Large-scale signal quality evaluation system for cellular network

Abstract

The invention discloses a large-scale signal quality evaluation system for a cellular network, relates to the technical field of signal processing and analysis, and mainly aims to solve the problem of large deviation of signal quality evaluation of wireless signal transmission in an existing multi-region cooperative working environment. A corresponding surface ID is distributed to each obstacle surface in the model; the tracking module is used for determining a propagation path of all wireless signals from the base station to the user in the three-dimensional scene model by adopting a ray tracing algorithm, and acquiring a surface ID and an incident angle of each reflecting surface on the propagation path; the reflection field fitting module is used for regression of reflection attenuation of all reflection surfaces on the propagation path based on the surface ID and the incident angle; and the radio frequency rendering module is used for determining the communication link receiving power of the user based on the reflection attenuation of all the reflecting surfaces on the propagation path, the free space path loss and the signal transmitting power so as to evaluate the signal quality received by the user.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing and analysis, and in particular to a large-scale signal quality evaluation system for a cellular network. Background Art

[0002] Cellular network multi-region signal quality assessment is an important part of modern communication network planning and optimization, aiming to improve the coverage and service quality of wireless networks. With the surge in mobile communication users and the continuous growth of data traffic, the coverage of a single base station can no longer meet the increasing demand. Therefore, multi-region collaborative work has become the key to improving network performance. By comprehensively analyzing the signal quality, user experience and resource utilization in different regions, the layout and configuration of base stations can be effectively optimized to ensure that the network can maintain stable performance in a high-density environment.

[0003] At present, for outdoor environments, street view images obtained from electronic maps can be used to construct three-dimensional material maps for millimeter wave communications. Combined with ray tracing, the three-dimensional material map can be used to determine the optimal direction of beamforming. The material recognition technology in computer vision is used to identify the material type in the real environment from the scene image, and then used to infer the material reflectivity. The wireless signal strength at the receiving end is then simulated based on the inferred reflectivity. However, the material reflectivity depends not only on the material type, but also on the incident angle and material thickness, factors that cannot be clearly inferred from the image. Therefore, the material reflectivity inferred by computer vision may deviate significantly from the actual value, resulting in a large deviation in the simulated wireless signal strength. Summary of the invention

[0004] In view of this, the present invention provides a cellular network large-scale signal quality assessment system, the main purpose of which is to solve the problem of large deviation in signal quality assessment of wireless signal transmission in the existing multi-region collaborative working environment.

[0005] According to one aspect of the present invention, a cellular network large-scale signal quality assessment system is provided, comprising:

[0006] A model preprocessing module, used to obtain a three-dimensional scene model and assign a corresponding surface ID to each obstacle surface in the three-dimensional scene model;

[0007] A tracking module, used to determine the propagation paths of all wireless signals from the base station to the user in the three-dimensional scene model using a ray tracing algorithm, and to obtain the surface ID and incident angle of each obstacle reflection surface on the propagation path;

[0008] A reflection field fitting module, used for regressing the reflection attenuation of all the reflection surfaces on the propagation path based on the surface ID and the incident angle;

[0009] A radio frequency rendering module is used to determine the user's communication link reception power based on the reflection attenuation of all the reflection surfaces on the propagation path, the free space path loss and the transmission power of the wireless signal, so as to evaluate the signal quality received by each user based on the communication link reception power.

[0010] Furthermore, the tracking module includes:

[0011] A definition unit, configured to define a base station in the three-dimensional scene model as a wireless signal transmitting end, and define the user as a wireless signal receiving end;

[0012] A ray tracing unit is used to use the ray tracing algorithm based on the wireless signal receiving end to trace the wireless signal back to the wireless signal sending end, so as to obtain all the propagation paths between the base station and the user.

[0013] Furthermore, the reflection field fitting module includes:

[0014] A model training unit, used for performing neural network model training processing based on the reflection simulation data to obtain a reflection attenuation model corresponding to each surface ID in the three-dimensional scene model;

[0015] A target determination unit, configured to determine a target reflection attenuation model corresponding to each obstacle surface on the propagation path based on the surface ID;

[0016] The reflection field fitting unit is used to perform fitting processing based on the target reflection attenuation model and the corresponding incident angle to obtain the reflection attenuation of all the reflection surfaces on the propagation path.

[0017] Furthermore, the reflection field fitting module further includes a single simulation unit, and the single simulation unit is used for:

[0018] Setting each obstacle surface in the three-dimensional scene model to be the same material to be analyzed;

[0019] Simulating a propagation path of the wireless signal from the base station to the user, and obtaining an incident angle of each obstacle surface on the propagation path;

[0020] The transmitting power of the base station and the receiving power of the user are acquired, and the transmitting power, the receiving power and the incident angle are integrated into reflection simulation data corresponding to the material to be analyzed.

[0021] Furthermore, the model training unit is also used for:

[0022] Configuring corresponding target material information for each of the surface IDs;

[0023] determining target reflection simulation data from the reflection simulation data based on the target material information;

[0024] A neural network model training process is performed based on the target reflection simulation data to obtain the reflection attenuation model corresponding to the surface ID.

[0025] Furthermore, the radio frequency rendering module includes:

[0026] A path loss determination unit, configured to determine the free space path loss corresponding to each propagation path based on the free propagation distance of each propagation path and the carrier wavelength of the wireless signal;

[0027] A single path rendering unit, configured to perform channel rendering processing on the corresponding propagation path based on the reflection attenuation of all the reflection surfaces on the propagation path and the free space path loss, to obtain a channel rendering result corresponding to the propagation path;

[0028] A receiving power determination unit is used to determine the single-link receiving power corresponding to each user based on the channel rendering result and the transmission power of the wireless signal, and integrate all the single-link receiving powers corresponding to the user to obtain the communication link receiving power of the user.

[0029] Furthermore, the system further comprises an interference assessment module, wherein the interference assessment module is configured to:

[0030] Simulating interference links corresponding to each user based on the three-dimensional scene model;

[0031] Determine the interference link receiving power corresponding to each user based on the reflection field fitting module and the radio frequency rendering module;

[0032] The signal-to-noise-interference ratio received by the user is determined based on the communication link received power and the interference link received power, so that the signal quality received by each user is evaluated based on the signal-to-noise-interference ratio.

[0033] According to another aspect of the present invention, a method for evaluating large-scale signal quality in a cellular network is provided, comprising:

[0034] Acquire a three-dimensional scene model, and assign a corresponding surface ID to each obstacle surface in the three-dimensional scene model;

[0035] Using a ray tracing algorithm to determine the propagation paths of all wireless signals from a base station to a user in the three-dimensional scene model, and obtaining the surface ID and incident angle of each obstacle reflection surface on the propagation path;

[0036] regressing the reflection attenuation of all the reflection surfaces on the propagation path based on the surface ID and the incident angle;

[0037] Determine the communication link receiving power of the user based on the reflection attenuation of all the reflection surfaces on the propagation path, the free space path loss and the transmission power of the wireless signal;

[0038] The received signal strength of each user is determined based on the communication link received power, and the interference link received power of each user is determined; the signal-to-noise-interference ratio received by the user is determined based on the received signal strength and the interference link received power, so that the signal quality received by each user is evaluated based on the signal-to-noise-interference ratio.

[0039] According to another aspect of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, wherein the executable instruction enables a processor to execute operations corresponding to the above-mentioned cellular network large-scale signal quality assessment system.

[0040] According to another aspect of the present invention, there is provided a computer device, comprising a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0041] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the above-mentioned cellular network large-scale signal quality assessment system.

[0042] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0043] The present invention provides a large-scale signal quality assessment system for a cellular network. Compared with the prior art, the present invention applies ray tracing technology to the simulation of the electromagnetic wave transmission process through a model preprocessing module and a tracing module to help evaluate the communication quality of multiple regions of a base station. The neural reflection field that has been successful in optics is also converted to the field of radio frequency (RF) through a reflection field fitting module and a radio frequency rendering module to learn the reflection coefficient under the receiving power of the radio frequency signal. After sampling the signal strength and training the reflection field fitting module, the present invention enables the reflection field fitting module to show significant performance in learning the material reflection coefficient of surfaces at different angles in a large scene, and the predicted receiving power matches the measured value, so that the system can predict the signal strength of different regions, especially the area with dense base stations, thereby identifying the area with weak signal strength. The present invention not only provides a more convenient solution for the reasonable deployment of base stations, but also provides a systematic method and scientific basis for the deployment itself.

[0044] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0046] Figure 1 A schematic diagram showing the structure of a cellular network large-scale signal quality assessment system provided by an embodiment of the present invention is shown;

[0047] Figure 2 A schematic diagram showing the composition and structure of another cellular network large-scale signal quality assessment system provided by an embodiment of the present invention is shown;

[0048] Figure 3 A schematic diagram of a flow chart of a method for evaluating large-scale signal quality in a cellular network provided by an embodiment of the present invention is shown;

[0049] Figure 4 A schematic structural diagram of a computer device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0050] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0051] The embodiment of the present invention provides a cellular network large-scale signal quality assessment system. Figure 1 As shown, the system includes:

[0052] A model preprocessing module 11 is used to obtain a three-dimensional scene model and assign a corresponding surface ID to each obstacle surface in the three-dimensional scene model;

[0053] In an embodiment of the present invention, the model preprocessing module obtains a three-dimensional scene model. The three-dimensional scene model can be constructed by reconstructing a three-dimensional scene using a street view image obtained from an electronic map, including the three-dimensional geometric shape of the object. The model preprocessing module also assigns a corresponding surface ID to each obstacle surface in the obtained three-dimensional scene model, wherein the surface ID is not only used to distinguish the surface type of each obstacle, but also to configure and process the attributes of each obstacle through the surface ID, which is not specifically limited in the embodiment of the present invention.

[0054] It should be noted that the three-dimensional scene model obtained by the model preprocessing module in the embodiment of the present invention may also include the material properties of the object. However, due to the problem of large deviation in the reflectivity calculated based on the material properties, the embodiment of the present invention only assumes that the geometric shape of the three-dimensional scene is known, and focuses on learning complex material reflectivity functions in the later stage to improve the accuracy of material reflectivity calculation. Therefore, in the embodiment of the present invention, the material reflectivity is not a constant for a given material, but a function under multiple influencing factors, including incident angle, thickness, etc., which is not specifically limited in the embodiment of the present invention.

[0055] A tracking module 12, configured to determine the propagation paths of all wireless signals from the base station to the user in the three-dimensional scene model by using a ray tracing algorithm, and to obtain the surface ID and incident angle of each obstacle reflection surface on the propagation path;

[0056] In the embodiment of the present invention, the tracking module uses a ray tracing algorithm to determine the propagation paths of all wireless signals from the base station to the user in the three-dimensional scene model. The ray tracing algorithm is a basic tool for accurately modeling electromagnetic (EM) wave propagation in complex environments.

[0057] It should be noted that the use of ray tracing in a real environment requires a high-quality scene model, including the geometry and material properties of objects. With the advancement of computer vision and LiDAR technology, three-dimensional scene geometry and material estimation have become more and more accurate and efficient. Material estimation in computer vision mainly focuses on identifying material properties that affect visual effects, such as surface color, roughness, and metallic properties. In contrast, in the embodiment of the present invention, the wireless radio frequency RF (Radio Frequency) ray tracing no longer uses computer vision to estimate the reflectivity of the material, but only focuses on the amplitude and phase changes of the RF signal before and after reflection. The reflectivity of the material is determined later through the reflection field fitting module, which is not specifically limited in the embodiment of the present invention.

[0058] In order to accurately simulate the reflectivity of the material in the reflection field fitting module, the tracking module in the embodiment of the present invention also needs to obtain the surface ID and incident angle of each obstacle reflective surface on the propagation path.

[0059] A reflection field fitting module 13, configured to regress the reflection attenuation of all the reflection surfaces on the propagation path based on the surface ID and the incident angle;

[0060] In an embodiment of the present invention, the reflection field fitting module performs regression analysis based on the surface ID and incident angle obtained by the tracking module to obtain the reflection attenuation of all reflecting surfaces on the propagation path. In recent years, neural radiation fields (NeRF) have achieved great success in ray tracing of light. It uses several scene images to learn the continuous volume attribute representation of the scene and generate new images from different angles without the need for a scene model. Its improved NeRF2 model (Neural Radio-Frequency RadianceFields) extends the neural radiation field from optics to electromagnetics, taking into account the reflection, diffraction and scattering of RF signals. NeRF2 represents the scene as a neural radiation field and optimizes the neural network through RF signal measurement. In the case of a transmitter with a known position, NeRF2 can accurately predict the received signal and can predict the power at any position if the data density is sufficient. However, NeRF2 has three main disadvantages. First, its performance depends on data density, and higher prediction errors may occur when the data density is low. Second, even in areas with high data density, it cannot achieve accurate predictions where there are sudden changes in the measurement. Most importantly, although the predicted received power is accurate, the learned volume scene function cannot reflect the real wireless propagation environment.

[0061] In the embodiment of the present invention, the powerful description ability of the neural network is used to propose a neural reflection field for real environment materials, wherein the neural reflection field can learn the material reflection coefficient under different incident angles. In order to achieve efficient and scalable measurement, the received power of the user terminal equipment (UEs) in the cellular network and the three-dimensional scene geometry provided by the electronic map can be used to learn the neural reflection field in a complex outdoor environment, which is not specifically limited in the embodiment of the present invention.

[0062] The radio frequency rendering module 14 is used to determine the user's communication link reception power based on the reflection attenuation of all the reflection surfaces on the propagation path, the free space path loss and the transmission power of the wireless signal, so as to evaluate the signal quality received by each user based on the communication link reception power.

[0063] In an embodiment of the present invention, the radio frequency rendering module determines the user's communication link receiving power based on the reflection attenuation of all the reflection surfaces on the propagation path, the free space path loss and the transmission power of the wireless signal, that is, integrating the reflection attenuation into the ray and link levels, and finally the communication link receiving power at the receiving end (user). The communication link receiving power at the user can reflect the receiving strength of the RF signal. If the receiving strength is low, the base station in the entire three-dimensional scene model can be adjusted, which is not specifically limited in the embodiment of the present invention.

[0064] Further, as a refinement and extension of the specific implementation of the above embodiment, in order to apply the ray tracing technology to the field of wireless signal propagation and accurately determine the propagation direction of the wireless signal, another cellular network large-scale signal quality assessment system is provided, such as Figure 2 As shown, the tracking module 12 in the system includes:

[0065] A definition unit 12_1, configured to define a base station in the three-dimensional scene model as a wireless signal transmitting end, and define the user as a wireless signal receiving end;

[0066] The ray tracing unit 12_2 is used to use the ray tracing algorithm based on the wireless signal receiving end to trace the wireless signal back to the wireless signal sending end, and obtain all the propagation paths between the base station and the user.

[0067] In an embodiment of the present invention, the definition unit defines all base stations in the three-dimensional scene model as wireless signal transmitters (analogous to light sources in light propagation), and defines users as wireless signal receivers (analogous to cameras or imaging terminals in light propagation). In light propagation, the basic principle of the ray tracing algorithm is to start from the camera, trace along the opposite direction of the light, calculate the intersection of the light with the objects in the scene, and consider the influence of the light source and the material of the object, and finally determine the color of the pixel of the camera or imaging terminal. In this embodiment, the ray tracing algorithm starts from the wireless signal receiving terminal, traces along the opposite direction of the wireless signal propagation, and obtains the propagation path from the base station to the user by calculating the reflection points of the wireless signal and the obstacle surface in the three-dimensional scene model.

[0068] It should be noted that the ray tracing algorithm in the embodiment of the present invention is intended to obtain the accurate propagation direction of the wireless signal. When the material of the surface of each obstacle in the three-dimensional scene model is unclear, the receiving power of the receiving end is not directly obtained.

[0069] Further, as a refinement and extension of the specific implementation of the above embodiment, in order to more accurately determine the reflection attenuation of the wireless signal on the obstacle surface in the three-dimensional scene model, another cellular network large-scale signal quality assessment system is provided, such as Figure 2As shown, the reflection field fitting module 13 in the system includes:

[0070] A model training unit 13_1 is used to perform neural network model training processing based on the reflection simulation data to obtain a reflection attenuation model corresponding to each surface ID in the three-dimensional scene model;

[0071] In an embodiment of the present invention, the model training unit performs neural network model training processing based on the reflection simulation data. The neural network model is composed of a multi-layer perceptron, such as a classic multi-layer perceptron (MLP). The MLP consists of eight fully connected layers, using a ReLU activation function and 256 channels, in which one skip connection connects the input to the activation value of the fifth layer.

[0072] It should be noted that as a data-driven method, MLP faces the problem of data density. Usually, the estimation accuracy of reflection attenuation in areas with dense wireless communication is better than that in areas with sparse wireless communication. Therefore, it is necessary to ensure data density through reflection simulation. In the embodiment of the present invention, the system uses a single simulation unit 13_4 to perform reflection simulation processing, specifically:

[0073] (1) Setting each obstacle surface in the 3D scene model to the same material to be analyzed;

[0074] (2) Simulate the propagation path of wireless signals from the base station to the user and obtain the incident angle of each obstacle surface on the propagation path;

[0075] (3) Obtaining the transmission power of the base station and the receiving power of the user, and integrating the transmission power, receiving power and incident angle into reflection simulation data corresponding to the material to be analyzed.

[0076] In the embodiment of the present invention, the model training unit obtains the incident angle, transmission power and reception power of the ray for each reflection point on the transmission path to perform model training, and obtains the reflection attenuation model corresponding to each material to be analyzed.

[0077] However, the materials of each obstacle surface in the three-dimensional scene model simulating the real environment are not necessarily the same, so it is necessary to configure the target material information corresponding to each surface ID for each obstacle surface; then, determine the target reflection simulation data from the reflection simulation data based on the configured target material information; and then perform neural network model training processing based on the target reflection simulation data. The above method can be used to obtain the reflection attenuation model corresponding to the surface ID.

[0078] The reflection attenuation model regresses the attenuation of all reflection points according to the incident angle and surface ID, where θ is the incident angle of the ray at the reflection point and s is the surface ID. The reflection attenuation model learns the amplitude and phase changes of all reflection points. For example, let ΔA(θ,s) and ΔΘ(θ,s) represent the amplitude and phase changes of the ray incident on the sth obstacle surface at an angle θ, respectively. The reflection coefficient can be expressed as δ(θ,s) = ΔA(θ,s)exp (jΔΘ(θ,s)) .

[0079] Unlike NeRF2, the reflection attenuation model uses the incident angle of the ray as input, because the reflectivity of the material depends on the incident angle. Other factors that affect the reflectivity of the material, such as thickness, are excluded from the reflection attenuation model because these factors are static. In order to consider the multipath effect in the RF field, the amplitude and phase of the RF signal need to be considered at the same time. Therefore, a replicated neural network model can also be used to fit the reflection field, where the input of the replicated neural network model is the incident angle of the reflection and the number of the corresponding reflection surface, and the output is the amplitude attenuation and phase attenuation of the reflection.

[0080] A target determination unit 13_2, configured to determine a target reflection attenuation model corresponding to each obstacle surface on the propagation path based on the surface ID;

[0081] In the embodiment of the present invention, since the reflection attenuation model corresponding to each material can fit the reflection attenuation of different materials, for an obstacle surface of a given material, it is necessary to determine the reflection attenuation model corresponding to the material, that is, to determine the target reflection attenuation model corresponding to each obstacle surface on the propagation path based on the surface ID.

[0082] The reflection field fitting unit 13_3 is used to perform fitting processing based on the target reflection attenuation model and the corresponding incident angle to obtain the reflection attenuation of all the reflection surfaces on the propagation path.

[0083] Further, as a refinement and extension of the specific implementation of the above embodiment, in order to accurately obtain the communication link receiving power at a specified location so as to evaluate the signal quality, another cellular network large-scale signal quality evaluation system is provided, such as Figure 2 As shown, the radio frequency rendering module 14 in the system includes:

[0084] A path loss determination unit 14_1, configured to determine a free space path loss corresponding to each propagation path based on a free propagation distance of each propagation path and a carrier wavelength of the wireless signal;

[0085] In the embodiment of the present invention, the path loss determination unit determines the free space path loss corresponding to each propagation path based on the free propagation distance of each propagation path and the carrier wavelength of the wireless signal, wherein the free space path loss refers to the spatial attenuation of the radio electromagnetic wave when it propagates in an unobstructed space, and is usually positively correlated with the transmission distance, that is, the longer the spatial propagation distance, the greater the loss. The specific calculation formula is as follows:

[0086]

[0087] in, represents the free space path loss corresponding to the i-th propagation path from the transmitter to the receiver, λ c Indicates the carrier wavelength of the wireless signal, d i represents the free propagation distance of the i-th propagation path. It represents the phase change caused by the wireless signal propagating from the transmitter to the receiver.

[0088] A single path rendering unit 14_2, configured to perform channel rendering processing on the corresponding propagation path based on the reflection attenuation of all the reflection surfaces on the propagation path and the free space path loss, to obtain a channel rendering result corresponding to the propagation path;

[0089] In the embodiment of the present invention, the single-path rendering unit first determines the reflection attenuation of all reflection surfaces on the propagation path, and the calculation formula is as follows:

[0090]

[0091] Among them, δ k represents the total reflection attenuation on the kth propagation path, r k，l represents the lth reflection point on the kth propagation path, P k represents the set of all reflection points on the kth propagation path, θ(r k，l ) represents the reflection point r k，l The incident angle at k，l ) represents the reflection point r k，l The surface ID, ΔA(θ(r k，l ), s(r k，l )) represents the amplitude change of the signal, ΔΘ(θ(r k，l )) represents the phase change of the signal caused by the reflection at the reflection point.

[0092] Next, multiply the free space path loss by the reflection attenuation, which is The channel rendering result corresponding to the propagation path is obtained, where represents the free space path loss of the kth propagation path.

[0093] It should be noted that for all propagation paths between the transmitter i and the receiver d, the channel rendering results of all paths between the transmitter i and the receiver d can be obtained by summing them up. The formula is as follows:

[0094]

[0095] Among them, R i，d represents the set of propagation paths between the transmitter i and the receiver d, H i，d Represents the channel rendering results of all paths from the sender i and the receiver d.

[0096] The receiving power determination unit 14_3 is used to determine the single-link receiving power corresponding to each user based on the channel rendering result and the transmission power of the wireless signal, and integrate all the single-link receiving powers corresponding to the user to obtain the communication link receiving power of the user.

[0097] In the embodiment of the present invention, the receiving power determining unit determines the single-link receiving power corresponding to each user based on the channel rendering result and the transmission power of the wireless signal, which is expressed as in, Represents the transmission power of the wireless signal of the i-th transmitter.

[0098] It should be noted that the actual cellular network is a network with multiple transmitters, where the i-th transmitter has a power of Let B be the set of transmitters in the cellular network, H i，d The channel rendering results for all paths between sender i and receiver d. The signal received at receiver d is a combination of the signals received from all senders, which can be expressed as:

[0099]

[0100] in, It indicates the wireless radio frequency signal of the kth ray received by the receiving end, where k represents the number of the wireless signal ray.

[0101] Since the received signal is complex-valued, the received power can be calculated as:

[0102]

[0103] in, is the power of the kth ray received by the receiver, which is the square of the modulus of the wireless RF signal corresponding to the kth ray. Here, "modulus" refers to the amplitude of a complex signal, and the corresponding English word is modulus.

[0104] Combining all the above equations, the received power at the receiving end d can be rewritten as follows:

[0105]

[0106] Furthermore, as a refinement and extension of the above-mentioned specific implementation methods, in order to further evaluate the signal quality in the cellular network, another cellular network large-scale signal quality evaluation system is provided, such as Figure 2 As shown, the system further includes an interference assessment module 15, and the interference assessment module 15 is used to:

[0107] Simulating interference links corresponding to each user based on the three-dimensional scene model;

[0108] Determine the interference link receiving power corresponding to each user based on the reflection field fitting module and the radio frequency rendering module;

[0109] The signal-to-noise-interference ratio received by the user is determined based on the communication link received power and the interference link received power, so that the signal quality received by each user is evaluated based on the signal-to-noise-interference ratio.

[0110] In an embodiment of the present invention, after training with a neural network, the system can predict the received power at any position in the environment and deployment of the three-dimensional scene model. Since the system of the present invention can simultaneously predict the received power from multiple transmitters to the receiving end, the received power of the interference link can be obtained while obtaining the received power of the communication link, and the signal-to-noise-interference ratio of the communication at this time can be obtained, which is the signal strength of the communication link divided by the power of the interference link. According to the theory of wireless communication, the communication quality of the communication link is positively correlated with the signal-to-noise-interference ratio, so the signal quality received by each user can be evaluated based on the signal-to-noise-interference ratio.

[0111] In an embodiment of the present invention, after obtaining the distribution of the signal-to-noise-interference ratio in the environment, it is possible to understand the areas with poor signal quality under the current base station deployment. In this case, the quality of the areas with poor signals can be improved by adjusting the deployment of the base station, while ensuring that the signal quality of other areas continues to remain good. Specifically, the global optimal solution can be found in the parameter space of the base station by searching, and the new deployment parameters can use ray tracing to verify the signal quality. In this way, the deployment of the base station can be improved after the signal quality evaluation is quickly obtained within a limited time. There is no need to judge the signal quality of a certain area in an exhaustive manner, nor is it necessary to re-measure after deployment to confirm the quality, thereby greatly improving the efficiency of base station operation and maintenance personnel. This is very beneficial for the current dense 5G base station deployment improvement, and for the future 6G network, the base station density will further increase. The solution of the present invention can help save a lot of time cost and manpower and material resources to help quickly find the optimal solution for base station deployment, thereby guiding the actual deployment of base stations.

[0112] The embodiment of the present invention provides a large-scale signal quality assessment system for a cellular network. Compared with the prior art, the present invention applies the ray tracing technology to the simulation of the electromagnetic wave transmission process through a model preprocessing module and a tracing module to help evaluate the communication quality of multiple regions of the base station. The neural reflection field that has been successful in optics is also converted to the field of radio frequency (RF) through a reflection field fitting module and a radio frequency rendering module to learn the reflection coefficient under the receiving power of the radio frequency signal. After sampling the signal strength and training the reflection field fitting module, the reflection field fitting module of the present invention shows significant performance in learning the material reflection coefficient of surfaces at different angles in a large scene, and the predicted receiving power matches the measured value, so that the system can predict the signal strength of different regions, especially the areas with dense base stations, thereby identifying areas with weak signal strength. The present invention not only provides a more convenient solution for the reasonable deployment of base stations, but also provides a systematic method and scientific basis for the deployment itself.

[0113] As the above Figure 1 The embodiment of the present invention provides a method for evaluating the quality of large-scale signals in a cellular network. Figure 3 As shown, the method includes:

[0114] 101. Acquire a three-dimensional scene model, and assign a corresponding surface ID to each obstacle surface in the three-dimensional scene model;

[0115] 102. Determine the propagation paths of all wireless signals from the base station to the user in the three-dimensional scene model using a ray tracing algorithm, and obtain the surface ID and incident angle of each obstacle reflection surface on the propagation path;

[0116] 103. Regressing the reflection attenuation of all the reflection surfaces on the propagation path based on the surface ID and the incident angle;

[0117] 104. Determine the user's communication link reception power based on the reflection attenuation of all the reflection surfaces on the propagation path, the free space path loss and the transmission power of the wireless signal, so as to evaluate the signal quality received by each user based on the communication link reception power.

[0118] The embodiment of the present invention provides a large-scale signal quality assessment method for a cellular network. Compared with the prior art, the present invention obtains a three-dimensional scene model and assigns a corresponding surface ID to each obstacle surface in the three-dimensional scene model; uses a ray tracing algorithm to determine the propagation path of all wireless signals from the base station to the user in the three-dimensional scene model, and applies the ray tracing technology to the simulation of the electromagnetic wave transmission process to help evaluate the communication quality of multiple regions of the base station. In addition, by regressing the reflection attenuation of all the reflection surfaces on the propagation path based on the surface ID and the incident angle, the neural reflection field that has been successful in optics is converted to the field of radio frequency (RF) to learn the reflection coefficient under the receiving power of the radio frequency signal. After sampling the signal strength and training the reflection attenuation, the present invention shows significant performance in learning the material reflection coefficient of surfaces at different angles in a large scene, and the predicted receiving power matches the measured value, so that the system can predict the signal strength of different regions, especially the area with dense base stations, so as to identify the area with weak signal strength. The present invention not only provides a more convenient solution for the reasonable deployment of base stations, but also provides a systematic method and scientific basis for the deployment itself.

[0119] According to one embodiment of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, and the computer executable instruction can execute the cellular network large-scale signal quality assessment system in any of the above system embodiments.

[0120] Figure 4 A schematic diagram of the structure of a computer device provided according to an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the computer device.

[0121] like Figure 4 As shown, the computer device may include: a processor (processor) 202 , a communication interface (Communications Interface) 204 , a memory (memory) 206 , and a communication bus 208 .

[0122] The processor 202 , the communication interface 204 , and the memory 206 communicate with each other via the communication bus 208 .

[0123] The communication interface 204 is used to communicate with other devices such as clients or other servers.

[0124] The processor 202 is used to execute the program 210, and specifically can execute the relevant steps of the above-mentioned cellular network large-scale signal quality assessment system.

[0125] Specifically, the program 210 may include program codes, which include computer operation instructions.

[0126] The processor 202 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. The one or more processors included in the computer device may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0127] The memory 206 is used to store the program 210. The memory 206 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0128] The program 210 may be specifically configured to enable the processor 202 to perform the following operations:

[0129] A model preprocessing module, used to obtain a three-dimensional scene model and assign a corresponding surface ID to each obstacle surface in the three-dimensional scene model;

[0130] A tracking module, used to determine the propagation paths of all wireless signals from the base station to the user in the three-dimensional scene model using a ray tracing algorithm, and to obtain the surface ID and incident angle of each obstacle reflection surface on the propagation path;

[0131] A reflection field fitting module, used for regressing the reflection attenuation of all the reflection surfaces on the propagation path based on the surface ID and the incident angle;

[0132] A radio frequency rendering module is used to determine the user's communication link reception power based on the reflection attenuation of all the reflection surfaces on the propagation path, the free space path loss and the transmission power of the wireless signal, so as to evaluate the signal quality received by each user based on the communication link reception power.

[0133] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0134] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A cellular network large-scale signal quality assessment system, characterized in that: include: A model preprocessing module, used to obtain a three-dimensional scene model and assign a corresponding surface ID to each obstacle surface in the three-dimensional scene model; A tracking module, used to determine the propagation paths of all wireless signals from the base station to the user in the three-dimensional scene model using a ray tracing algorithm, and to obtain the surface ID and incident angle of each obstacle reflection surface on the propagation path; A reflection field fitting module, used for regressing the reflection attenuation of all the reflection surfaces on the propagation path based on the surface ID and the incident angle; A radio frequency rendering module is used to determine the user's communication link reception power based on the reflection attenuation of all the reflection surfaces on the propagation path, the free space path loss and the transmission power of the wireless signal, so as to evaluate the signal quality received by each user based on the communication link reception power.

2. The system according to claim 1, characterized in that The tracking module includes: A definition unit, configured to define a base station in the three-dimensional scene model as a wireless signal transmitting end, and define the user as a wireless signal receiving end; A ray tracing unit is used to use the ray tracing algorithm based on the wireless signal receiving end to trace the wireless signal back to the wireless signal sending end, so as to obtain all the propagation paths between the base station and the user.

3. The system according to claim 1, characterized in that The reflection field fitting module comprises: A model training unit, used for performing neural network model training processing based on the reflection simulation data to obtain a reflection attenuation model corresponding to each surface ID in the three-dimensional scene model; A target determination unit, configured to determine a target reflection attenuation model corresponding to each obstacle surface on the propagation path based on the surface ID; The reflection field fitting unit is used to perform fitting processing based on the target reflection attenuation model and the corresponding incident angle to obtain the reflection attenuation of all the reflection surfaces on the propagation path.

4. The system according to claim 3, characterized in that The reflection field fitting module further includes a single simulation unit, and the single simulation unit is used for: Setting each obstacle surface in the three-dimensional scene model to be the same material to be analyzed; Simulating a propagation path of the wireless signal from the base station to the user, and obtaining an incident angle of each obstacle surface on the propagation path; The transmitting power of the base station and the receiving power of the user are acquired, and the transmitting power, the receiving power and the incident angle are integrated into reflection simulation data corresponding to the material to be analyzed.

5. The system according to claim 4, characterized in that The model training unit is also used for: Configuring corresponding target material information for each of the surface IDs; determining target reflection simulation data from the reflection simulation data based on the target material information; A neural network model training process is performed based on the target reflection simulation data to obtain the reflection attenuation model corresponding to the surface ID.

6. The system according to claim 1, characterized in that The radio frequency rendering module includes: A path loss determination unit, configured to determine the free space path loss corresponding to each propagation path based on the free propagation distance of each propagation path and the carrier wavelength of the wireless signal; A single path rendering unit, configured to perform channel rendering processing on the corresponding propagation path based on the reflection attenuation of all the reflection surfaces on the propagation path and the free space path loss, to obtain a channel rendering result corresponding to the propagation path; A receiving power determination unit is used to determine the single-link receiving power corresponding to each user based on the channel rendering result and the transmission power of the wireless signal, and integrate all the single-link receiving powers corresponding to the user to obtain the communication link receiving power of the user.

7. The system according to any one of claims 1 to 6, characterized in that: The system further includes an interference assessment module, wherein the interference assessment module is configured to: Simulating interference links corresponding to each user based on the three-dimensional scene model; Determine the interference link receiving power corresponding to each user based on the reflection field fitting module and the radio frequency rendering module; The signal-to-noise-interference ratio received by the user is determined based on the communication link received power and the interference link received power, so that the signal quality received by each user is evaluated based on the signal-to-noise-interference ratio.

8. A method for evaluating large-scale signal quality in a cellular network, characterized in that: include: Acquire a three-dimensional scene model, and assign a corresponding surface ID to each obstacle surface in the three-dimensional scene model; Using a ray tracing algorithm to determine the propagation paths of all wireless signals from a base station to a user in the three-dimensional scene model, and obtaining the surface ID and incident angle of each obstacle reflection surface on the propagation path; regressing the reflection attenuation of all the reflection surfaces on the propagation path based on the surface ID and the incident angle; Determine the communication link receiving power of the user based on the reflection attenuation of all the reflection surfaces on the propagation path, the free space path loss and the transmission power of the wireless signal; The received signal strength of each user is determined based on the communication link received power, and the interference link received power of each user is determined; the signal-to-noise-interference ratio received by the user is determined based on the received signal strength and the interference link received power, so that the signal quality received by each user is evaluated based on the signal-to-noise-interference ratio.

9. A storage medium, wherein at least one executable instruction is stored in the storage medium, and the executable instruction executes operations corresponding to the cellular network large-scale signal quality assessment system according to any one of claims 1 to 7.

10. A computer device, comprising a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the cellular network large-scale signal quality assessment system according to any one of claims 1-7.

Images