Urban low-altitude channel and line-of-sight propagation simulation system fusing BIM and GIS data
By integrating BIM and GIS data, a high-fidelity 3D scene was constructed and deterministic ray tracing was performed, which solved the problems of low accuracy and long computation time in urban low-altitude channel simulation, and realized real-time response and efficient simulation analysis of dynamic environment.
Patent Information
- Application Number
- CN202610161964.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-02-04
- Publication Date
- 2026-05-15
AI Technical Summary
Existing technologies cannot accurately characterize the complex reflection, diffraction, and shading effects of urban low-altitude channels, cannot support real-time responses to dynamic environmental changes, and consume too much computation time, making it difficult to meet the needs of rapid planning and evaluation.
By integrating BIM and GIS data, a high-fidelity 3D scene map is constructed. A deterministic ray tracing algorithm is adopted, combined with the mirror method and the bouncing ray algorithm, and parallel computing is performed through a CPU and GPU collaborative architecture to achieve channel simulation.
It achieves an order-of-magnitude improvement in channel prediction accuracy, supports real-time response in dynamic environments, is suitable for high-reliability control and low-latency data link planning of UAVs, expands the application scope of the system, and provides a complete simulation verification environment.
Smart Images
Figure CN122052948A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of urban channel simulation, specifically to an urban low-altitude channel and line-of-sight propagation simulation system that integrates BIM and GIS data. Background Technology
[0002] In recent years, low-altitude aircraft such as unmanned aerial vehicles (UAVs) and electric vertical takeoff and landing (eVTOL) aircraft have flourished globally, finding widespread application in urban air traffic, emergency rescue, logistics distribution, and environmental monitoring. Low-altitude communication networks are the core and safety guarantee for the operation of these aircraft. However, the signal propagation environment in urban low-altitude channels is extremely complex. Traditional channel simulation techniques mainly utilize empirical or semi-empirical models such as the Okumura-Hata model or the COST 231 model, simulating channels based on macroscopic statistical data or using ray tracing simulation tools. However, these common types of channel simulation suffer from the following problems in practical applications:
[0003] First, it cannot accurately depict the complex reflection, diffraction, and shading effects caused by the precise shape, layout, and materials of buildings in a specific urban area, resulting in low prediction accuracy and poor simulation performance.
[0004] Secondly, the real urban environment is dynamic and changing. Factors such as moving vehicles, temporary buildings, and changing vegetation can directly or indirectly affect the propagation of radio waves. Most existing simulation systems are based on static scenes and cannot effectively support real-time or near-real-time response to dynamic environmental changes. They are also difficult to perform dynamic closed-loop simulation with UAV flight control systems.
[0005] Finally, when simulating a large number of application scenarios involving the dynamic trajectories of drones, the computation time of existing technologies is too long, making it difficult to meet the needs of rapid planning and evaluation. Summary of the Invention
[0006] The purpose of this invention is to provide an urban low-altitude channel and line-of-sight propagation simulation system that integrates BIM and GIS data, in order to solve the problems mentioned in the background art, such as low accuracy, inability to effectively support real-time or near-real-time response to dynamic environmental changes, and excessive calculation time, which make it difficult to meet the needs of rapid planning and evaluation.
[0007] Therefore, the present invention provides an urban low-altitude channel and line-of-sight propagation simulation system that integrates BIM and GIS data, comprising:
[0008] The data fusion processing module is used to receive BIM and GIS data, and to preprocess the received data by performing coordinate system I, semantic mapping and physical parameter assignment, and output a unified scene map.
[0009] A 3D scene construction module is used to receive a scene graph and convert it into a highly optimized spatial index structure that supports sub-millisecond ray casting.
[0010] A deterministic propagation calculation module is used to calculate the propagation path of a signal based on a spatial index structure, combined with the mirror method and the bouncing ray algorithm.
[0011] A hybrid channel modeling module is used to generate a comprehensive description of channel characteristics based on the calculated effective path information.
[0012] A parallel computing and scheduling module, which is used to perform parallel ray tracing using a collaborative architecture of CPU and GPU;
[0013] A visualization analysis module is used to provide human-computer interaction and results.
[0014] Preferably, the specific architecture of the data fusion processing module includes a data interface layer, a core processing layer, and an output layer;
[0015] The data interface layer is used for reading, parsing, and temporarily storing raw data, including BIM data interface, GIS data interface, and temporary storage. The BIM data interface is used to parse industrial basic files and extract the geometric characteristics, semantics, and material properties of building components; the GIS data interface is used to parse CityGML, Shapefile, and raster digital elevation models.
[0016] The core processing layer is used to unify the data obtained by the data interface layer in a three-dimensional coordinate system. In the unified coordinate system, it repairs, simplifies and aligns the geometric model with semantic information, and assigns precise electromagnetic physical parameters to each geometric patch. The core processing layer has a built-in material electromagnetic parameter database divided by frequency band.
[0017] The output layer is used to generate and output a unified scene graph for use by subsequent ray tracing modules.
[0018] Preferably, the specific data processing flow of the data fusion processing module is as follows:
[0019] First, data input parsing: For BIM files input from the BIM data interface and GIS datasets input from the GIS data interface, the corresponding parser is called to parse the input data;
[0020] Next, a unified coordinate system is established. At least three easily identifiable and highly accurate feature points with the same name are selected on the analyzed BIM model and GIS map, and the coordinates of the feature points in the BIM local coordinate system are obtained. and coordinates in the GIS projected coordinate system Using the seven-parameter Helmert transform, the expression is: In the formula is the translation vector, s is the scale factor used to compensate for scale differences in measurements, R is the rotation matrix of the transformation model, representing rotation along three axes, and the seven parameters include three translation parameters, one scale parameter, and three rotation parameters, which are obtained by least squares adjustment.
[0021] Using the calculated seven parameters, coordinate transformation is performed on all geometric vertices in the BIM model to obtain data spatially aligned with GIS data.
[0022] Then, semantic mapping associates the refined component information of BIM with the macro-geographic entity classification of GIS, assigning a unified semantic label to provide semantic labels for subsequent physical calculations.
[0023] Next, physical property values are assigned. For regions with semantic tags, the electromagnetic parameter database of materials is queried according to the simulation center frequency set by the user to obtain the corresponding electromagnetic parameters, including the phase velocity of the influencing wave, the conductivity of the material and the vacuum dielectric constant. The obtained electromagnetic parameters are then attached as data.
[0024] Finally, the data from different sources processed in the above steps are organized into a logically unified scene graph. The root node of the scene graph represents the entire simulation region, and each leaf node contains one or more sets of triangular faces. Each face includes vertex coordinates, normal vectors, semantic labels, and electromagnetic physical properties. The constructed scene graph is then output.
[0025] Preferably, the 3D scene construction module includes a scene management layer, a core acceleration structure construction layer, and a serialization output layer;
[0026] The scene management layer is used to receive a unified scene graph, perform overall scene analysis, calculate the global bounding box in the world coordinate system, and preset construction parameters according to the scene scale and complexity.
[0027] The core acceleration structure building layer is used to perform recursive and uniform octree partitioning of the entire scene space from top to bottom on a macroscopic level, quickly indexing spatial locations and eliminating large empty areas. On a microscopic level, it constructs a locally optimized BVH for the set of triangular faces managed by each octree leaf node containing geometry from bottom to top.
[0028] The serialization output layer is used to flatten and serialize the complex pointer-linked structure that is built and exists in main memory, converting it into a linear memory layout suitable for efficient GPU access. The output includes geometric data, physical attribute data, and most importantly, hybrid accelerated structure data that can be directly traversed by GPU kernel functions.
[0029] Preferably, the deterministic propagation calculation module includes
[0030] The simulation control unit is used to parse the simulation tasks input by the user, initialize the global simulation state, and coordinate the execution of the simulation process by each unit.
[0031] The ray emitter unit is used to generate an initial set of rays originating from the emission source according to the selected mode. The modes are divided into the mirror mode and the SBR mode. The mirror mode is for specified low-order reflections and does not emit wide-angle rays. Based on the principles of geometric optics, it calculates the mirror points to accurately solve the specific specular reflection path. The SBR mode is used to capture high-order reflections, diffuse reflections and diffraction paths. It uses an algorithm based on geodesic dome or uniform solid angle sampling to emit discrete rays into the entire upper half space.
[0032] A ray-based scene interaction engine is used to drive the propagation, interaction, and evolution of each ray in the scene.
[0033] The physical interaction calculation unit includes a reflection, transmission calculator and a diffraction calculator, which are used to calculate the reflection coefficient and transmission coefficient of rays at the interface of media and to handle the diffraction phenomenon of rays at the sharp edges of buildings.
[0034] A ray tracing unit is used to capture propagating rays and record their geometric and electromagnetic information.
[0035] The ray period management unit is used to track the status of each ray and decides to terminate the ray when the ray energy is lower than a preset threshold, reaches the maximum reflection or diffraction order, or leaves the simulation area.
[0036] Preferably, the reflection and transmission calculator uses Fresnel's equations for calculation, and the expression for calculating the reflection coefficient is: In the formula Angle of incidence and The reflection coefficient is the polarization of the incident angle. By Snell's Law The complex refractive index is calculated. , Let be the complex permittivity; the expression for calculating the transmission coefficient is: In the formula This refers to the impedance characteristics of the medium for the propagation of electromagnetic waves.
[0037] The diffraction calculator uses uniform diffraction theory to determine a diffraction point on the edge of the wedge. Uniform diffraction theory provides a set of complex diffraction coefficients D, and the diffraction field strength is... In the formula For a strong incident field, For diffusion factor, This represents the distance from the diffraction point to the field point.
[0038] Preferably, the hybrid channel modeling module includes
[0039] The path data processing unit receives, verifies, and preprocesses raw path data from the ray tracing unit, including complex amplitude. Propagation delay Departure angle, arrival angle, and polarization information;
[0040] The large-scale fading calculation unit is used to coherently superimpose the complex amplitudes of all effective paths to obtain the total received electric field. Then calculate the total received power. and path loss In the formula This refers to the transmission power.
[0041] Small-scale fading calculation units are used to calculate the merged paths according to their time delays. Sort the paths and calculate the power of each path;
[0042] The Doppler spectrum calculation unit is used to calculate the Doppler frequency shift for each path of the mobile terminal and synthesize the Doppler power spectrum of the channel.
[0043] That is, for a path, the angle of arrival and the terminal velocity vector are known. The Doppler frequency shift of this path is In the formula For carrier frequency, At the speed of light, For wavelength, The power of each path is represented by a unit vector from the terminal toward the incident direction. Assigned to the corresponding Doppler frequency shift Then, the frequency domain is binned and the power is superimposed to obtain the discrete Doppler power spectrum. ;
[0044] The time-varying channel impulse response unit is used to recalculate the delay, amplitude, and phase of each path at each sampling time based on the terminal position at that time for a given terminal trajectory. Then, these paths are superimposed to obtain the channel impulse response sequence at that time, which is used for time-domain simulation.
[0045] The channel parameter output and visualization unit outputs the calculated channel parameters and model in a standard format and provides them to the visualization module for graphical display.
[0046] Preferably, the parallel computing and scheduling module includes:
[0047] The task scheduling and resource management layer runs on the CPU main thread. It is used to parse simulation tasks, perform macro-level task graph decomposition, and monitor CPU / GPU utilization, memory usage, and PCIe bandwidth in real time to provide decision-making basis for dynamic scheduling.
[0048] The CPU parallel execution layer is used to execute tasks suitable for CPU parallelization.
[0049] The GPU compute and memory management layer is used for the allocation, release, and lifecycle management of GPU memory, as well as configuring and launching GPU kernel functions and optimizing the dimensions of mesh and thread blocks.
[0050] The data transmission and synchronization layer is used for data synchronization and transmission.
[0051] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements all the functions of the above-mentioned urban low-altitude channel and line-of-sight propagation simulation system that integrates BIM and GIS data.
[0052] Meanwhile, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements all the functions of the above-mentioned system.
[0053] The present invention proposes an urban low-altitude channel and line-of-sight propagation simulation system that integrates BIM and GIS data, the advantages of which are as follows:
[0054] This system can integrate the detailed building component information provided by BIM with the macro-geographic information provided by GIS under a unified coordinate system and assign precise electromagnetic physical parameters. It not only achieves the alignment of geometric and semantic information, but also constructs a high-fidelity 3D scene map containing material properties and electromagnetic parameters, builds a digital twin environment of the city with complete physical properties, and uses a deterministic ray tracing algorithm based on physical optics for simulation. It can accurately characterize the reflection, transmission, diffraction and shading effects caused by the fine structure of building facades, so that the channel prediction accuracy is improved by orders of magnitude compared with traditional models, which can meet the planning requirements of high-reliability control and low-latency data links for UAVs.
[0055] Through modular design, it can access external signals such as UAV flight control data and traffic flow information in real time, dynamically update the target position and environmental status in the simulation scenario, and trigger incremental calculations. This is crucial for simulating real-time channel changes in UAV maneuvering flight, assessing the impact of dynamic obstacles, and conducting joint simulations with real flight control systems, greatly expanding the application scope of the system.
[0056] The hybrid channel modeling module, based on ray tracing results, can not only calculate large-scale path loss, but also generate small-scale fading characteristics, Doppler power spectrum, and time-varying channel impulse response. The system supports terminal mobile trajectory input and can dynamically update channel state. It is suitable for dynamic channel simulation of mobile platforms such as UAVs, and provides reliable simulation basis for communication system design, link budget and network planning.
[0057] The system features a clear modular design with well-defined interfaces between modules, facilitating functional expansion and algorithm replacement. The standardized output format supports integration with external systems such as UAV simulation platforms and communication protocol stacks, enabling closed-loop simulation and joint testing. It provides a complete simulation verification environment for applications such as urban air traffic and low-altitude IoT. Through the visualization analysis module, users can view multi-dimensional simulation results in real time, including ray paths, signal strength distribution, and channel parameters. The system also supports human-computer interaction and scene parameter adjustment, greatly enhancing its practicality and ease of use in planning, debugging, and educational demonstrations. Attached Figure Description
[0058] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0059] Figure 1 This is a flowchart of the simulation system of the present invention. Detailed Implementation
[0060] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described in this specification are merely for explaining the invention and are not intended to limit the invention.
[0061] Example:
[0062] Please see Figure 1 This invention provides a simulation system for urban low-altitude channel and line-of-sight propagation that integrates BIM and GIS data. This embodiment simulates a public safety incident occurring in the core area of a city, in which a drone equipped with emergency communication relay equipment flies along a predetermined route, establishes a link with the ground command center base station, and evaluates the dynamic changes in its communication quality.
[0063] The data fusion processing module receives BIM and GIS data, performs preprocessing on the received data including coordinate system I, semantic mapping, and physical parameter assignment, and outputs a unified scene map.
[0064] After the system starts, users can import IFC format BIM model files and GIS datasets for the target area through the system interface. These include CityGML building models, Shapefile road data, and raster DEM terrain data. The data interface layer of the data fusion processing module parses the BIM and GIS data respectively. The core processing layer automatically identifies three common feature points in the BIM model and GIS map, and uses a seven-parameter Helmholtz transform model to solve for translation, rotation, and scale parameters using the least squares method. Using the solved parameters, all geometric vertices in the BIM model are transformed to a global or local projection coordinate system consistent with the GIS data, achieving spatial alignment with millimeter-level precision.
[0065] The system associates the semantics of components such as "concrete exterior walls and glass curtain walls" in BIM with the "building" entity in GIS and uniformly labels them as "building facades". Based on the user-defined simulation center frequency of 2.4 GHz, the core processing layer queries the built-in material electromagnetic parameter database and automatically assigns corresponding electromagnetic parameters such as complex permittivity and conductivity to semantic areas such as "building facades". The output layer generates a unified scene map, which is organized in a tree structure. The root node is the entire simulation area, and the leaf nodes are sets of triangular patches with vertices, normals, semantic labels and electromagnetic properties.
[0066] The 3D scene construction module receives the scene graph and converts it into a highly optimized spatial index structure that supports sub-millisecond ray casting.
[0067] The module receives the scene graph and first recursively constructs an octree, setting the threshold for the number of triangular faces in the leaf nodes to 1024. After partitioning, an octree of depth 8 is generated, in which a large number of nodes in high-altitude and street areas are marked as empty. Subsequently, the BVH (Binary Value Hierarchy) is constructed in parallel for each leaf node of the octree containing geometry. For a node containing a building facade, a surface area heuristic algorithm is used to construct a high-quality BVH binary tree for its 5237 triangular faces. The constructed hybrid acceleration structure and geometric data are serialized to prepare for GPU computing.
[0068] The deterministic propagation calculation module is used to calculate the propagation path of a signal based on a spatial index structure, combined with the mirror method and the bouncing ray algorithm.
[0069] The user sets the transmitter in the GUI as a ground command center base station, located on a rooftop, with coordinates (X, Y, Z), and sets the transmission power. The antenna is an omnidirectional antenna; the receiver is an airborne terminal for a UAV, and the simulation parameters are set to the carrier frequency. The maximum reflection order is 2, and the number of emitted rays is... .
[0070] When the user starts the simulation, the system receives data through the dynamic data interface. The external flight simulator begins to push real-time UAV pose data, including timestamp, latitude and longitude, roll angle, pitch angle and yaw angle. After receiving the first pose data packet, the system immediately converts it to the simulation coordinate system, places the UAV model in the corresponding position, and starts the channel simulation task for this time point.
[0071] The parallel computing and scheduling module is used to perform parallel ray tracing using a collaborative architecture of CPU and GPU.
[0072] The CPU-side task, primarily the main thread, prepares the current Tx and Rx positions, simulation parameters, and the "scene package" already loaded into memory. Through multiple CUDA streams, it asynchronously copies the ray emission start information required for this calculation to the GPU's global memory. The CPU then launches a large-scale CUDA kernel function, with each GPU thread responsible for tracing one initial ray. First, it quickly determines whether there is occlusion on the line-of-sight path from Tx to Rx. At this moment, the drone is located on the side of the building, and the LOS is occluded. The GPU compresses the captured path information and sends it back to the CPU. The CPU-side thread then aggregates all paths, including strong paths calculated using the image method, forming the current path list.
[0073] The hybrid channel modeling module is used to generate a comprehensive description of channel characteristics based on the calculated effective path information.
[0074] The hybrid channel modeling module processes the path list, generating a power delay profile and channel impulse response (PDP) based on path delay and complex amplitude. At this point, the PDP displays multiple significant reflection and diffraction peaks. The channel parameters are physically derived, and the power of the coherent superposition of all paths is calculated. Path loss, Doppler shift, and time delay.
[0075] The visualization and analysis module provides human-computer interaction and results. It displays the city model and the current location of the drone, and updates and displays a heatmap of received signal strength in real time on a horizontal slice of the drone's flight altitude. When the drone flies into the shadow area behind a building, that area immediately turns dark red on the heatmap, and the PDP curve of the current link is plotted in real time, displaying the calculated channel parameters. The system continuously processes new poses of MQTT inflows at a rate of approximately 1 second per point, repeating the above steps to achieve dynamic and continuous evaluation of channel characteristics.
[0076] This embodiment fully verifies the complete capabilities of the system of the present invention in processing multi-source heterogeneous data fusion, large-scale scene efficient simulation, physical self-consistent channel modeling, and dynamic real-time access and visualization. It realizes high-fidelity and high-efficiency simulation analysis of urban low-altitude dynamic communication links, providing a basis for low-altitude communication network planning.
[0077] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.
Claims
1. A simulation system for urban low-altitude channel and line-of-sight propagation integrating BIM and GIS data, characterized in that: include: The data fusion processing module is used to receive BIM and GIS data, and to preprocess the received data by performing coordinate system I, semantic mapping and physical parameter assignment, and output a unified scene map. A 3D scene construction module is used to receive a scene graph and convert it into a highly optimized spatial index structure that supports sub-millisecond ray casting. A deterministic propagation calculation module is used to calculate the propagation path of a signal based on a spatial index structure, combined with the mirror method and the bouncing ray algorithm. A hybrid channel modeling module is used to generate a comprehensive description of channel characteristics based on the calculated effective path information. A parallel computing and scheduling module, which is used to perform parallel ray tracing using a collaborative architecture of CPU and GPU; A visualization analysis module is used to provide human-computer interaction and results.
2. The urban low-altitude channel and line-of-sight propagation simulation system integrating BIM and GIS data according to claim 1, characterized in that: The specific architecture of the data fusion processing module includes a data interface layer, a core processing layer, and an output layer; The data interface layer is used for reading, parsing, and temporarily storing raw data, including BIM data interface, GIS data interface, and temporary storage. The BIM data interface is used to parse industrial basic files and extract the geometric characteristics, semantics, and material properties of building components; the GIS data interface is used to parse CityGML, Shapefile, and raster digital elevation models. The core processing layer is used to unify the data obtained by the data interface layer in a three-dimensional coordinate system. In the unified coordinate system, it repairs, simplifies and aligns the geometric model with semantic information, and assigns precise electromagnetic physical parameters to each geometric patch. The core processing layer has a built-in material electromagnetic parameter database divided by frequency band. The output layer is used to generate and output a unified scene graph for use by subsequent ray tracing modules.
3. The urban low-altitude channel and line-of-sight propagation simulation system integrating BIM and GIS data according to claim 2, characterized in that: The specific data processing flow of the data fusion processing module is as follows: First, data input parsing: For BIM files input from the BIM data interface and GIS datasets input from the GIS data interface, the corresponding parser is called to parse the input data; Next, a unified coordinate system is established. At least three easily identifiable and highly accurate feature points with the same name are selected on the analyzed BIM model and GIS map, and the coordinates of the feature points in the BIM local coordinate system are obtained. and coordinates in the GIS projected coordinate system Using the seven-parameter Helmert transform, the expression is: In the formula is the translation vector, s is the scale factor used to compensate for scale differences in measurements, R is the rotation matrix of the transformation model, representing rotation along three axes, and the seven parameters include three translation parameters, one scale parameter, and three rotation parameters, which are obtained by least squares adjustment. Using the calculated seven parameters, coordinate transformation is performed on all geometric vertices in the BIM model to obtain data spatially aligned with GIS data. Then, semantic mapping associates the refined component information of BIM with the macro-geographic entity classification of GIS, assigning a unified semantic label to provide semantic labels for subsequent physical calculations. Next, physical property values are assigned. For regions with semantic tags, the electromagnetic parameter database of materials is queried according to the simulation center frequency set by the user to obtain the corresponding electromagnetic parameters, including the phase velocity of the influencing wave, the conductivity of the material and the vacuum dielectric constant. The obtained electromagnetic parameters are then attached as data. Finally, the data from different sources processed in the above steps are organized into a logically unified scene graph. The root node of the scene graph represents the entire simulation region, and each leaf node contains one or more sets of triangular faces. Each face includes vertex coordinates, normal vectors, semantic labels, and electromagnetic physical properties. The constructed scene graph is then output.
4. The urban low-altitude channel and line-of-sight propagation simulation system integrating BIM and GIS data according to claim 1, characterized in that: The 3D scene construction module includes a scene management layer, a core acceleration structure construction layer, and a serialization output layer; The scene management layer is used to receive a unified scene graph, perform overall scene analysis, calculate the global bounding box in the world coordinate system, and preset construction parameters according to the scene scale and complexity. The core acceleration structure building layer is used to perform recursive and uniform octree partitioning of the entire scene space from top to bottom on a macroscopic level, quickly indexing spatial locations and eliminating large empty areas. On a microscopic level, it constructs a locally optimized BVH for the set of triangular faces managed by each octree leaf node containing geometry from bottom to top. The serialization output layer is used to flatten and serialize the complex pointer-linked structure that is built and exists in main memory, converting it into a linear memory layout suitable for efficient GPU access. The output includes geometric data, physical attribute data, and most importantly, hybrid accelerated structure data that can be directly traversed by GPU kernel functions.
5. The urban low-altitude channel and line-of-sight propagation simulation system integrating BIM and GIS data according to claim 1, characterized in that: The deterministic propagation calculation module includes The simulation control unit is used to parse the simulation tasks input by the user, initialize the global simulation state, and coordinate the execution of the simulation process by each unit. The ray emitter unit is used to generate an initial set of rays originating from the emission source according to the selected mode. The modes are divided into the mirror mode and the SBR mode. The mirror mode is for specified low-order reflections and does not emit wide-angle rays. Based on the principles of geometric optics, it calculates the mirror points to accurately solve the specific specular reflection path. The SBR mode is used to capture high-order reflections, diffuse reflections and diffraction paths. It uses an algorithm based on geodesic dome or uniform solid angle sampling to emit discrete rays into the entire upper half space. A ray-based scene interaction engine is used to drive the propagation, interaction, and evolution of each ray in the scene. The physical interaction calculation unit includes a reflection, transmission calculator and a diffraction calculator, which are used to calculate the reflection coefficient and transmission coefficient of rays at the interface of media and to handle the diffraction phenomenon of rays at the sharp edges of buildings. A ray tracing unit is used to capture propagating rays and record their geometric and electromagnetic information. The ray period management unit is used to track the status of each ray and decides to terminate the ray when the ray energy is lower than a preset threshold, reaches the maximum reflection or diffraction order, or leaves the simulation area.
6. The urban low-altitude channel and line-of-sight propagation simulation system integrating BIM and GIS data according to claim 5, characterized in that: The reflection and transmission calculator uses Fresnel's equations for calculation. The expression for calculating the reflection coefficient is: In the formula Angle of incidence and The reflection coefficient is the polarization of the incident angle. By Snell's Law The complex refractive index is calculated. , Where is the complex permittivity; the expression for calculating the transmission coefficient is: In the formula This refers to the impedance characteristics of the medium for the propagation of electromagnetic waves. The diffraction calculator uses uniform diffraction theory to determine a diffraction point on the edge of the wedge. Uniform diffraction theory provides a set of complex diffraction coefficients D, and the diffraction field strength is... In the formula For a strong incident field, For diffusion factor, This represents the distance from the diffraction point to the field point.
7. A simulation system for urban low-altitude channel and line-of-sight propagation integrating BIM and GIS data as described in claim 6, characterized in that: The hybrid channel modeling module includes The path data processing unit receives, verifies, and preprocesses raw path data from the ray tracing unit, including complex amplitude. Propagation delay Departure angle, arrival angle, and polarization information; The large-scale fading calculation unit is used to coherently superimpose the complex amplitudes of all effective paths to obtain the total received electric field. Then calculate the total received power. and path loss In the formula This refers to the transmission power. Small-scale fading calculation units are used to calculate the merged paths according to their time delays. Sort the paths and calculate the power of each path; The Doppler spectrum calculation unit is used to calculate the Doppler frequency shift for each path of the mobile terminal and synthesize the Doppler power spectrum of the channel. That is, for a path, the angle of arrival and the terminal velocity vector are known. , The Doppler frequency shift of this path is In the formula For carrier frequency, At the speed of light, For wavelength, The power of each path is represented by a unit vector from the terminal toward the incident direction. Assigned to the corresponding Doppler frequency shift Then, the frequency domain is binned and the power is superimposed to obtain the discrete Doppler power spectrum. ; The time-varying channel impulse response unit is used to recalculate the delay, amplitude, and phase of each path at each sampling time based on the terminal position at that time for a given terminal trajectory. Then, these paths are superimposed to obtain the channel impulse response sequence at that time, which is used for time-domain simulation. The channel parameter output and visualization unit outputs the calculated channel parameters and model in a standard format and provides them to the visualization module for graphical display.
8. The urban low-altitude channel and line-of-sight propagation simulation system integrating BIM and GIS data according to claim 1, characterized in that: The parallel computing and scheduling module includes: The task scheduling and resource management layer runs on the CPU main thread. It is used to parse simulation tasks, perform macro-level task graph decomposition, and monitor CPU / GPU utilization, memory usage, and PCIe bandwidth in real time to provide decision-making basis for dynamic scheduling. The CPU parallel execution layer is used to execute tasks suitable for CPU parallelization. The GPU compute and memory management layer is used for the allocation, release, and lifecycle management of GPU memory, as well as configuring and launching GPU kernel functions and optimizing the dimensions of mesh and thread blocks. The data transmission and synchronization layer is used for data synchronization and transmission.
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 program, it implements the system according to any one of claims 1-8.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the system described in any one of claims 1-8.