Network deployment strategy determination method and device and computer equipment

Abstract

The invention relates to the technical field of communication, in particular to a network deployment strategy determination method and device and computer equipment. The method comprises the following steps: acquiring a network use demand of a user in a real scene and scene description information of the real scene; solving the configuration parameters of the network equipment in the real scene by taking the scene description information as a constraint, and taking the objective function value to meet the network use requirement and maximize the objective function value as a target, so as to obtain a network deployment strategy of the real scene; wherein the objective function value is determined based on a network quality score of the communication network provided by the network equipment, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a method, apparatus and computer device for determining a network deployment strategy. Background Art

[0002] Indoor wireless network deployment refers to achieving full coverage of wireless networks in an indoor environment through reasonable planning and configuration to meet the users' access needs for wireless networks in the area. Indoor wireless network deployment is a comprehensive task that requires comprehensive consideration and implementation from multiple aspects such as demand analysis, network architecture design, equipment selection and configuration, wiring and installation, and testing and optimization. Through reasonable deployment and optimization, full coverage and efficient operation of indoor wireless networks can be achieved, providing users with convenient and stable wireless network services.

[0003] However, traditional indoor wireless network deployment methods usually take signal coverage as the main optimization goal, ignoring the actual user experience and the complexity of the wireless environment, leaving room for improvement in network deployment. Summary of the invention

[0004] Based on this, it is necessary to provide a network deployment strategy determination method, device and computer equipment that can improve the network deployment effect in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for determining a network deployment strategy, the method comprising:

[0006] Obtain the user's network usage requirements in real scenarios and scenario description information of real scenarios;

[0007] With the scenario description information as a constraint, the objective function value meets the network usage requirements, and the objective function value is maximized, the configuration parameters of the network devices in the real scenario are solved to obtain the network deployment strategy for the real scenario;

[0008] The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

[0009] In one of the embodiments, the configuration parameters include at least antenna orientation, channel parameters and transmission power parameters of the network device; the network usage requirements include a focus area in a real scenario and a target communication quality corresponding to the focus area.

[0010] In one embodiment, the network quality score is obtained by inputting the objective quality score and the subjective quality score corresponding to the communication network into a quality evaluation model and calculating the quality evaluation model.

[0011] In one embodiment, both the objective quality score and the subjective quality score are determined based on configuration parameters.

[0012] In one embodiment, the scene description information of the real scene is a two-dimensional image corresponding to the real scene; and the method further includes:

[0013] According to the two-dimensional image of the real scene, a virtual scene model corresponding to the real scene is constructed;

[0014] According to the configuration parameters, a virtual scene model is used to perform signal simulation processing to obtain an objective quality score.

[0015] In one embodiment, constructing a virtual scene model corresponding to the real scene according to the two-dimensional image of the real scene includes:

[0016] Based on the two-dimensional image of the real scene, a digital twin model corresponding to the real scene is constructed;

[0017] Reconstruct the real scene to obtain a wireless reconstruction model;

[0018] The digital twin model and the wireless reconstruction model are fused to obtain a virtual scene model corresponding to the real scene.

[0019] In one embodiment, the real scene is reconstructed to obtain a wireless reconstruction model, including:

[0020] Construct a three-dimensional scene model corresponding to the real scene;

[0021] Construct a ray tracing model based on the three-dimensional scene model;

[0022] Configure the 3D scene model according to the material property information of the real scene;

[0023] Using the ray tracing model, the signal propagation simulation is performed in the configured three-dimensional scene model to obtain the simulation results;

[0024] According to the simulation results, the configured three-dimensional scene model is optimized and adjusted to obtain the wireless reconstruction model.

[0025] In one embodiment, with the scenario description information as a constraint, with the objective function value satisfying the network usage requirements and with the objective of maximizing the objective function value, the configuration parameters of the network devices in the real scenario are solved to obtain the network deployment strategy of the real scenario, including:

[0026] With the scenario description information as a constraint, the objective function value meets the network usage requirements, and the objective function value is maximized, the configuration parameters of the network devices in the real scenario are solved to obtain the deployment strategy to be verified in the real scenario;

[0027] Use the verification model to conduct simulation tests on the deployment strategy to be verified;

[0028] According to the simulation results, the network deployment strategy of the real scene is determined from the deployment strategies to be verified; wherein the verification model is constructed based on the two-dimensional image of the real scene, and the verification model is different from the virtual scene model.

[0029] In a second aspect, the present application further provides a network deployment strategy determination device, the device comprising:

[0030] An acquisition module is used to acquire the network usage requirements of the user in the real scene and the scene description information of the real scene;

[0031] A strategy determination module is used to solve the configuration parameters of network devices in real scenarios with scenario description information as a constraint, with the objective function value satisfying the network usage requirements and with the objective of maximizing the objective function value, to obtain the network deployment strategy for the real scenario;

[0032] The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

[0033] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0034] Obtain the user's network usage requirements in real scenarios and scenario description information of real scenarios;

[0035] With the scenario description information as a constraint, the objective function value meets the network usage requirements, and the objective function value is maximized, the configuration parameters of the network devices in the real scenario are solved to obtain the network deployment strategy for the real scenario;

[0036] The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

[0037] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:

[0038] Obtain the user's network usage requirements in real scenarios and scenario description information of real scenarios;

[0039] With the scenario description information as a constraint, the objective function value meets the network usage requirements, and the objective function value is maximized, the configuration parameters of the network devices in the real scenario are solved to obtain the network deployment strategy for the real scenario;

[0040] The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

[0041] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0042] Obtain the user's network usage requirements in real scenarios and scenario description information of real scenarios;

[0043] With the scenario description information as a constraint, the objective function value meets the network usage requirements, and the objective function value is maximized, the configuration parameters of the network devices in the real scenario are solved to obtain the network deployment strategy for the real scenario;

[0044] The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

[0045] The above-mentioned network deployment strategy determination method, device and computer equipment, the objective function value in the present application is determined based on the network quality score of the communication network provided by the network device. On the basis of satisfying the network quality score, the objective function can also use other multiple objectives as joint optimization conditions. The problem of the complexity of the wireless environment is overcome through multi-objective optimization. In addition, the network quality score is determined according to the objective quality score and subjective quality score corresponding to the communication network. The actual usage experience of the user is taken into account through the subjective quality score, thereby improving the accuracy of network deployment optimization. Furthermore, with the scenario description information as a constraint, it is now possible to generate a network deployment strategy for the real scenario in a targeted manner, thereby further improving the accuracy of network deployment optimization and improving the network deployment optimization effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0047] FIG1 is a schematic diagram of a flow chart of a method for determining a network deployment strategy in an embodiment;

[0048] Figure 2 A schematic diagram of a flow chart of a step of obtaining an objective quality score in one embodiment;

[0049] Figure 3 A schematic diagram of a process for constructing a virtual scene model in one embodiment;

[0050] Figure 4 A schematic diagram of a flow chart of steps for obtaining a wireless reconstruction model in an embodiment;

[0051] Figure 5 A schematic diagram of a flow chart of steps for obtaining a network deployment strategy for a real scenario in an embodiment;

[0052] Figure 6 A structural block diagram of a network deployment strategy determination device in one embodiment;

[0053] FIG. 7 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] Traditional indoor wireless network deployment methods usually take signal coverage as the main optimization target, ignoring the actual user experience and the complexity of the wireless environment. In recent years, with the rapid development of wireless network technology, wireless environment reconstruction technology and user perception models have gradually been applied to the field of wireless network deployment optimization, but there are still some problems to be solved, such as:

[0056] 1. Lack of correspondence between user subjective experience and network performance parameters: There is often a large difference between user subjective feedback and the results of expert analysis based on network statistical data in existing network optimization, which makes it difficult to measure the real user experience in actual situations.

[0057] 2. Limitations of optimization methods: Existing optimization methods usually only perform multi-objective optimization on the entire network or individual users. Targeting individual users ignores network performance capabilities, while targeting the entire network sacrifices user experience. There is a lack of consideration of the balance between the entire network and individual users. It is necessary to form a collaborative optimization capability that takes into account both the overall network and individual experience.

[0058] In an exemplary embodiment, as shown in FIG1 , a network deployment strategy determination method is provided, which is described by taking the method applied to a computer device as an example, and includes the following S101 and S102:

[0059] S101, obtaining the network usage requirements of the user in the real scenario and the scenario description information of the real scenario.

[0060] The real scene refers to the actual environment where the user is located, such as an indoor scene. This scene can include various indoor environments such as home, office, factory, school, etc.

[0061] Network usage requirements refer to the specific requirements and expectations that users put forward for network services, performance, functions, etc. in specific scenarios or environments. For example, if the real scenario is an industrial scenario, the network usage requirements proposed by the user require the network to have high reliability, real-time and anti-interference capabilities to support applications such as automated production lines and remote monitoring. In the construction of smart cities, the network needs to support large-scale device access, low-latency video transmission and other features to support applications such as smart transportation and smart security. In a home environment, users may pay more attention to the network's coverage, ease of use and the interconnection capabilities of smart home devices.

[0062] Furthermore, the scene description information of a real scene refers to a detailed and accurate description and explanation of a specific environment or situation so that others can clearly understand and imagine the scene. This description information usually includes multiple aspects such as the physical environment, layout, facilities, atmosphere, and character behavior of the scene. For network use, the scene description information of a real scene may include the following aspects:

[0063] 1) Physical environment: describes the physical space where the scene is located, such as office, home, public place, etc., as well as the size, shape, layout and decoration of the space.

[0064] 2) Network facilities: List the network devices available in the scenario, such as routers, switches, wireless access points, etc., and explain their configuration, performance, and connection methods.

[0065] 3) User behavior: describes how users in the scenario use the Internet, including the types of devices they use (such as mobile phones, computers, tablets, etc.), the Internet services they access (such as web browsing, video streaming, online games, etc.), and their Internet usage habits.

[0066] 4) Network performance provides information about the scenario's network performance, such as bandwidth, latency, packet loss rate, etc., and how these performance parameters affect the user's network experience.

[0067] 5) Security status: Describe the network security measures in the scenario, such as firewalls, intrusion detection systems, encryption technology, etc., and point out possible security risks and vulnerabilities.

[0068] 6) Other relevant factors: Other factors that may affect network usage in the scenario may also be included, such as power supply, network coverage, electromagnetic interference, etc.

[0069] S102, with the scenario description information as a constraint, with the objective function value satisfying the network usage requirements and with the goal of maximizing the objective function value, the configuration parameters of the network devices in the real scenario are solved to obtain the network deployment strategy for the real scenario.

[0070] What is possible is that, with the scenario description information as a constraint, with the objective function value satisfying the network usage requirements, and with the goal of maximizing the objective function value, solving the configuration parameters of the network equipment in the real scenario means: predicting the actual performance of the communication network in the real scenario under the corresponding configuration parameters (i.e., the objective function value) based on the scenario description information, with the actual performance satisfying the network usage requirements, and with the goal of maximizing the objective function value, determining the optimal solution for the configuration parameters.

[0071] Optionally, the configuration parameters may include: 1) Channel selection: Use a wireless analysis tool (to detect the surrounding wireless environment and select the channel with the least interference. 2) Bandwidth setting. For example, in the 5GHz band, if there is less interference around, you can set it to a higher bandwidth (such as 80MHz or 160MHz) to obtain a higher transmission rate. 3) Signal strength and transmission power: Adjust the transmission power of the router according to the actual environment to ensure signal coverage while reducing unnecessary interference. 4) Modulation mode: In the router settings, you can select a modulation mode that supports a higher data rate to improve the performance of the wireless network. 5) Location parameters: The location of the network device directly affects the coverage and quality of the wireless signal. For example, placing the router in the corner of the room or close to the wall may cause the signal to be blocked or reflected, thereby reducing the penetration and coverage of the signal. If there are other electronic devices around the network device (such as microwave ovens, televisions, computers, etc.), the electromagnetic radiation generated by these devices when working may interfere with the transmission of the network signal, resulting in unstable signals or reduced speed.

[0072] The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

[0073] In one possible implementation, the objective function value may be a function constructed with the network quality score of the communication network provided by the network device as an independent variable. The function may also include characteristics such as network utilization and network device energy consumption as independent variables. For example, the higher the network quality score, the larger the objective function value; conversely, the lower the network quality score, the smaller the objective function value; in another possible implementation, the objective function value may also be the network quality score itself.

[0074] In this embodiment, the objective quality score corresponding to the communication network refers to evaluating the performance and quality of the communication network through a series of quantitative indicators. These scores are usually based on the actual operation data of the network, using professional testing methods and evaluation standards, and can objectively reflect the performance of the network. The objective quality score indicators of the communication network may include:

[0075] 1) Network capacity: refers to the maximum number of users that the network can support or the amount of data that can be processed simultaneously. The larger the network capacity, the higher the load that the network can bear, and the better the user experience will usually be.

[0076] 2) Transfer rate: refers to the speed at which data is transmitted in the network. The faster the transfer rate, the lower the delay users experience when performing activities such as file downloads and video streaming.

[0077] 3) Latency: refers to the time required for data to travel from the sender to the receiver. The lower the latency, the faster the network response, which is especially important for applications with high real-time requirements (such as online games and video conferencing).

[0078] 4) Packet loss rate: The subjective quality score corresponding to the communication network refers to the proportion of data packets lost during network transmission. The lower the packet loss rate, the higher the reliability of network transmission and the better the data integrity is guaranteed.

[0079] 5) Signal strength and quality: For wireless communication networks (such as cellular networks and Wi-Fi networks), signal strength (such as RSRP value) and quality (such as SINR value) are important indicators for measuring network coverage and performance. The higher the signal strength and the better the quality, the more stable the signal received by the user and the higher the communication quality.

[0080] 6) Access success rate and disconnection rate: The access success rate reflects the proportion of users who succeed when trying to access the network, while the disconnection rate reflects the proportion of users who are accidentally disconnected during the communication process. These two indicators together measure the stability and reliability of the network.

[0081] 7) Voice quality score (such as MOS value): For communication networks that provide voice services, voice quality score is an important indicator to measure the user's call experience. MOS value is calculated through subjective testing or objective algorithms, and can quantitatively reflect the clarity, delay and distortion of voice.

[0082] 8) Other special test indicators: Depending on the specific network type and business requirements, other special test indicators may also be included, such as throughput, jitter, spectrum efficiency, and coverage uniformity.

[0083] Furthermore, the subjective quality score corresponding to the communication network refers to the subjective feeling and evaluation of the communication network performance and service quality from the user's perspective. This scoring method usually relies on the user's direct experience and perception, and is therefore closer to the user's actual needs and usage scenarios. The subjective quality score can reflect the user's overall satisfaction with the communication network, including but not limited to the following aspects:

[0084] 1) Voice quality: For communication networks that provide voice services, users will evaluate the clarity, delay, and noise of the call based on their own auditory perception. For example, in mobile communication networks, users may give subjective scores based on whether there are echoes, noise, or interruptions during the call.

[0085] 2) Data transmission speed: When users browse the web, watch videos or download files, they will evaluate the data transmission quality of the communication network based on the data loading speed and smoothness. Fast and stable data transmission can improve user satisfaction.

[0086] 3) Network coverage: When users use the network in different locations within the coverage area, they will evaluate the quality of network coverage based on the strength and stability of the signal. Good network coverage can ensure that users can obtain stable and reliable communication services in various environments.

[0087] 4) Stability and reliability: Users will evaluate the stability and reliability of the communication network based on its performance over a long period of use. For example, frequent disconnections, inability to access, or data loss will reduce user satisfaction.

[0088] 5) Ease of use and service support: Users also give subjective ratings based on the ease of use of the communication network and the quality of service support. This includes aspects such as the complexity of network setup, timeliness of troubleshooting, and the quality of customer service.

[0089] Subjective quality scores are usually collected through questionnaires, user feedback, or online reviews. Operators can use these subjective scores to understand user needs and expectations, and then optimize network performance and service quality. It should be noted that subjective quality scores are affected by many factors, including user personal preferences, usage habits, and network environment, so multiple factors need to be considered when conducting an evaluation. Optionally, the above subjective quality scores can be obtained through questionnaires.

[0090] Furthermore, this embodiment can assign corresponding objective weights and subjective weights to the objective quality score and subjective quality score of the communication network, respectively, and process the objective quality score and subjective quality score based on the objective weight and subjective weight to obtain a network quality score of the communication network.

[0091] In the above-mentioned network deployment strategy determination method, the objective function value in this application is determined based on the network quality score of the communication network provided by the network device. On the basis of satisfying the network quality score, the objective function can also use other multiple objectives as joint optimization conditions. The problem of the complexity of the wireless environment is overcome through multi-objective optimization. In addition, the network quality score is determined according to the objective quality score and subjective quality score corresponding to the communication network. The actual usage experience of the user is taken into account through the subjective quality score, thereby improving the accuracy of network deployment optimization. Furthermore, with the scenario description information as a constraint, it is now possible to generate targeted network deployment strategies for real scenarios, thereby further improving the accuracy of network deployment optimization.

[0092] In an exemplary embodiment, the configuration parameters include at least location parameters, antenna orientation, channel parameters, and transmission power parameters of the network device.

[0093] It is understandable that 1) Location parameters: The location of network equipment will directly affect the propagation and coverage of the signal. Placing the device in the center or high up usually results in better signal coverage. Avoid placing the device in corners, near walls or obstacles to reduce signal blockage and reflection.

[0094] 2) Antenna orientation: For network devices with adjustable antennas, the orientation of the antenna will significantly affect the transmission and reception of signals. By adjusting the antenna direction, the coverage and quality of the signal can be optimized to ensure that the signal can reach the target area more effectively.

[0095] 3) Channel parameters: Channel is the "channel" for wireless signal transmission. Different channels use different frequency ranges. Selecting the channel with the least interference can reduce interference and conflict during signal transmission and improve network stability and transmission speed.

[0096] 4) Transmit power parameter: Transmit power determines the strength of wireless signals sent by network devices. Properly increasing the transmit power can expand the coverage of the signal, but too high a transmit power may cause signal interference and increase energy consumption.

[0097] It is understandable that after changing the configuration parameters of the network equipment, it is necessary to analyze the changes in network performance through network simulation or actual testing. The simulation analysis can be performed in a virtual environment to predict performance changes by simulating network behaviors in different scenarios.

[0098] In an exemplary embodiment, both the objective quality score and the subjective quality score are determined according to the configuration parameters. In this embodiment, the performance change of the communication network in the real scene is simulated by building a virtual environment of the real scene. At this time, it is necessary to first build a virtual scene model corresponding to the real scene. Figure 2As shown, the scene description information of the real scene is a two-dimensional image corresponding to the real scene; the method also includes:

[0099] S201, constructing a virtual scene model corresponding to the real scene according to the two-dimensional image of the real scene.

[0100] Among them, a camera or professional image acquisition device is used to obtain a two-dimensional image of a real scene. The format of the two-dimensional image can be JPEG (Joint Photographic Experts Group) format or GIF (Graphics Interchange Format) format. If there are multiple two-dimensional images, they need to be stitched or fused to form a complete image. First, a series of preprocessing operations are performed on the digital image, including noise removal, image enhancement, binarization, etc., in order to highlight the graphics and line details in the image, laying the foundation for subsequent feature extraction and object recognition. A deep learning model (such as a convolutional neural network) is used to conduct an in-depth analysis of the preprocessed image to extract key features such as shape, line structure, and specific symbols. Combined with the results of feature extraction, the target detection algorithm is used to identify elements such as walls, doors, windows, and furniture in the image. Then, the attribute information of each object is extracted, specifically including: by identifying the dimension annotations (such as numbers and text descriptions) in the image, the dimension information is extracted using optical character recognition technology. For the dimensions of elements such as walls, doors, windows, etc., it may be necessary to combine their shape and position information for calculation. The extraction of material information is relatively complex because the detailed data of the material may not be directly contained in the floor plan.

[0101] Furthermore, based on the extracted attribute information, a 3D (Three-dimensional) modeling engine is used to automatically generate a three-dimensional scene model. Walls, doors, windows, furniture and other elements are presented in three-dimensional form and given corresponding material properties for subsequent signal propagation simulation. The creator can check and adjust the generated model to ensure that the model accurately reflects the real scene. The creator can add, delete, move and modify the size, material and other properties of the elements in the model through the interactive interface, thus constructing a virtual scene model corresponding to the real scene.

[0102] S202, performing signal simulation processing using a virtual scene model according to the configuration parameters to obtain an objective quality score.

[0103] Optionally, the virtual scene model can be an electromagnetic propagation model, which can reflect the signal attenuation and multipath effect in the real scene. In the design process of the communication system, the electromagnetic propagation model can be used to estimate the signal coverage, strength and possible interference. This helps designers optimize the configuration parameters of network equipment (such as antenna layout and communication parameters, etc.).

[0104] Specifically, the virtual scene model includes the network device model corresponding to the network device. After modifying the configuration parameters of the network device model, the virtual scene model is run to simulate the propagation process of electromagnetic waves in the network. The objective quality indicators generated during the simulation process, such as signal strength, coverage, interference, etc., are recorded; and the objective quality score is determined based on the objective quality indicators.

[0105] Furthermore, in order to construct an accurate virtual scene model, in this embodiment, the method of constructing the virtual scene model is as follows, including S301 to S303:

[0106] S301, constructing a digital twin model corresponding to the real scene based on the two-dimensional image of the real scene.

[0107] It can be understood that the digital twin model corresponding to the real scene refers to a virtual model created by using advanced modeling, simulation and information technology to fully and accurately digitally replicate and dynamically simulate specific scenes in the real world. This virtual model not only contains static features such as the three-dimensional geometric form and physical properties of the real scene, but also can reflect the dynamic changes in the scene in real time, such as equipment operation status, personnel activities, environmental conditions, etc.

[0108] Specifically, building a digital twin model corresponding to a real scene can include: using 3D modeling software and 3D reconstruction algorithms to build a 3D scene model of the real scene based on the collected 3D geometric data. This step may require a combination of multiple modeling methods, such as BIM (Building Information Modeling) modeling, oblique photography modeling, laser point cloud modeling, etc. Based on the physical characteristics and operating laws of the real scene, a mathematical model is established to describe its behavior and dynamic changes. This may include physical equations, dynamic models, statistical models, etc., depending on the complexity and requirements of the scene. Integrate the 3D scene model with the physical model, behavior model, data acquisition system, simulation tools, etc. to form a complete digital twin system.

[0109] S302, reconstructing the real scene to obtain a wireless reconstruction model.

[0110] It can be understood that the wireless reconstruction model corresponding to the real scenario refers to a virtual model that dynamically adjusts and optimizes the wireless communication network in a specific real scenario. This model achieves flexible reconstruction and optimal configuration of the wireless communication network by simulating and analyzing factors such as wireless signal propagation characteristics, interference conditions, and user needs in the real scenario.

[0111] In one possible implementation, the real scene is reconstructed to obtain a wireless reconstruction model, such as Figure 4 As shown, including:

[0112] S401, constructing a three-dimensional scene model corresponding to the real scene.

[0113] Optionally, constructing a wireless reconstruction model corresponding to a real scene may include: using 3D modeling software to construct a 3D scene model of the scene based on the collected geometric data, and considering the impact of factors such as the material, height, and layout of the building on the propagation of wireless signals in the 3D scene model.

[0114] S402: construct a ray tracing model according to the three-dimensional scene model.

[0115] Among them, the three-dimensional scene model is a numerical model based on the principle of geometric optics, which is mainly used to simulate the propagation behavior of radio waves in complex environments. It calculates the signal strength at the receiving point by tracing all possible ray paths from the transmitter to the receiver, including direct, reflected and refracted paths. Among them, the three-dimensional scene model includes the three-dimensional scene model based on geometric optics and the three-dimensional scene model based on physical optics.

[0116] S403: configuring the three-dimensional scene model according to the material property information of the real scene.

[0117] Understandably, wireless environment reconstruction is a complex process that involves simulating and reproducing the real-world wireless environment. In this process, the setting of material properties is a crucial step, which directly affects the propagation characteristics of wireless signals in the environment. Different materials have different effects on wireless signals. For example, walls may block or weaken signals, while furniture may reflect or scatter signals. Therefore, accurately setting the electromagnetic properties of materials is key to ensuring the authenticity of the reconstructed environment.

[0118] For example, 1) Penetration loss: For obstacles such as walls, it is necessary to set their penetration loss, that is, the degree of attenuation of the signal when it passes through the material. This is usually a function related to the thickness of the material and the electromagnetic frequency. 2) Reflection coefficient: Furniture, floors, and other surfaces may reflect wireless signals. The reflection coefficient describes the proportion of the signal reflected on the surface of the material. This coefficient is usually a complex number that contains the amplitude and phase information of the reflected signal. 3) Scattering coefficient: Certain materials, such as rough surfaces or fabrics, may scatter wireless signals. The scattering coefficient describes the degree and directionality of signal scattering on the material. 4) Other properties: In addition to the above properties, electromagnetic properties such as the dielectric constant and magnetic permeability of the material may also need to be considered, which will also affect the propagation of wireless signals.

[0119] Specifically, the electromagnetic properties (ie, material property information of the real scene) extracted and measured are applied to the corresponding materials of the above three-dimensional scene model, that is, these property values ​​are input into the three-dimensional scene model.

[0120] S404, using a ray tracing model, performing signal propagation simulation in the configured three-dimensional scene model to obtain a simulation result.

[0121] After the three-dimensional scene model is constructed in S401, a large number of rays can be emitted from the emission source (such as a base station antenna, light source or sound source) to the surrounding space. These rays will propagate in the three-dimensional scene according to the principle of geometric optics, and will intersect, reflect, refract and diffract with objects in the scene. A computer program is used to track the propagation path of each ray and record information such as its intersection with the object, reflection surface, refraction surface and diffraction point.

[0122] In the process of tracing the propagation path of rays, it is necessary to calculate the ray's amplitude, phase, delay, polarization and other parameters according to the propagation path of each ray and the properties of the object. The calculation of these parameters usually depends on electromagnetic field theory, optical principles or acoustic principles, etc. Finally, the parameters of all rays at the receiving point are coherently synthesized to obtain the total signal strength or other related parameters of the receiving point.

[0123] The 3D scene model is used to trace a large number of ray paths and calculate related parameters, so the amount of calculation is usually very large. In order to improve the calculation efficiency, some acceleration algorithms and optimization techniques can be used. For example, the spatial index mechanism is used to quickly locate the intersection of the ray and the object; the partition method, dimensionality reduction method or multi-mirror ray acceleration technology is used to reduce unnecessary calculations; parallel computing resources such as GPU are used to accelerate the calculation process, etc.

[0124] S405, optimizing and adjusting the configured three-dimensional scene model according to the simulation result to obtain a wireless reconstruction model.

[0125] First, analyze the simulation results in detail, including indicators such as signal strength distribution, delay, and jitter. Compare the expected target with the actual simulation results to identify areas with insufficient signal coverage, severe interference, or abnormal signal attenuation. First, analyze the simulation results in detail, including indicators such as signal strength distribution, delay, and jitter. Compare the expected target with the actual simulation results to identify areas with insufficient signal coverage, severe interference, or abnormal signal attenuation.

[0126] Secondly, adjust the material properties, including: 1) Penetration loss: If the simulation results show that the penetration loss of certain walls or obstacles is too large or too small, you need to adjust the penetration loss value in its electromagnetic properties. 2) Reflection coefficient: For surfaces with strong reflection, such as metal or smooth glass, you may need to adjust its reflection coefficient to more accurately simulate the signal reflection. 3) Scattering coefficient: For rough surfaces or complex materials, consider adding scattering effects and adjusting the scattering coefficient to simulate multipath propagation of signals. 4) Move obstacles: If the simulation results show that an obstacle has caused serious interference to signal propagation, consider moving the position of the obstacle in the model or reducing its size. 5) Add or remove buildings: Adjust the layout and number of buildings based on the simulation results to improve signal coverage or reduce interference.

[0127] Then, refine the model details, including: 1) Increase model accuracy: If the simulation results have high requirements for model accuracy, you can consider improving the accuracy of the 3D scene model, such as increasing the level of detail (LOD) of the building, refining the terrain model, etc. 2) Optimize ray tracing parameters: Adjust the parameters in the ray tracing model, such as ray density, tracing depth, etc., to improve the accuracy of the simulation results and calculation efficiency.

[0128] Finally, after completing the above adjustments, rerun the ray tracing model for simulation, compare and analyze the new simulation results with the previous results, and evaluate the effect of the adjustments. Continue to optimize the 3D scene model based on the new simulation results until it meets the expected accuracy and practicality requirements.

[0129] S303, fusing the digital twin model and the wireless reconstruction model to obtain a virtual scene model corresponding to the real scene.

[0130] Optionally, the digital twin model and the wireless reconstruction model are fused, including the following dimensions:

[0131] 1) Data alignment: Ensure that the spatial coordinate systems in the digital twin model and the wireless reconstruction model are consistent. Match the objects in the two models to ensure that they represent the same entity in the real scene.

[0132] 2) Information integration:

[0133] Combine the physical environment information (such as buildings, obstacles, terrain, etc.) in the digital twin model with the wireless propagation information (such as signal strength, delay, jitter, etc.) in the wireless reconstruction model.

[0134] Each object or region is assigned a unique identifier for tracking and reference in the fused model.

[0135] 3) Conflict resolution: During the fusion process, you may encounter inconsistent information in the two models. For example, the digital twin model may show that an area is open, while the wireless reconstruction model may show that the area has signal obstruction. A conflict resolution strategy needs to be developed, such as giving priority to more reliable data sources or performing manual verification.

[0136] 4) Model optimization: Optimize the fused model to improve its performance and accuracy. For example, the model can be simplified to reduce computational complexity, or parameters can be adjusted to better match the actual scenario.

[0137] 5) Verification and testing: Verify and test the fused virtual scene model to ensure that it accurately reflects the physical and wireless characteristics of the real scene. Field test data or expert review can be used to verify the accuracy of the model.

[0138] 6) Output: The output after fusion processing is a virtual scene model that contains the digital twin model and wireless reconstruction model information. This model can be used for various analysis and optimization tasks of wireless communication systems, such as signal coverage prediction, interference analysis, network planning, etc. Then, select the appropriate simulation tool to build a simulation platform. Import the wireless reconstruction model into the simulation platform.

[0139] In this embodiment, through the fusion processing in step S303, a virtual scene model including both physical environment details and wireless propagation characteristics is obtained, which provides a powerful tool for the design and optimization of wireless communication systems.

[0140] In an exemplary embodiment, the network quality score is obtained by inputting the objective quality score and the subjective quality score corresponding to the communication network into the quality evaluation model and then calculating the quality evaluation model. The objective quality score and the subjective quality score are both determined according to the configuration parameters.

[0141] Among them, in the above-constructed virtual scene model, the configuration parameters are input, and the output result obtained is the subjective quality score corresponding to the configuration parameters; correspondingly, the configuration parameters are input into the objective scoring model, and the output of the objective scoring model is the objective quality score; wherein, the training samples of the objective scoring model are real objective scores under different network environments.

[0142] The quality evaluation model can be a multilayer perceptron (MLP), a convolutional neural network (CNN), or a recurrent neural network (RNN), etc. The input of the quality evaluation model is an objective quality score and a subjective quality score, and the output is a network quality score.

[0143] When training the quality evaluation model, firstly, subjective quality indicators fed back by a large number of users in different training samples (different network environments) are collected, and subjective quality scores of different network environments are determined according to the subjective quality indicators. For example, the subjective quality indicators are shown in Table 1 below:

[0144] Table 1

[0145]

[0146] Further, objective quality indicators collected under different training samples (different network environments) are obtained, and objective quality scores of different network environments are determined according to the objective quality indicators; illustratively, the objective quality indicators are shown in the following Table 2:

[0147] Table 2

[0148]

[0149] For each network environment (training sample), the network quality is scored in two aspects, combining the above subjective quality score and objective quality score, aiming to more comprehensively and accurately reflect the actual experience of users in the wireless network environment. First, by analyzing the subjective quality indicators of network fluency, network stability, network security, network speed, network usability and other dimensions, the subjective quality score is determined, and the subjective quality level is determined based on the subjective quality score. The scoring criteria for the subjective quality indicators are shown in Table 3 below:

[0150] Table 3

[0151]

[0152] Secondly, the objective quality score is determined by analyzing objective quality indicators such as signal strength, delay, jitter, and packet loss rate. Specifically, each objective quality indicator is scored based on the objective quality indicator, with a score range of 1-5, where 5 represents the best performance and 1 represents the worst performance. Specifically, the scoring criteria for subjective quality indicators are shown in Table 4 below:

[0153] Table 4

[0154]

[0155] Combined with the scoring standards in Tables 3 and 4 above, the quality scoring model trained in this embodiment combines the subjective quality score U (such as questionnaire surveys, complaint records, etc.) and the objective network score P based on the network statistical data of the same period when calculating the network quality score V. In order to generalize the prediction of the network quality score results, the quality scoring model is a deep learning model, which constructs a nonlinear mapping relationship between the network quality score V and the objective network score P and the subjective quality score U, as shown in the following formula (1):

[0156] V = U ⋅ [ 1 − U ( U + P ) ] [ 1 − ( | U − P | / 100 ) ] + P ⋅ [ 1 − P ( U + P ) ] [ 1 − ( | U − P | / 100 ) ] (1)

[0157] When designing the model structure of the quality scoring model, the model structure may include input layer, hidden layer, output layer, etc., as well as parameters such as activation function and loss function. The input layer contains network performance indicators and user behavior data, including but not limited to signal strength, delay, jitter, user location, usage time, etc. The hidden layer is used to set the number of hidden layers and the number of neurons according to the data characteristics and model complexity. The output layer is used to output the network quality score V, and the activation function is used to map the output value to between 0 and 1 for normalization. When training the quality scoring model, an optimization algorithm is used, the gradient descent method is used to train the model, and a suitable loss function is selected. The mean squared error (MSE) is used to measure the gap between the model prediction result and the network quality score V in the training sample.

[0158] In an exemplary embodiment, the network usage requirement includes a focus area in a real scenario and a target communication quality corresponding to the focus area.

[0159] It is understandable that the focus area usually refers to a geographical area or specific place where the user or organization pays special attention to and needs high-quality communication services during network use. For different focus areas, the user or organization will set corresponding target communication quality indicators to ensure that the network service meets actual needs.

[0160] In the specific implementation process, a virtual scene model corresponding to the real scene can be generated. In the virtual scene model, the user can intuitively mark the areas that need to be focused on to ensure the experience, such as desks, sofas, beds, etc. through the interactive interface. The marking method can be point selection, box selection, drawing polygons, etc., which is convenient for the user to operate flexibly. Furthermore, the user sets the expected network performance indicators for each marked area, such as signal strength threshold, delay upper limit, jitter tolerance, etc. Preset indicator options can be provided, and the user can also customize the indicators according to their own needs.

[0161] Correspondingly, the network quality score V includes sub-network quality scores V1, V2, V3, etc. corresponding to each area of ​​interest. n (where n is any area of ​​interest) are determined by the above quality scoring model. The sub-network quality scores V1, V2, V3, etc. are weighted to obtain a network quality score V; further, each area of ​​interest may correspond to a corresponding scoring threshold.

[0162] In an exemplary embodiment, Figure 5 As shown, with the scenario description information as a constraint, the objective function value meets the network usage requirements, and the objective function value is maximized, the configuration parameters of the network devices in the real scenario are solved to obtain the network deployment strategy of the real scenario, including:

[0163] S501, with the scenario description information as a constraint, with the objective function value satisfying the network usage requirements and with the goal of maximizing the objective function value, the configuration parameters of the network devices in the real scenario are solved to obtain the deployment strategy to be verified in the real scenario.

[0164] Specifically, this embodiment adopts a multi-objective reinforcement learning algorithm when solving the configuration parameters of network devices in a real scene, that is, based on the above virtual scene model, an environment model for multi-objective reinforcement learning algorithm learning is constructed. The environment model includes a state space, an action space, a reward function, etc. Among them:

[0165] 1) State space: The state space defines all possible states in the reinforcement learning environment. In this embodiment, the state may include the current configuration parameters of the network device, signal strength distribution, network traffic conditions, etc.

[0166] 2) Action Space: The action space defines all possible actions or decisions in a reinforcement learning environment.

[0167] In this embodiment, the action may include adjusting the transmit power of the network device, changing the antenna direction, adjusting the channel allocation, etc. The action space should be operable and able to have a real impact on the environment.

[0168] 3) Reward function: The reward function defines the reward or penalty for each action in the reinforcement learning environment. In this embodiment, the reward function may be based on multiple objectives, such as maximizing network coverage, minimizing interference, optimizing network capacity, etc. The reward function should be measurable and accurately reflect the impact of the action on the environment.

[0169] Based on the above components, a multi-objective reinforcement learning environment model can be constructed. In this environment model, the multi-objective reinforcement learning algorithm will try to find the optimal strategy, that is, a series of actions to maximize the cumulative reward and meet multiple goals. By continuously interacting and learning with the environment, the algorithm will gradually optimize the configuration parameters of the network devices to improve the performance and efficiency of the network.

[0170] First, the selection process of the above multi-objective reinforcement learning algorithm is:

[0171] 1) Select the Multi-Objective Deep Q-learning (MODQL) algorithm to optimize multiple objectives at the same time and find the best compromise solution based on the weights between the objectives. The algorithm uses multiple Q functions to approximate each objective function and finds the Pareto optimal solution set by the Pareto optimization method (the examples of the Pareto optimization method used include but are not limited to: genetic algorithm, multi-objective particle swarm optimization algorithm).

[0172] 2) Design an agent and use a deep neural network to approximate the Q function and learn the optimal network deployment strategy. This approach reduces the complexity of the final model. An agent is a system that can make decisions based on the state of the environment.

[0173] 3) Deep Neural Network: Use a deep neural network to approximate the Q function, which can predict the future rewards based on the current network state and the actions performed.

[0174] The specific training process is: use a multi-objective reinforcement learning algorithm to train the agent so that it can select the optimal action according to the state of the environment and maximize the reward function. The agent continuously learns in the simulated environment, tries different actions, and updates the Q function according to the rewards obtained, and finally learns the optimal network deployment strategy.

[0175] The reward function should encourage the agent to find a solution that not only meets user needs but also efficiently utilizes network resources. A possible embodiment includes, but is not limited to, designing a reward function based on factors such as network quality score V, network resource utilization, and device energy consumption.

[0176] The process of using multi-objective reinforcement learning is as follows: based on the trained reinforcement learning agent, according to the user's network usage requirements in the real scenario and the scenario description information of the real scenario, with the scenario description information as the constraint, the objective function value meets the network usage requirements, and the objective function value is maximized as the goal, the configuration parameters of the network equipment in the real scenario are solved, and the optimal network deployment strategy for the real scenario is selected.

[0177] S502, using the verification model to perform a simulation test on the deployment strategy to be verified.

[0178] The verification model is constructed based on the two-dimensional image of the real scene, and the verification model is different from the virtual scene model.

[0179] Optionally, the validation model refers to a simulation environment or framework used to evaluate and verify the effectiveness of the deployment strategy. This model is usually based on a previously constructed virtual scene model and may incorporate a multi-objective reinforcement learning network or other optimization algorithm.

[0180] Specifically, according to the deployment strategy to be verified, configure the relevant parameters and conditions in the verification model to ensure that the model can accurately simulate the actual network environment. Implement the deployment strategy to be verified in the verification model to simulate actual operations such as the configuration of network devices and the distribution of network traffic. During the simulation test, collect data on network performance, such as coverage, throughput, latency, etc., for subsequent analysis and evaluation.

[0181] S503: Determine a network deployment strategy for a real scenario from the deployment strategies to be verified according to the simulation results.

[0182] Specifically, based on the collected performance data, the effect of the deployment strategy to be verified is analyzed to evaluate whether it meets the expected network performance requirements. If the deployment strategy to be verified does not perform well in the simulation test, the strategy can be optimized and adjusted based on the analysis results, and then the simulation test can be conducted again until a satisfactory solution is found.

[0183] Furthermore, after determining the network deployment strategy, the changes in the designer's network quality score V before and after optimization are displayed in the form of a chart, so that the designer can intuitively understand the optimization effect. The designer can fine-tune the network deployment strategy based on the visualization results, such as adjusting the antenna position or adding a new antenna. The designer can evaluate the optimization plan based on the visualization results and provide feedback. Based on the designer's feedback, the digital twin model, the designer's perception model, the optimization algorithm, etc. can be adjusted to further improve the optimization effect.

[0184] In an exemplary embodiment, based on the above network deployment strategy determination method, the network deployment solution determination may be performed in the following scenarios:

[0185] Scenario 1:

[0186] 1) The user uploads the house floor plan and generates a 3D scene model corresponding to the house floor plan;

[0187] 2) The user marks the sofa area (area of ​​interest) in the 3D scene model, sets the expected signal strength threshold to -65 dBm, and sets other indicators, such as a delay limit of 20 milliseconds and a jitter tolerance of 5 milliseconds (i.e., network usage requirements).

[0188] 3) Perform wireless reconstruction on the three-dimensional scene model to obtain a wireless reconstruction model, and construct a digital twin model corresponding to the three-dimensional scene model. Based on the wireless reconstruction model and the digital twin model, a virtual scene model is constructed;

[0189] 4) Use the virtual scene model to simulate the propagation of wireless signals in the room, taking into account the impact of walls, furniture, etc. on the signal. Based on the virtual environment model, calculate the objective quality score of the sofa area under the current configuration parameters of the communication network;

[0190] 5) Determine the subjective quality score of the communication network under the current configuration parameters through the subjective scoring model according to the objective quality score configuration parameters corresponding to the communication network under the current configuration parameters;

[0191] 6) Input the objective quality score and the subjective quality score into the quality score model to obtain the network quality score V;

[0192] 7) A multi-objective reinforcement learning algorithm is used to solve the configuration parameters of network devices in real scenarios with the goal of ensuring that the network quality score V meets the network usage requirements and maximizing the network quality score V, and the network deployment strategy for the real scenario is obtained.

[0193] Scenario 2:

[0194] 1) The user uploads the office floor plan and generates a 3D scene model corresponding to the office floor plan;

[0195] 2) The user marks the desk area (area of ​​interest) in the 3D scene model and sets the expected signal strength threshold to -65 dBm, and sets other indicators, such as a delay limit of 20 milliseconds and a jitter tolerance of 5 milliseconds (i.e., network usage requirements).

[0196] 3) Perform wireless reconstruction on the three-dimensional scene model to obtain a wireless reconstruction model, and construct a digital twin model corresponding to the three-dimensional scene model. Based on the wireless reconstruction model and the digital twin model, a virtual scene model is constructed;

[0197] 4) Use the virtual scene model to simulate the propagation of wireless signals in the room, taking into account the impact of walls, furniture, etc. on the signal. Based on the virtual environment model, calculate the objective quality score of the sofa area under the current configuration parameters of the communication network;

[0198] 5) Determine the subjective quality score of the communication network under the current configuration parameters through the subjective scoring model according to the objective quality score configuration parameters corresponding to the communication network under the current configuration parameters;

[0199] 6) Input the objective quality score and the subjective quality score into the quality score model to obtain the network quality score V;

[0200] 7) A multi-objective reinforcement learning algorithm is used to solve the configuration parameters of network devices in real scenarios with the goal of ensuring that the network quality score V meets the network usage requirements and maximizing the network quality score V, and the network deployment strategy for the real scenario is obtained.

[0201] Scenario 3:

[0202] 1) The user uploads the office floor plan and generates a 3D scene model corresponding to the office floor plan;

[0203] 2) The user marks the customer-dense areas (areas of interest) in the three-dimensional scene model and sets the expected signal strength threshold to -65 dBm, and sets other indicators, such as a delay upper limit of 20 milliseconds and a jitter tolerance of 5 milliseconds (i.e., network usage requirements).

[0204] 3) Perform wireless reconstruction on the three-dimensional scene model to obtain a wireless reconstruction model, and construct a digital twin model corresponding to the three-dimensional scene model. Based on the wireless reconstruction model and the digital twin model, a virtual scene model is constructed;

[0205] 4) Use the virtual scene model to simulate the propagation of wireless signals in the room, taking into account the impact of walls, furniture, etc. on the signal. Based on the virtual environment model, calculate the objective quality score of the sofa area under the current configuration parameters of the communication network;

[0206] 5) Determine the subjective quality score of the communication network under the current configuration parameters through the subjective scoring model according to the objective quality score configuration parameters corresponding to the communication network under the current configuration parameters;

[0207] 6) Input the objective quality score and the subjective quality score into the quality score model to obtain the network quality score V;

[0208] 7) A multi-objective reinforcement learning algorithm is used to solve the configuration parameters of network devices in real scenarios with the goal of ensuring that the network quality score V meets the network usage requirements and maximizing the network quality score V, and the network deployment strategy for the real scenario is obtained.

[0209] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0210] Based on the same inventive concept, the embodiment of the present application also provides a network deployment strategy determination device for implementing the network deployment strategy determination method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more network deployment strategy determination devices provided below can refer to the limitations of the network deployment strategy determination method above, and will not be repeated here.

[0211] In an exemplary embodiment, Figure 6 As shown, a network deployment strategy determination device is provided, including: an acquisition module 11 and a strategy determination module 12, wherein:

[0212] An acquisition module 11 is used to acquire the network usage requirements of the user in a real scenario and the scenario description information of the real scenario;

[0213] A strategy determination module 12 is used to solve the configuration parameters of the network devices in the real scene with the scenario description information as a constraint, the objective function value satisfying the network usage requirements, and the objective function value maximization as a goal, to obtain the network deployment strategy for the real scene;

[0214] The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

[0215] In one of the embodiments, the configuration parameters include at least antenna orientation, channel parameters and transmission power parameters of the network device; the network usage requirements include a focus area in a real scenario and a target communication quality corresponding to the focus area.

[0216] In one embodiment, the network quality score is obtained by inputting the objective quality score and the subjective quality score corresponding to the communication network into the quality evaluation model, and the quality evaluation model calculates the network quality score.

[0217] In one embodiment, both the objective quality score and the subjective quality score are determined based on configuration parameters.

[0218] In one embodiment, the scene description information of the real scene is a two-dimensional image corresponding to the real scene; the device also includes a scoring module, which is used to: construct a virtual scene model corresponding to the real scene according to the two-dimensional image of the real scene;

[0219] According to the configuration parameters, a virtual scene model is used to perform signal simulation processing to obtain an objective quality score.

[0220] In one embodiment, the scoring module is further used to:

[0221] Based on the two-dimensional image of the real scene, a digital twin model corresponding to the real scene is constructed;

[0222] Reconstruct the real scene to obtain a wireless reconstruction model;

[0223] The digital twin model and the wireless reconstruction model are fused to obtain a virtual scene model corresponding to the real scene.

[0224] In one embodiment, the scoring module is further used to:

[0225] Construct a three-dimensional scene model corresponding to the real scene;

[0226] Construct a ray tracing model based on the three-dimensional scene model;

[0227] Configure the 3D scene model according to the material property information of the real scene;

[0228] Using the ray tracing model, the signal propagation simulation is performed in the configured three-dimensional scene model to obtain the simulation results;

[0229] According to the simulation results, the configured three-dimensional scene model is optimized and adjusted to obtain the wireless reconstruction model.

[0230] In one embodiment, the strategy determination module 12 is further used to: solve the configuration parameters of the network devices in the real scene with the scenario description information as a constraint, with the objective function value satisfying the network usage requirements, and with the objective of maximizing the objective function value, to obtain the deployment strategy to be verified in the real scene;

[0231] Use the verification model to conduct simulation tests on the deployment strategy to be verified;

[0232] According to the simulation results, the network deployment strategy of the real scene is determined from the deployment strategies to be verified; wherein the verification model is constructed based on the two-dimensional image of the real scene, and the verification model is different from the virtual scene model.

[0233] Each module in the above network deployment strategy determination device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module above.

[0234] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG7. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a network deployment strategy determination method is implemented. The display unit of the computer device is used to form a visually visible picture, which may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0235] Those skilled in the art will understand that the structure shown in FIG. 7 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0236] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0237] Obtain the user's network usage requirements in real scenarios and scenario description information of real scenarios;

[0238] With the scenario description information as a constraint, the objective function value meets the network usage requirements, and the objective function value is maximized, the configuration parameters of the network devices in the real scenario are solved to obtain the network deployment strategy for the real scenario;

[0239] The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

[0240] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0241] Obtain the user's network usage requirements in real scenarios and scenario description information of real scenarios;

[0242] With the scenario description information as a constraint, the objective function value meets the network usage requirements, and the objective function value is maximized, the configuration parameters of the network devices in the real scenario are solved to obtain the network deployment strategy for the real scenario;

[0243] The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

[0244] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0245] Obtain the user's network usage requirements in real scenarios and scenario description information of real scenarios;

[0246] With the scenario description information as a constraint, the objective function value meets the network usage requirements, and the objective function value is maximized, the configuration parameters of the network devices in the real scenario are solved to obtain the network deployment strategy for the real scenario;

[0247] The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

[0248] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0249] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0250] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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 application.

[0251] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for determining a network deployment strategy, It is characterized in that The method comprises: Obtaining the network usage requirements of the user in a real scenario and scenario description information of the real scenario; Taking the scenario description information as a constraint, taking the objective function value satisfying the network usage requirement, and maximizing the objective function value as a goal, solving the configuration parameters of the network devices in the real scenario to obtain the network deployment strategy for the real scenario; The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

2. The method according to claim 1, It is characterized in that The configuration parameters include at least the antenna orientation, channel parameters and transmission power parameters of the network device; the network usage requirements include the focus area in the real scene and the target communication quality corresponding to the focus area.

3. The method according to claim 1, It is characterized in that The network quality score is obtained by inputting the objective quality score and the subjective quality score corresponding to the communication network into a quality evaluation model and calculating the quality evaluation model.

4. The method according to claim 1, It is characterized in that The objective quality score and the subjective quality score are both determined according to the configuration parameters.

5. The method according to any one of claims 1 to 4, It is characterized in that The scene description information of the real scene is a two-dimensional image corresponding to the real scene; the method further includes: Constructing a virtual scene model corresponding to the real scene according to the two-dimensional image of the real scene; According to the configuration parameters, the virtual scene model is used to perform signal simulation processing to obtain the objective quality score.

6. The method according to claim 5, It is characterized in that The step of constructing a virtual scene model corresponding to the real scene according to the two-dimensional image of the real scene comprises: Constructing a digital twin model corresponding to the real scene according to the two-dimensional image of the real scene; Reconstructing the real scene to obtain a wireless reconstruction model; The digital twin model and the wireless reconstruction model are fused to obtain a virtual scene model corresponding to the real scene.

7. The method according to claim 6, It is characterized in that The reconstructing the real scene to obtain a wireless reconstruction model includes: Constructing a three-dimensional scene model corresponding to the real scene; Constructing a ray tracing model according to the three-dimensional scene model; configuring the three-dimensional scene model according to material property information of the real scene; Using the ray tracing model, the signal propagation simulation is performed in the configured three-dimensional scene model to obtain the simulation results; According to the simulation results, the configured three-dimensional scene model is optimized and adjusted to obtain a wireless reconstruction model.

8. The method according to claim 4, It is characterized in that The method of solving the configuration parameters of the network devices in the real scenario with the scenario description information as a constraint, with the objective function value satisfying the network usage requirement and with the objective of maximizing the objective function value, to obtain the network deployment strategy for the real scenario includes: Taking the scenario description information as a constraint, taking the objective function value satisfying the network usage requirement, and maximizing the objective function value as a goal, solving the configuration parameters of the network device in the real scenario to obtain the deployment strategy to be verified for the real scenario; Using a verification model, a simulation test is performed on the deployment strategy to be verified; According to the simulation results, the network deployment strategy of the real scene is determined from the deployment strategies to be verified; wherein the verification model is constructed based on the two-dimensional image of the real scene, and the verification model is different from the virtual scene model.

9. A network deployment strategy determination device, It is characterized in that The device comprises: An acquisition module, used to acquire the network usage requirements of the user in a real scenario and scenario description information of the real scenario; A strategy determination module is used to solve the configuration parameters of the network devices in the real scenario with the scenario description information as a constraint, with the objective function value satisfying the network usage requirements, and with the objective of maximizing the objective function value, to obtain the network deployment strategy for the real scenario; The objective function value is determined based on a network quality score of the communication network provided by the network device, and the network quality score is determined according to an objective quality score and a subjective quality score corresponding to the communication network.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

11. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

12. A computer program product comprising a computer program, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images