6G digital twinning online channel modeling method

Abstract

The invention discloses a 6G digital twinning online channel modeling method, which comprises the following steps of: based on information acquired from an environment data source, carrying out communication scene online identification by utilizing a machine learning model; matching channel parameters of a 6G universal channel model according to the identified communication scene, and mapping the channel parameters to the communication scene; reconstructing the digital twin online channel of the static scene in combination with the obtained channel parameters; dynamically sensing the physical environment, acquiring environment change information, and obtaining dynamic sensing data; and in combination with a static-dynamic mixed channel modeling algorithm and dynamic sensing data, dynamically updating the digital twin online channel of the reconstructed static scene to obtain a new real-time twin channel, and constructing a visual platform of the digital twin online channel. According to the method, the channel characteristics of the existing network communication environment can be reflected in real time, real-time sensing, prediction and decision making can be realized, the pilot frequency overhead is reduced, and the system performance is improved.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technologies, and in particular to a 6G digital twin online channel modeling method. Background Art

[0002] 6G (Sixth Generation) wireless communication networks are expected to have full coverage, full application, full digital, full spectrum, full sensory and strong security. Traditional network optimization relies on costly, time-consuming and risky methods such as channel measurement, trial and error and engineering experience. In addition, there are many adjustable parameters in actual wireless networks, and the nonlinear prediction of network performance and user traffic is extremely complex. Therefore, offline network optimization is not effective, and most networks can only achieve about 60% of their performance, and there is still significant room for improvement.

[0003] Online channel models that can accurately reflect the real physical environment and assist in online network optimization are crucial for the deployment and performance improvement of 6G networks. However, the test performance of the existing network communication system is inconsistent with the laboratory simulation performance. There are two main reasons: first, the scene of the existing network test has not been measured by the channel, and the channel model used for laboratory simulation does not match the actual measured environment (channel) of the existing network; second, even if there is a channel model that meets the actual measured scene of the existing network, its channel model parameters still face fixed limitations and are difficult to adapt to the time-varying actual measured scene of the existing network, resulting in the channel simulation model cannot match the scene of the existing network test in real time. Therefore, there is an urgent need for DTOCM (Digital Twin Online Channel Models) that can reflect the channel characteristics of the existing network communication environment in real time. Summary of the invention

[0004] The purpose of the present invention is to provide a 6G digital twin online channel modeling method, which can synchronize the performance between the simulated network and the actual network by realizing continuous visualization and accurate prediction of dynamic channel changes.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A 6G digital twin online channel modeling method includes the following steps:

[0007] S1. Based on the information obtained from the environmental data source, the machine learning model is used to perform online recognition of communication scenarios.

[0008] S2. According to the identified communication scenario, match the channel parameters of 6GPCM (6G Pervasive Channel Model) and map them to the communication scenario.

[0009] S3. Combined with the acquired channel parameters, the digital twin online channel of the static scene is reconstructed.

[0010] S4. Dynamically perceive the physical environment and obtain dynamic perception data.

[0011] S5. Combining the static-dynamic hybrid channel modeling algorithm and dynamic perception data, the digital twin online channel of the reconstructed static scene is dynamically updated to obtain a new real-time twin channel, including the physical features of the static scene and the objects of the dynamic scene.

[0012] S6. Build a visualization platform for the digital twin online channel.

[0013] Furthermore, in step S1, the online identification of the communication scenario includes the following sub-steps:

[0014] S101. Environmental data sources include electronic maps, remote sensing images and point cloud data; electronic maps provide information about terrain, buildings and roads; remote sensing images are obtained through satellites or drones and are used to identify terrain features and environmental changes; three-dimensional point cloud data are generated by lidar or other sensors to provide the three-dimensional spatial structure of the environment.

[0015] S102. Extract and process information from environmental data sources through machine learning models, classify communication scenarios into different categories, and further subdivide these categories.

[0016] Furthermore, in step S2, matching channel parameters includes the following:

[0017] The channel parameter matrix of 6GPCM is expressed as:

[0018] H=[PL·SH·BL·WE·AL] 1 / 2 ·H s

[0019] Among them, H is the channel matrix, PL is the path loss, SH is the shadow fading, BL is the blocking effect, AL is the atmospheric absorption loss, WE is the weather effect loss, and H s Small scale fading.

[0020] Small scale fading H s It is expressed as:

[0021]

[0022] Among them, M T is the number of antenna elements in the transmitting antenna array, M R is the number of antenna elements in the receiving antenna array, f c is the carrier frequency, R is the receiving end, T is the transmitting end, τ is the delay, is the pth transmitting antenna array element With the qth receiving antenna element The channel impulse response between .

[0023] is the line-of-sight LoS component Non-line-of-sight NLoS component The specific formula is:

[0024]

[0025]

[0026] Among them, K R (t) is the Rice factor at time t; []T is the transpose; For different frequency bands Directivity diagram corresponding to vertical polarization; For different frequency bands Directivity diagram corresponding to vertical polarization; For the tth moment to The elevation arrival angle corresponding to the LoS path; For the tth moment to The azimuth arrival angle corresponding to the LoS path; is the first transmitting antenna array element; is the first receiving antenna array element; V is vertical polarization; H is horizontal polarization; E is the elevation angle; L is the LoS path; A is the antenna unit; and All are random phases that obey the uniform distribution of (0,2π]; For different frequency bands Directivity diagram corresponding to horizontal polarization; For different frequency bands Directivity diagram corresponding to horizontal polarization; F r is the Faraday rotation matrix, ψ l,m is the Faraday rotation angle, ψ l,m =108 / f c 2 ; For the tth moment to The pitch departure angle corresponding to the LoS path; For the tth moment to The azimuth departure angle corresponding to the LoS path; is the delay of the LoS path at time t, For the tth moment and The vector distance between them, c is the speed of light; N qp (t) is the total number of paths at time t; M n (t) is the total number of sub-paths at the tth moment; is the inverse of the cross-polarization power ratio; μ is the joint polarization imbalance; δ() is the antenna element spacing; and At time t, they are to The azimuth departure angle and elevation departure angle corresponding to the mth sub-path; and At time t, they are to The azimuth departure angle and elevation departure angle corresponding to the mth sub-path; Under NLoS conditions arrive The power of the mth subpath in the nth path; For the tth moment and The delay of the mth sub-path in the nth path.

[0027] Furthermore, in step S3, reconstructing the digital twin online channel of the static scene includes the following sub-steps:

[0028] S301. Reconstruct the 3D static communication scene through Unity software based on the channel parameters.

[0029] S302, import the reconstructed scene into RT (Ray Tracing) software, and simulate the channel characteristics based on GO (Geometric Optics) and UTD (Uniform Theory of Diffraction). The calculation formula of the channel impulse response is:

[0030]

[0031] Where h(t) is the channel impulse response; N(t) is the total number of paths; A i is the amplitude of the ith path, P i is the power of the ith path; f D is the Doppler shift, B is the total number of ray action points, is the unit direction vector of the ray from the b-th action point to the b+1-th action point, is the velocity of the b-th action point, is the velocity of the b-1th action point, v is the propagation speed of the ray; τ i (t) is the delay of the ith path at time t, d i is the length of the i-th path; ψ is the random phase.

[0032] S303. Use the channel detection equipment to perform channel measurement at the set position and frequency band to obtain actual channel measurement data, compare the difference between the channel characteristic simulation results and the actual channel measurement data, calibrate the electromagnetic coefficients used in the simulation, and obtain the digital twin online channel of the reconstructed static scene.

[0033] Furthermore, in step S4, obtaining dynamic perception data includes the following sub-steps:

[0034] S401. Deploy various types of sensors in the target communication scenario to form an environment perception network.

[0035] S402. Collect environmental data in real time through the environmental perception network. The data includes the location information, speed, acceleration and size of the moving objects. Clean, fuse and preliminarily analyze the data to extract dynamic perception data and track its dynamic changes.

[0036] Furthermore, the sensors include cameras, lidars, millimeter wave radars, ultrasonic sensors, and synthetic aperture radar systems.

[0037] Cameras are used to capture visual information, LiDAR and millimeter-wave radar are used to provide distance and velocity measurements, and SAR systems are used to generate ground-level imaging data.

[0038] Furthermore, in step S5, the new real-time twin channel is obtained including the following contents:

[0039] The impulse response h of the new real-time twin channel qp (t,τ) is the static component and dynamic component The specific expression is:

[0040]

[0041] For static scenes, the RT geometric optics method is used to generate the LoS path and multipath components of static scatterers in the physical features. The specific expression is:

[0042]

[0043] in, and They are the LoS component and NLOS component of the static CIR (Channel Impulse Response), and All of them are random phases that obey the uniform distribution of (0,2π], K qp (t) is the total number of rays, is the inverse of the cross-polarization power ratio of the kth ray; P qp,k (t) is the pth transmitting antenna array element at the tth time The power between the kth ray, τ qp,k (t) is the pth transmitting antenna element at the tth time and the time delay between the kth ray.

[0044] For objects in dynamic scenes, the dynamic cluster parameters are extracted from the measured data, and GBSM is used to simulate the dynamic scattering channel. The CIR of the dynamic component is modeled using 6GPCM, and only the NLOS path is considered. The specific expression is:

[0045]

[0046] Furthermore, in step S6, building a visualization platform for the digital twin online channel includes the following sub-steps:

[0047] S601. Use Blender software to perform three-dimensional modeling on the physical features of the static scene and import them into the digital twin online channel of the reconstructed static scene.

[0048] S602. Use the Unity engine to integrate the dynamic perception data and the digital twin online channel of the constructed static scene into the scene, and realize the real motion trajectory and behavior simulation of the moving object through programming and physics engine.

[0049] S603. Integrate static scenes and dynamic scenes, build a digital twin channel visualization platform, output channel characteristics, and update them in real time based on user interactions and scene changes.

[0050] Furthermore, the present invention also proposes an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the 6G digital twin online channel modeling method when executing the computer program.

[0051] Furthermore, the present invention also proposes a computer-readable storage medium, which stores a computer program, and the computer program executes the 6G digital twin online channel modeling method when run by a processor.

[0052] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0053] 1. Reduce pilot overhead: By accessing the database or based on position, direction and antenna parameters, DTOCM can preload basic CSI (Channel State Information), allowing the device to obtain channel parameters when passing through continuous 3D space, significantly reducing pilot overhead and improving 6G network performance.

[0054] 2. Real-time CSI provision and prediction: DTOCM has strong real-time channel information extraction and prediction capabilities, adapting to dynamic network conditions. This capability ensures that the 6G communication network remains responsive to environmental changes, helps calibrate the performance between channel model simulation and actual environmental conditions, and thus improves the reliability and robustness of network performance.

[0055] 3. Channel Information Visualization: DTOCM provides visualization of channel information changes, predicting and highlighting changes in the communication environment. It allows display of node location and motion status, providing real-time monitoring of channel characteristics and other relevant information. This visualization enhances environmental perception and decision-making capabilities in future 6G networks.

[0056] 4. Optimizing network performance: DTOCM can simulate scenarios in a virtual environment, providing perception and feedback for optimizing real-world communication network deployment. This enables comprehensive testing in a controlled virtual environment, allowing for rapid iteration and improvement. By leveraging the information gained from these simulations, future 6G networks can be effectively optimized to achieve a balance between performance and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is an overall implementation flow chart of the present invention.

[0058] Figure 2 It is the digital twin online channel modeling flow chart of the present invention.

[0059] Figure 3 It is a schematic diagram of classification and matching of communication scenarios of the present invention.

[0060] Figure 4 It is the RT simulation result of the static scene of the embodiment of the present invention.

[0061] Figure 5 This is a distribution diagram of the environment perception sensors according to an embodiment of the present invention.

[0062] Figure 6 It is a digital twin online channel modeling visualization platform of an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0064] To achieve the above objectives, the present invention proposes a 6G digital twin online channel modeling method, such as Figure 1 As shown, the specific steps are as follows:

[0065] S1. The current 5G standardized channel model has a relatively rough division of communication scenarios and cannot fully describe all communication scenarios of the future 6G network under the vision of full frequency band, full coverage and full application. In order to build a real-time and accurate digital twin online channel model applicable to all frequency bands and scenarios, it is necessary to classify 6G wireless communication scenarios in detail and explore the channel characteristics of different communication scenarios.

[0066] Based on the information obtained from the environmental data source, a machine learning model (such as a neural network) is used to perform online recognition of communication scenarios; the specific contents are as follows:

[0067] S101. Environmental data sources include electronic maps, remote sensing images and point cloud data; electronic maps provide information about terrain, buildings and roads, which helps to understand the geographical and urban environment; high-resolution remote sensing images are obtained through satellites or drones to identify terrain features and environmental changes; three-dimensional point cloud data generated by lidar or other sensors provide an accurate description of the three-dimensional spatial structure of the environment.

[0068] S102, such as Figure 2 As shown, the information in the environmental data source is extracted and processed through a machine learning model, and the communication scenarios are classified into different categories, such as aviation communication, space communication, land communication, marine communication scenarios and communication conversion scenarios. These categories are further subdivided into more specific environments, such as satellites, space stations, aircraft, drones, indoors, cities, sea islands, indoor-outdoor, outdoor-indoor, etc. The neural network-based classification network has a more detailed understanding of various environments.

[0069] S2. According to the identified communication scenario, match the channel parameters of the 6G universal channel model and map them to the communication scenario, so that accurate multi-scenario channel measurement can be performed and compatibility with the 6G channel model is ensured; the specific contents are as follows:

[0070] The channel parameter matrix of 6GPCM (6G Pervasive Channel Model) is expressed as:

[0071] H=[PL·SH·BL·WE·AL] 1 / 2 ·H s

[0072] Among them, H is the channel matrix, PL is the path loss, SH is the shadow fading, BL is the blocking effect, AL is the atmospheric absorption loss, WE is the weather effect loss, and H s Small scale fading.

[0073] Small scale fading H s It is expressed as:

[0074]

[0075] Among them, M T is the number of antenna elements in the transmitting antenna array, M R is the number of antenna elements in the receiving antenna array, f c is the carrier frequency, R is the receiving end, T is the transmitting end, τ is the delay, is the pth transmitting antenna array element With the qth receiving antenna element The channel impulse response between .

[0076] is the line-of-sight LoS component Non-line-of-sight NLoS component The specific formula is:

[0077]

[0078]

[0079] Among them, K R (t) is the Rice factor at time t, that is, the ratio of the power of the LoS component to the power of the NLoS component, [] T For transposition, For different frequency bands The directivity diagram corresponding to vertical polarization is For different frequency bands The directivity diagram corresponding to vertical polarization is For the tth moment to The elevation arrival angle corresponding to the LoS path is For the tth moment to The azimuth arrival angle corresponding to the LoS path is is the first transmitting antenna array element, is the first receiving antenna array element, V is the vertical polarization, H is the horizontal polarization, E is the elevation angle, L is the LoS path, A is the antenna unit, and All of them are random phases that obey the uniform distribution of (0,2π], For different frequency bands The directivity diagram corresponding to horizontal polarization is For different frequency bands The directivity pattern corresponding to horizontal polarization, F r is the Faraday rotation matrix, ψ l,m is the Faraday rotation angle, ψ l,m =108 / f c 2 , where the Faraday rotation angle is calculated c The unit is GHz, For the tth moment to The pitch departure angle corresponding to the LoS path, For the tth moment to The azimuth departure angle corresponding to the LoS path, is the delay of the LoS path at time t, For the tth moment and The vector distance between them, c is the speed of light, N qp (t) is the total number of paths at time t, M n (t) is the total number of sub-paths at the tth moment, is the inverse of the cross-polarization power ratio, μ is the joint polarization imbalance, δ() is the antenna element spacing, and At time t, they are to The azimuth departure angle and elevation departure angle corresponding to the mth subpath, and At time t, they are to The azimuth departure angle and elevation departure angle corresponding to the mth subpath, Under NLoS conditions arrive The power of the mth subpath in the nth path, For the tth moment and The delay of the mth sub-path in the nth path.

[0080] S3. Reconstruct the digital twin online channel of the static scene based on the acquired channel parameters; the specific contents are as follows:

[0081] S301, such as Figure 3 As shown in Figure 1, based on the channel parameters, the 3D static communication scene is reconstructed through the Unity software. The static scene RT (Ray Tracing) simulation in this example is as follows: Figure 4As shown, the dynamic environment perception in the wireless valley outdoor example is demonstrated. Figure 4 (a) and Figure 4 (b) describes the test site and sensor settings respectively.

[0082] S302, import the reconstructed scene into RT software, and simulate the channel characteristics based on GO (Geometric Optics) and UTD (Uniform Theory of Diffraction), which can accurately consider various propagation mechanisms of electromagnetic waves, such as direct radiation, reflection, transmission, diffraction, scattering, etc. Figure 5 As shown, the multipath parameters can be obtained in the RT twin channel according to the location of the transmitting and receiving ends and the environment;

[0083] The channel impulse response is calculated as:

[0084]

[0085] Where h(t) is the channel impulse response; N(t) is the total number of paths; A i is the amplitude of the ith path, P i is the power of the ith path; f D is the Doppler shift, B represents the total number of ray action points, is the unit direction vector of the ray from the b-th action point to the b+1-th action point, is the velocity of the b-th action point, is the transmitting end moving speed, is the velocity of the b-1th action point, v is the propagation speed of the ray; τ i (t) is the delay of the ith path at time t, d i is the length of the i-th path; ψ is the random phase.

[0086] S303. Use channel detection equipment (such as vector network analyzer, spectrum analyzer, etc.) to perform channel measurement at a specific location and frequency band to obtain actual channel measurement data, compare the differences between the channel characteristic simulation results and the actual channel measurement data (such as mean square error), calibrate the electromagnetic coefficients used in the simulation, improve the accuracy of RT channel reconstruction, and obtain the digital twin online channel of the reconstructed static scene, providing accurate model support for subsequent channel analysis and system design.

[0087] In addition to the twin environment, the digital twin layer also contains twin data. Generative models are used to create twins of different scenarios from measurement data to provide a priori support for channel measurement and interference suppression.

[0088] S4. Dynamically perceive the physical environment and obtain dynamic perception data; the specific content is:

[0089] S401. Deploy multiple types of sensors in the target communication scenario, such as cameras, lidar, millimeter-wave radar, ultrasonic sensors, and synthetic aperture radar (SAR) systems, which together form a high-precision, multi-modal environmental perception network. Cameras can capture visual information, lidar and millimeter-wave radar can provide high-precision distance and speed measurements, and SAR systems are used to generate high-resolution ground imaging data.

[0090] S402: Collect environmental data in real time through the environmental perception network. This data includes the location information, speed, acceleration, size, etc. of moving objects. Clean, fuse and preliminarily analyze the data to extract environmental change information that is meaningful for channel modeling, i.e., dynamic perception data, such as identifying moving vehicles, pedestrians, buildings, trees and other physical objects, and tracking their dynamic changes.

[0091] S5. Combine the static-dynamic hybrid channel modeling algorithm and dynamic perception data to dynamically update the digital twin online channel of the reconstructed static scene to obtain a new real-time twin channel, including the physical characteristics of the static scene and the objects of the dynamic scene; the specific content is:

[0092] The impulse response h of the new real-time twin channel qp (t,τ) is the static component and dynamic component The specific expression is:

[0093]

[0094] For static scenes, the RT geometric optics method is used to generate the LoS paths and multipath components of static scatterers in physical features such as buildings, trees, and the ground. The specific expressions are:

[0095]

[0096]

[0097] in, and They are the LoS component and NLOS component of the static CIR (Channel Impulse Response), and All of them are random phases that obey the uniform distribution of (0,2π], K qp (t) is the total number of rays, is the inverse of the cross-polarization power ratio of the kth ray; P qp,k (t) is the pth transmitting antenna element at the tth time The power between the kth ray, τ qp,k (t) is the pth transmitting antenna element at the tth time The time delay between the ray and the k-th ray can be obtained through RT simulation.

[0098] For objects in dynamic scenes, the dynamic cluster parameters are extracted from the measured data, and GBSM is used to simulate the dynamic scattering channel. For the CIR of the dynamic component, 6GPCM is mainly used for modeling, and only the NLOS path is considered. The specific expression is:

[0099]

[0100] S6. Build a visualization platform for the digital twin online channel. The specific contents are as follows:

[0101] S601. Use Blender software to perform detailed 3D modeling of the physical features of the static scene, such as terrain, buildings, roads, etc., and import them into the digital twin online channel of the reconstructed static scene to ensure that the modeling results are highly consistent with the actual environment.

[0102] S602. Use the Unity engine to integrate the dynamic perception data and the digital twin online channel of the constructed static scene into the scene. Through programming and the physical engine, the real motion trajectory and behavior simulation of the moving object are realized, and the changes in the channel characteristics are dynamically displayed.

[0103] S603, integrate static scenes and dynamic scenes, build a digital twin channel visualization platform, the system outputs detailed channel characteristics, including CSI, path loss and other data, and updates in real time based on user interaction and scene changes.

[0104] In this example, measured data from a location in Jiangning District, Nanjing City is mainly used. The 3D images captured by drones are then reconstructed in scale using Blender software, and textures are extracted from the aerial view, while ensuring that the electromagnetic properties of the 3D scene are aligned with the actual situation. On this basis, the output generated by Blender is given electromagnetic properties and integrated into Wireless Insite for accurate ray tracing simulation to generate accurate channel state information. To improve the accuracy of the predictions, these simulation results are calibrated with the measured data.

[0105] like Figure 6As shown in the figure, the visualization of channel characteristics is presented in the visualization module in the form of heat maps. The interface supports the aesthetic design of custom heat maps, the display of emission positions, and the selection of channel characteristics, thereby enhancing the analysis and understanding of channel characteristics. In addition, an integrated 3D-EADTOCMP platform was developed through Unity to integrate static models and channel characteristics in an interactive user interface, allowing users to intuitively explore and analyze 3D environments and their channel characteristics. In the environmental perception part of dynamic scene updates, the video stream is divided into two parallel processing parts: one is to detect the category and two-dimensional coordinates of objects using the YOLO (You-Only-Look-Once) network, and the other is to obtain depth information through the depth camera program. By synchronizing the timestamps, the 3D coordinate data containing depth indicators and object classifications are transmitted to the Unity platform, so that objects can be accurately instantiated in the virtual environment to ensure their correct spatial position.

[0106] The system outputs detailed channel characteristics, including CSI, path loss and other relevant data, and dynamically updates according to user interaction and scene changes. At the same time, RT and GBSM are used to accurately predict and simulate wireless channel changes to achieve real-time updates. Figure 6 As can be seen from (a), it can perceive environmental changes in real time and detect pedestrians. Figure 6 (b) shows the corresponding digital twin environment, which can be updated in real time to reflect the movement of pedestrians. Figure 6 (c) shows the depth analysis through a depth map, with different colors representing distance and depth. Figure 6 (d) shows the corresponding channel characteristics, including PSDs (Power Spectral Densities) in azimuth and elevation, providing detailed information on the wireless channel characteristics.

[0107] The embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. It should be noted that when the processor executes the computer program, the specific steps of the method provided in the embodiment of the present invention are corresponding to the specific steps of the method provided in the embodiment of the present invention, and the processor has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided in the embodiment of the present invention.

[0108] The embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program. It should be noted that when the computer program is executed by the processor, it corresponds to the specific steps of the method provided in the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided in the embodiment of the present invention.

[0109] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A 6G digital twin online channel modeling method, characterized in that: include: S1. Based on the information obtained from the environmental data source, the machine learning model is used to perform online recognition of communication scenarios; S2. According to the identified communication scenario, match the channel parameters of the 6G universal channel model and map them to the communication scenario; S3. Reconstruct the digital twin online channel of the static scene based on the acquired channel parameters; S4, dynamically perceive the physical environment and obtain dynamic perception data; S5. Combining the static-dynamic hybrid channel modeling algorithm and dynamic perception data, dynamically update the digital twin online channel of the reconstructed static scene to obtain a new real-time twin channel, including the physical features of the static scene and the objects of the dynamic scene; S6. Build a visualization platform for the digital twin online channel.

2. The 6G digital twin online channel modeling method according to claim 1 is characterized in that: In step S1, online identification of communication scenarios includes the following sub-steps: S101. Environmental data sources include electronic maps, remote sensing images and point cloud data; Electronic maps provide information about terrain, buildings and roads; remote sensing images are acquired through satellites or drones to identify terrain features and environmental changes; three-dimensional point cloud data is generated by lidar or other sensors to provide the three-dimensional spatial structure of the environment; S102. Extract and process information from environmental data sources through machine learning models, classify communication scenarios into different categories, and further subdivide these categories.

3. The 6G digital twin online channel modeling method according to claim 1 is characterized in that: In step S2, matching channel parameters includes the following: The channel parameter matrix of the 6G universal channel model is expressed as: H=[PL·SH·BL·WE·AL] 1 / 2 ·H s Among them, H is the channel matrix, PL is the path loss, SH is the shadow fading, BL is the blocking effect, AL is the atmospheric absorption loss, WE is the weather effect loss, and H s It is a small-scale fading; Small scale fading H s It is expressed as: Among them, M T is the number of antenna elements in the transmitting antenna array, M R is the number of antenna elements in the receiving antenna array, f c is the carrier frequency, R is the receiving end, T is the transmitting end, τ is the delay, is the pth transmitting antenna array element With the qth receiving antenna element The channel impulse response between is the line-of-sight LoS component Non-line-of-sight NLoS component The specific formula is: Among them, K R (t) is the Rice factor at the tth moment; [] T is transposed; F q,fc,V () is in different frequency bands Directivity diagram corresponding to vertical polarization; For different frequency bands Directivity diagram corresponding to vertical polarization; For the tth moment to The elevation arrival angle corresponding to the LoS path; For the tth moment to The azimuth arrival angle corresponding to the LoS path; is the first transmitting antenna array element; is the first receiving antenna array element; V is vertical polarization; H is horizontal polarization; E is the elevation angle; L is the LoS path; A is the antenna unit; and All are random phases that obey the uniform distribution of (0,2π]; For different frequency bands Directivity diagram corresponding to horizontal polarization; For different frequency bands Directivity diagram corresponding to horizontal polarization; F r is the Faraday rotation matrix, ψ l,m is the Faraday rotation angle, ψ l,m =108 / f c 2 ; For the tth moment to The pitch departure angle corresponding to the LoS path; For the tth moment to The azimuth departure angle corresponding to the LoS path; is the delay of the LoS path at time t, For the tth moment and The vector distance between them, c is the speed of light; N qp (t) is the total number of paths at time t; M n (t) is the total number of sub-paths at the tth moment; is the inverse of the cross-polarization power ratio; μ is the joint polarization imbalance; δ( ) is the antenna element spacing; and At time t, they are to The azimuth departure angle and elevation departure angle corresponding to the mth sub-path; and At time t, they are to The azimuth departure angle and elevation departure angle corresponding to the mth sub-path; Under NLoS conditions arrive The power of the mth subpath in the nth path; For the tth moment and The delay of the mth sub-path in the nth path.

4. The 6G digital twin online channel modeling method according to claim 1 is characterized in that: In step S3, the reconstruction of the digital twin online channel of the static scene includes the following sub-steps: S301, reconstructing the 3D static communication scene through unity software based on the channel parameters; S302, import the reconstructed scene into the ray tracing software, and simulate the channel characteristics based on geometric optics and consistent diffraction theory. The calculation formula of the channel impulse response is: Where h(t) is the channel impulse response; N(t) is the total number of paths; A i is the amplitude of the ith path, P i is the power of the ith path; f c is the carrier frequency; δ( ) is the antenna array element spacing; f D is the Doppler shift, B is the total number of ray action points, is the unit direction vector of the ray from the b-th action point to the b+1-th action point, is the velocity of the b-th action point, is the velocity of the b-1th action point, v is the propagation speed of the ray; τ i (t) is the delay of the ith path at time t, d i is the length of the i-th path; ψ is the random phase; S303. Use the channel detection equipment to perform channel measurement at the set position and frequency band to obtain actual channel measurement data, compare the difference between the channel characteristic simulation results and the actual channel measurement data, calibrate the electromagnetic coefficients used in the simulation, and obtain the digital twin online channel of the reconstructed static scene.

5. The 6G digital twin online channel modeling method according to claim 1 is characterized in that: In step S4, obtaining dynamic perception data includes the following sub-steps: S401, deploying various types of sensors in the target communication scenario to form an environment perception network; S402. Collect environmental data in real time through the environmental perception network. The data includes the location information, speed, acceleration and size of the moving objects. Clean, fuse and preliminarily analyze the data to extract dynamic perception data and track its dynamic changes.

6. The 6G digital twin online channel modeling method according to claim 5 is characterized in that: Sensors include cameras, lidar, millimeter-wave radar, ultrasonic sensors, and synthetic aperture radar systems; Cameras are used to capture visual information, LiDAR and millimeter-wave radar are used to provide distance and velocity measurements, and SAR systems are used to generate ground-level imaging data.

7. The 6G digital twin online channel modeling method according to claim 1 is characterized in that: In step S5, a new real-time twin channel is obtained including the following contents: The impulse response h of the new real-time twin channel qp (t,τ) is the static component and dynamic component The specific expression is: Among them, K R is the Rice factor, which is the ratio of the power of the LoS component to the power of the NLoS component; For static scenes, the incident ray tracing geometric optics method is used to generate the LoS path and multipath components of static scatterers in the physical features. The specific expression is: in, and are the LoS component and NLOS component of the static channel impulse response, respectively, [] T For transposition, For different frequency bands The directivity diagram corresponding to vertical polarization is For different frequency bands The directivity diagram corresponding to vertical polarization is is the delay of the LoS path at time t, For the tth moment to The elevation arrival angle corresponding to the LoS path is For the tth moment to The azimuth arrival angle corresponding to the LoS path is and All of them are random phases that obey the uniform distribution of (0,2π], For the tth moment to The pitch departure angle corresponding to the LoS path, is the first transmitting antenna array element, is the first receiving antenna element, For the tth moment to The azimuth departure angle corresponding to the LoS path, μ is the joint polarization imbalance, K qp (t) is the total number of rays, is the inverse of the cross-polarization power ratio of the kth ray; P qp,k (t) is the pth transmitting antenna array element at the tth time The power between the kth ray, τ qp,k (t) is the pth transmitting antenna array element at the tth time and the time delay between the kth ray; For objects in dynamic scenes, the dynamic cluster parameters are extracted from the measured data, and GBSM is used to simulate the dynamic scattering channel. The channel impulse response of the dynamic component is modeled using 6GPCM, and only the NLOS path is considered. The specific expression is:

8. The 6G digital twin online channel modeling method according to claim 1 is characterized in that: In step S6, building a visualization platform for the digital twin online channel includes the following sub-steps: S601. Using Blender software, three-dimensionally model the physical features of the static scene and import them into the digital twin online channel of the reconstructed static scene; S602. Use the Unity engine to integrate the dynamic perception data and the digital twin online channel of the constructed static scene into the scene, and realize the real motion trajectory and behavior simulation of the moving object through programming and physical engine; S603. Integrate static scenes and dynamic scenes, build a digital twin channel visualization platform, output channel characteristics, and update them in real time based on user interactions and scene changes.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the 6G digital twin online channel modeling method described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the 6G digital twin online channel modeling method described in any one of claims 1 to 8 is executed.

Images