an adjacency matrix. Based on this, the degree matrix is ​​calculated. ,in .

[0053] Graph Convolutional Networks (GCNs) are commonly used to represent the degree of connection between nodes in a graph. Therefore, we can draw upon GCN methods for solving this problem, and the result can be expressed as follows:

[0054] in, , , They represent time The layer outputs the fusion matrix, adjacency matrix, and weight matrix.

[0055] Optionally, the service simulation generation method based on dynamic connectivity provided in this application embodiment, wherein the sequence convolution module based on the strength of dynamic connectivity obtains user service time features under different connectivity strengths, specifically includes: The time series of user services are divided based on different communication strengths; By using sequence convolution operation functions, we can learn the time characteristics of user services under different connectivity strengths.

[0056] Specifically, there is a hidden correspondence between connectivity and service size, and the number of communications within a time interval. The stronger the connectivity, the greater the number of communications within the time interval. Learning this hidden correlation will help further improve the accuracy of communication service generation parameters. Therefore, based on the conventional time series convolution module, connectivity strength is introduced to separately learn the hidden correspondence between connectivity and service size, and the number of communications within a time interval. Figure 3 As shown.

[0057] A sequence learning framework based on connectivity strength can be internally divided into sections according to connectivity. There are several modules, each corresponding to a different communication relationship. These can be distinguished based on the specific values ​​assigned in the communication relationship matrix. For example, the first module learns the corresponding... The second module teaches the parameter patterns under the association relationship. The parameter patterns under the association relationship follow the same logic and can be flexibly adjusted as needed in actual use. During the learning process, the information of each module is independent. The input source includes two parts: one is the size and interval of the business generated under the current association relationship, and the other is the output value of other association relationships closely related to this association relationship. The information value of the corresponding module in the hidden layer of this association relationship can be obtained through non-linear processing of the two using gate functions and activation functions.

[0058] The sequence learning framework based on connectivity relationships essentially divides the hidden layer into modules according to the strength of connectivity relationships. Each module corresponds to a connectivity relationship, and the modules are not interconnected. The information of each module comes from the output of the hidden layer modules of the current connectivity relationship and other connectivity relationships closely related to that connectivity relationship. Each module contains communication service information of different connectivity relationships. Finally, the output of each module is weighted through backpropagation of the loss function to obtain the characteristic patterns of communication service parameters.

[0059] Suppose that for a sequence of communication service matrices, the scale set of communication relationships is... Indicate, then Hidden layer module The information of each module can be represented as: As shown in the following formula:

[0060] in, This represents the sequence convolution operation function. Information sources include communication relationships Input under conditions, and with Closely related No. Hidden information corresponding to each module.

[0061] By extracting communication service features under dynamic connectivity relationships, the hidden layer obtains information from multiple modules, each representing a communication service with different connectivity strengths. Finally, to improve the accuracy of communication service parameter feature extraction, the feature information from different modules can be fused. This is achieved by adding a fully connected layer after the hidden layer, using the fully connected layer to process each module's information. The final output is obtained by weighting the results. During training, backpropagation can be used to minimize the loss function and update the weight parameters of the fully connected layers. Furthermore, the weight parameters can be adjusted as needed, such as... Figure 4 As shown.

[0062] Optionally, the service simulation generation method based on dynamic communication relationships provided in this application embodiment includes obtaining a logical relationship matrix between users, which involves establishing a logical relationship matrix based on communication relationships between users, and the logical relationship matrix is ​​used to describe the strength of communication relationships between users.

[0063] Specifically, the logical relationship matrix is ​​used to characterize the strength of communication connections between users. It records indicators such as communication frequency, duration, and volume of transactions between users in matrix form, converting these indicators into specific numerical values ​​to intuitively reflect the closeness of communication between users. It not only captures the regularity of communication between users but also dynamically reflects changes in communication relationships, providing data support for communication network analysis.

[0064] In communication service simulation, the logical relationship matrix is ​​a crucial foundation. It provides a characteristic description of user communication behavior for the service generation model, enabling the simulated services to more closely resemble real-world scenarios. By quantifying communication relationships, this matrix supports tasks such as complex network analysis, resource optimization, and path planning, improving the accuracy and reliability of communication network performance simulation verification.

[0065] In the model training process, communication service data is first collected from the network management system, and a user service generation matrix and a connectivity matrix are constructed based on this data. The user service generation matrix records the communication service characteristics between users in different time periods, such as service size and communication frequency; while the connectivity matrix describes the strength of communication connections between users. Subsequently, a spatial convolution module uses a graph convolutional neural network to process the fused matrix, extracting the spatial features of user services and learning the impact of connectivity relationships between users on service generation. Simultaneously, a sequence convolution module divides the time series of user services according to the strength of connectivity relationships and extracts the temporal features under different connectivity strengths through convolution operations, capturing the temporal regularity of service generation. Finally, the extracted spatial and temporal features are fused, and the model parameters are optimized through backpropagation of the loss function to complete the model training. The trained model can generate simulated services with similar characteristics to real services in a simulation platform based on real-time data, providing accurate technical support for communication network performance simulation verification.

[0066] In the model application process, the trained and optimized user service generation model is first deployed to the simulation platform. Before the simulation begins, simulation scenario parameters are set according to actual needs, including network topology, user distribution, simulation duration, etc., and communication relationship data between users is input. This data is used to define the communication interaction patterns between users in the simulation environment. During the simulation, the model dynamically generates simulated user services based on the input communication relationship data and the previously trained user service generation pattern parameters. These simulated services include features such as service size, number of communications, and service type, which can reflect the communication behavior between different users in different time periods. The generated simulated service data is fed back to the simulation platform in real time, driving the simulation engine to run, thereby reproducing service interaction scenarios similar to those in real communication networks in the simulation environment.

[0067] This application significantly improves the accuracy and adaptability of communication service simulation through an innovative service simulation generation method based on dynamic connectivity relationships. First, this method constructs a user service generation feature model, combining spatial and sequential convolutional modules to learn the spatial and temporal features of user services, thereby capturing communication behavior patterns between users in greater detail. This multi-dimensional feature learning approach not only improves the consistency between simulated and real service features but also enhances the model's generalization ability across different communication scenarios. Second, through efficient graph convolutional neural networks (GCNs) and sequential convolutional operations, the model can quickly extract key features and generate simulated services in real time, optimizing simulation efficiency, reducing computational resource requirements, and making it suitable for large-scale network simulation verification. Furthermore, this method can be flexibly applied to various communication network scenarios, including high-density networks, low-bandwidth networks, and bursty traffic scenarios. By dynamically adjusting connectivity relationships and service feature parameters, it stably generates simulated services consistent with real service patterns, providing reliable support for network performance verification in different scenarios. Simultaneously, intelligent service generation and simulation verification reduce dependence on the real network environment, lower testing costs, and improve testing effectiveness. In summary, this application provides an efficient, accurate, and adaptable technical means for performance simulation verification of communication networks, significantly improving the overall performance of communication service simulation.

[0068] Optionally, embodiments of this application also provide a service simulation generation apparatus based on dynamic communication relationships, comprising: The matrix generation module is used to establish a user service matrix for different time periods based on the service data collected by the network management system. The user service matrix includes a service size matrix, a service generation frequency matrix, and a communication relationship matrix. The model generation module is used to obtain the logical relationship matrix between users based on the user business matrix and generate a user business generation feature model. The model training module is used to construct a model training set based on the collected communication service data and train the user service generation feature model. The service generation module is used to simulate and generate communication services in the simulation platform based on real-collected data and through a trained service generation feature model.

[0069] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0070] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0071] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0072] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some service interfaces; indirect couplings or communication connections between devices or units may be electrical or other forms.

[0073] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0074] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0075] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0076] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0077] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of embodiments of this disclosure upon considering the specification and practicing the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described herein. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

[0078] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0079] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A service simulation generation method based on dynamic communication relationships, characterized in that, Includes the following steps: Based on the service data collected by the network management system, a user service matrix for different time periods is established. The user service matrix includes a service size matrix, a service occurrence frequency matrix, and a communication relationship matrix. Based on the user business matrix, obtain the logical relationship matrix between users and generate a user business generation feature model. Based on the collected communication service data, a model training set is constructed, and the user service generation feature model is trained. In the simulation platform, communication services are simulated and generated based on real-world data and a trained service generation feature model.

2. The service simulation generation method based on dynamic communication relationships as described in claim 1, characterized in that, The establishment of user service matrices for different time periods specifically includes: Establish a service size matrix based on service size data collected by the network management system; A communication frequency matrix is ​​established based on the service generation frequency data collected by the network management system; Establish a communication relationship matrix based on user ID data collected by the network management system.

3. The service simulation generation method based on dynamic communication relationships as described in claim 1, characterized in that, The generated user service feature model includes generating a user service feature parameter learning framework based on connectivity relationships. The framework includes a spatial convolution module based on connectivity relationships to learn and obtain spatial features of user services, and a sequence convolution module based on dynamic connectivity relationship strength to obtain temporal features of user services under different connectivity relationship strengths.

4. The service simulation generation method based on dynamic communication relationships as described in claim 3, characterized in that, The spatial convolutional module based on connectivity learns and acquires the spatial features of user services, specifically including: Construct a fusion matrix based on the user business size matrix and logical relationship matrix; Spatial features of user services are obtained by extracting spatial features from the fusion matrix using a graph convolutional neural network.

5. The service simulation generation method based on dynamic communication relationships as described in claim 3, characterized in that, The sequence convolution module based on dynamic connectivity strength obtains user service time features under different connectivity strengths, specifically including: The time series of user services are divided based on different communication strengths; By using sequence convolution operation functions, we can learn the time characteristics of user services under different connectivity strengths.

6. The service simulation generation method based on dynamic communication relationships as described in claim 1, characterized in that, The process of obtaining the logical relationship matrix between users includes establishing a logical relationship matrix based on the communication relationships between users, wherein the logical relationship matrix is ​​used to describe the strength of the communication relationship between users.

7. A service simulation generation device based on dynamic communication relationships, characterized in that, include: The matrix generation module is used to establish a user service matrix for different time periods based on the service data collected by the network management system. The user service matrix includes a service size matrix, a service generation frequency matrix, and a communication relationship matrix. The model generation module is used to obtain the logical relationship matrix between users based on the user business matrix and generate a user business generation feature model. The model training module is used to construct a model training set based on the collected communication service data and train the user service generation feature model. The service generation module is used to simulate and generate communication services in the simulation platform based on real-collected data and through a trained service generation feature model.

8. A service simulation generation device based on dynamic communication relationships, characterized in that, It includes at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program that, when executed by the processing unit, causes the processing unit to perform the steps of the method according to any one of claims 1 to 6.

9. A storage medium, characterized in that, It stores a computer program executable by a service simulation generation device based on dynamic connectivity relationships. When the computer program is run on the service simulation generation device based on dynamic connectivity relationships, it causes the service simulation generation device based on dynamic connectivity relationships to perform the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 6.

Images