A hybrid wi-fi channel modeling method and system based on environment perception

Abstract

The application discloses a kind of based on environmental perception's mixed Wi-Fi channel modeling method and system, belong to wireless communication technical field, comprising: the real-time data of current communication environment is obtained, according to real-time data acquisition multimodal fusion output;According to multimodal fusion output, the object in communication environment is identified, to classify communication environment, classification category includes: quasi-static environment, local dynamic environment, global dynamic environment;According to the selected strategy in strategy library according to classification category, strategy includes high-precision mode, fast update mode, statistical auxiliary mode;According to the selected strategy execution channel parameter calculation, to obtain channel impulse response;Channel impulse response is formatted and packaged and output. In prior art, channel modeling cannot be completed under low computational complexity. Compared with prior art, the application adopts hybrid architecture, accurate modeling is realized under low computational complexity.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a hybrid Wi-Fi channel modeling method and system based on environment awareness. Background Technology

[0002] In wireless communication system design, network planning, and optimization, accurate channel modeling is a prerequisite for evaluating system performance and designing transmission schemes. The goal of Wi-Fi channel modeling is to accurately describe the variation characteristics of parameters such as amplitude, phase, delay, and angle of arrival of radio waves in the propagation environment.

[0003] Currently, the mainstream Wi-Fi channel modeling methods can be mainly divided into three categories: Statistical channel modeling: Based on a large amount of measured data, statistical features such as path loss, delay spread, and angle spread are extracted to establish standardized models such as 3GPP SCM and IEEE 802.11 TGn. Its advantages are simple computation and strong generalization ability, but its disadvantages are that it is not sensitive to the geometric features of specific environments and cannot accurately reflect the real propagation details in specific scenarios (such as offices and factories), resulting in limited simulation accuracy.

[0004] Deterministic channel modeling, represented by ray tracing, calculates the propagation path of each multipath using geometric optics and uniform diffraction theory, based on a precise 3D geometric model of the environment and material electromagnetic parameters. Its advantages include high accuracy and the ability to reflect scene-specific characteristics. However, its disadvantages include extremely high computational complexity, the requirement for a high-precision 3D environmental model beforehand, and difficulty in adapting to dynamically changing environments.

[0005] Hybrid channel modeling attempts to simply superimpose statistical and deterministic models, for example, using a statistical model to generate scattering components and a deterministic model to calculate direct and reflected paths. However, existing hybrid methods suffer from rigid coupling and insufficient utilization of environmental information. When local changes occur in the environment (such as opening and closing of doors and windows, or movement of people), the model cannot be updated quickly, making it difficult to balance real-time performance and accuracy.

[0006] In summary, the main drawbacks of existing technologies can be summarized as follows: they cannot achieve accurate, fast, and adaptive Wi-Fi channel modeling for specific physical environments while maintaining low computational complexity.

[0007] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0008] This invention provides a hybrid Wi-Fi channel modeling method based on environment awareness, including: acquiring real-time data of the current communication environment, the real-time data including point cloud geometric data and coarse-grained radio data, and acquiring multi-modal fusion output based on the real-time data; Based on the multimodal fusion output, objects within the communication environment are identified to classify the communication environment. The classification categories include: quasi-static environment, local dynamic environment, and global dynamic environment. Select the appropriate strategy from the strategy library based on the classification category. The strategies include high-precision mode, fast update mode, and statistical assistance mode. Channel parameters are calculated based on the selected strategy to obtain the channel impulse response; The channel impulse response is formatted, encapsulated, and output.

[0009] Furthermore, the point cloud geometric data is as follows:

[0010] in: Let i be the spatial coordinates of the i-th point in the global coordinate system. The total number of point clouds, For timestamps; It is a three-dimensional real number space; The coarse-grained data for radio is as follows:

[0011] in: For receiving signal strength indication, This refers to the signal arrival time; By fusing point cloud geometric data and coarse-grained radio data, a multimodal fusion output is obtained as follows: .

[0012] Furthermore, based on the multimodal fusion output, the identification of objects within the communication environment includes the following steps: Voxel filtering is used to reduce the point cloud density of the point cloud geometric data. A random sample consensus algorithm is then used to iteratively extract the maximum plane. The formula for the plane model of the maximum plane is:

[0013] in: It is a plane normal vector; Let be any point in the point cloud geometric data; The distance between the plane and the origin; A KD-tree spatial index is constructed for the foreground residual point cloud, and Euclidean clustering is performed. When the decision criterion is met, the objects are determined to belong to the same object cluster. The decision criterion is as follows:

[0014] in: This is the clustering distance threshold; Points satisfying connectivity are aggregated into K independent clusters using a breadth-first traversal, using the following formula:

[0015] in: Foreground residual point clouds; each cluster Corresponding to a physical object; For each cluster Calculate the bounding box size and centroid height, and then divide the physical objects into a static background set based on the bounding box size and centroid height. Dynamic object collection .

[0016] Furthermore, the communication environment is classified, including the following steps: Calculate dynamic object sets The displacement of the centroid of the k-th dynamic object is given by the following formula:

[0017] in, For the first The centroid coordinates of a dynamic object; The total environmental change is calculated based on the centroid displacement using the following formula:

[0018] in: As weight; A collection of dynamic objects The number of elements in the middle; By comparing the total amount of environmental changes with a threshold, the communication environment is classified based on the comparison results to obtain the classification category. The formula used is as follows: .

[0019] Furthermore, select the appropriate strategy from the strategy library based on the classification category, including the following steps: The selection mapping relationship of the defined strategy is represented by the following formula:

[0020] in: To select the mapping relationship, ; The target signal-to-noise ratio; For calculating resource indicators; The maximum number of reflections is configured based on the selected mapping relationship:

[0021] The maximum number of diffractions is configured based on the selected mapping relationship: .

[0022] Furthermore, channel parameter calculation is performed according to the selected strategy, including the following steps: The general channel impulse response model is constructed as follows:

[0023] in: For the number of multipaths; Let the amplitude be the i-th diameter; The phase of the i-th path; The time delay of the i-th path; (·) represents the Dirac function; j is the imaginary unit; The formula for calculating a deterministic electric field is as follows:

[0024] in: This serves as the launch reference field. Wave number; This refers to the direct range; Let q be the Fresnel reflection coefficient of the q-th reflection; This represents the total path of the m-th reflection path; Here is the diffraction coefficient matrix; The diffraction diffusion factor; The length of the nth diffraction path; Calculate the amplitude of the i-th diameter based on the deterministic electric field The formula used is:

[0025] in: Let be the magnitude of the complex electric field vector of the i-th path; The magnitude of the reference electric field at the transmitting end; To obtain the phase angle of the vertical polarization component; The scalar complex attenuation coefficient is decomposed into amplitude and phase using the following formula:

[0026]

[0027] The channel impulse response is calculated as follows:

[0028] in: This refers to the previous moment.

[0029] Furthermore, the channel impulse response is formatted, encapsulated, and output, including the following steps: The channel impulse response is converted into matrix form using the following formula:

[0030] in: Angle of arrival; Departure angle; For Doppler frequency shift; Perform a Fourier transform to output the frequency domain channel response:

[0031] in: For the frequency domain independent variable in the Fourier transform; {·} is the Fourier transform operator.

[0032] A hybrid Wi-Fi channel modeling system based on environment awareness, using the method described above, includes: an environment awareness module, a scene parsing and classification module, a hybrid modeling strategy library, a channel parameter calculation engine, and a model output module; The environment perception module is used to acquire real-time data of the current communication environment; The scene analysis and classification module is used to classify the communication environment based on real-time data and convert the classification results into control signals to be output to the hybrid modeling strategy library; A hybrid modeling strategy library is used to select strategies based on control signals and output the selection results to the channel parameter calculation engine. The channel parameter calculation engine is used to perform channel impulse response calculations according to the strategy. The model output module is used to convert the calculation results of the channel impulse response into a standard format output for simulation software or hardware simulation.

[0033] Compared with the prior art, the present invention has the following advantages: This invention achieves Wi-Fi channel modeling through a hybrid architecture based on environment awareness and policy configurability: when the environment is static, it provides accuracy comparable to purely deterministic modeling; when the environment changes, it quickly refreshes the model with far less computational overhead than re-performing global ray tracing, significantly improving the real-time performance and accuracy of channel modeling in dynamic scenarios.

[0034] This invention can automatically match appropriate modeling strategies based on the type of environmental change and available computing power without human intervention. Therefore, through adaptive strategy switching, this invention significantly reduces the average computational load over long periods while ensuring accuracy in critical scenarios, achieving a dynamic optimal balance between accuracy and efficiency.

[0035] The channel impulse response calculation process combines historical buffer complex coefficients with local incremental recalculation, and performs ray tracing calculations only on the changing regions in a local dynamic environment, reducing the channel update time from minutes to milliseconds.

[0036] This invention will use real-time sensing of object volume occupancy rate As a correction factor, the time delay spread automatically increases when objects are densely packed in the room (high volume occupancy), reflecting richer physical laws of multipath reflection; when the room is empty, the time delay spread tends to the baseline value. This correction makes the output of the statistical model dynamically change with the density of the real environment, making it more consistent with the actual propagation environment than a purely statistical model. Comparison with measured data shows that the accuracy can be improved by about 20%.

[0037] This invention achieves this by synchronously acquiring point cloud geometric data. With coarse-grained radio data merge into This enables complementary dual-mode perception capabilities of vision and radio, allowing for stable environmental perception even when a single sensing method fails.

[0038] This invention defines Normalized complex attenuation coefficient and its direction and The decomposition mapping realizes the transformation from deterministic electric field scalar complex attenuation coefficient Then and The mathematical chain establishes a reproducible quantitative correspondence between the microscopic physical quantities and macroscopic parameters of ray tracing. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the overall process in this embodiment.

[0040] Figure 2 : System architecture diagram of this embodiment. Detailed Implementation

[0041] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0042] Example 1: This invention provides an environment-aware hybrid Wi-Fi channel modeling method, referring to... Figure 1 , Figure 1 This is a schematic diagram of the process of the present invention. It includes the following steps: acquiring real-time data of the current communication environment, whereby the real-time data includes point cloud geometric data and coarse-grained radio data. The point cloud geometric data is as follows:

[0043] in: Let i be the spatial coordinates of the i-th point in the global coordinate system. The total number of point clouds, For timestamps; It is a three-dimensional real number space; The coarse-grained data for radio is as follows:

[0044] in: For receiving signal strength indication, This refers to the signal arrival time; By fusing point cloud geometric data and coarse-grained radio data, a multimodal fusion output is obtained as follows: .

[0045] In summary, this invention achieves its goal by simultaneously acquiring point cloud geometric data. With coarse-grained radio data merge into This enables complementary dual-mode perception capabilities of vision and radio, allowing for stable environmental perception even when a single sensing method fails.

[0046] Processing multimodal fusion output The process involves segmenting individual objects, tracking their positional changes, and quantitatively calculating the degree of environmental change. Point cloud geometric data Voxel mesh filtering is performed, dividing the 3D space into a cubic mesh with a fixed side length, retaining only the center point within each mesh, thus obtaining a sparse point cloud. .

[0047] Using the random sampling consensus algorithm from sparse point clouds The maximum plane is extracted iteratively, and the formula for the plane model of the maximum plane is:

[0048] in: It is a plane normal vector; Let be any point in the point cloud geometric data; This represents the distance from the plane to the origin. This process is repeated by randomly selecting three points to fit the plane, and counting the distances of interior points (points whose distance to the plane is less than...)... As a preferred option, Take a quantity of 0.02m, and retain the plane with the most interior points as the largest plane, such as walls, floors, and ceilings, and classify it as a static background. Then, these points are removed from the point cloud. This process is repeated until the number of remaining points is less than a threshold, thus separating large structural surfaces. At this point, the foreground residual point cloud after removing large structures is obtained. .

[0049] Foreground residual point cloud Construct a KD-tree spatial index and perform Euclidean clustering. When the decision criterion is met, the objects are determined to belong to the same object cluster. The decision criterion is as follows:

[0050] in: This is the clustering distance threshold; Points satisfying connectivity are aggregated into K independent clusters using a breadth-first traversal, using the following formula:

[0051] in: Foreground residual point clouds; each cluster It corresponds to a physical object, such as: a person, a movable partition, a table and chairs, etc.

[0052] For each cluster Calculate the bounding box dimensions Center of mass height .

[0053] If volume Furthermore, the centroid height fluctuates significantly over time (historical frame displacement standard deviation > 0.15m), and is marked as personnel.

[0054] If the volume is large and the shape is flat, that is And located in the door frame trajectory area (which can be pre-marked according to the house layout), it is marked as a door / window.

[0055] Larger objects of stable dimensions are labeled as movable furniture.

[0056] In summary, physical objects can be categorized into a set of static backgrounds based on their bounding box size and centroid height. Dynamic object collection .

[0057] In obtaining a collection of dynamic objects Then, calculate the dynamic object set. The displacement of the centroid of the k-th dynamic object is given by the following formula:

[0058] in, For the first The coordinates of the centroid of a dynamic object.

[0059] The total environmental change is calculated based on the centroid displacement using the following formula:

[0060] in: As weighted, for preferred components, personnel are weighted at 1.0, doors and windows at 0.8, and movable furniture at 0.5. A collection of dynamic objects The number of elements in the text.

[0061] By comparing the total amount of environmental changes with a threshold, the communication environment is classified based on the comparison results to obtain the classification category. The formula used is as follows: .

[0062] thus, For a quasi-static environment, For local dynamic environments, It is a global dynamic environment.

[0063] Then, select the appropriate strategy from the strategy library according to the classification category, including the following steps: The selection mapping relationship of the defined strategy is represented by the following formula:

[0064] in: To select the mapping relationship, ; The target signal-to-noise ratio; For calculating resource indicators; The specific decision-making logic is as follows:

[0065] The maximum number of reflections is configured based on the selected mapping relationship:

[0066] The maximum number of diffractions is configured based on the selected mapping relationship: .

[0067] in, This is the high-precision mode. For rapid update mode. This is a statistical auxiliary mode.

[0068] In summary, this invention can automatically match appropriate modeling strategies based on the type of environmental change and available computing power without human intervention. When the environment is static, the system operates with the highest accuracy ( 6 reflections =2 diffraction) provides accuracy approaching that of purely deterministic modeling; when the environment changes locally, the system achieves moderate accuracy ( 3 reflections =1 diffraction) for rapid tracking of changes; when the environment changes drastically or computing power is strained, the system completely shuts down the ray tracing core ( =1 diffraction). 0、 =0), relying solely on statistical models to maintain the output. Therefore, this invention, through adaptive strategy switching, significantly reduces the average computational load over long periods while ensuring accuracy in critical scenarios, achieving a dynamic optimal balance between accuracy and efficiency.

[0069] Channel parameters are calculated based on the selected strategy to obtain the channel impulse response. The specific process is as follows: A general channel impulse response model is constructed as follows:

[0070] in: The number of multipaths at time t; Let the amplitude be the i-th diameter; The phase of the i-th path; The time delay of the i-th path; (·) represents the Dirac function; j is the imaginary unit; When strategy A is selected, the deterministic electric field is calculated using the following formula:

[0071] in: This serves as the launch reference field. Wave number; This refers to the direct range; Let q be the Fresnel reflection coefficient of the q-th reflection; This represents the total path of the m-th reflection path; Here is the diffraction coefficient matrix; The diffraction diffusion factor; The length of the nth diffraction path; When strategy B is selected, the deterministic electric field must be calculated first, and then the following process must be performed: The scalar complex attenuation coefficient is calculated based on a deterministic electric field using the following formula:

[0072] in: Let be the magnitude of the complex electric field vector of the i-th path; The magnitude of the reference electric field at the transmitting end; To obtain the phase angle of the vertical polarization component; After the aforementioned mapping, the scalar complex attenuation coefficient is decomposed into amplitude and phase, respectively:

[0073]

[0074] The channel impulse response is calculated as follows: .

[0075] in: This refers to the previous moment.

[0076] As shown in the formula above, the channel impulse response is divided into two terms by curly braces. The first term can be directly read from the buffer using historical complex coefficients. No recalculation is required. The second item, for each path... ,calculate , then calculate , Meanwhile, the path length is obtained through ray tracing. calculate .

[0077] In summary, this invention defines... Normalized complex attenuation coefficient and its direction and The decomposition mapping realizes the transformation from deterministic electric field scalar complex attenuation coefficient Then and The mathematical chain establishes a reproducible quantitative correspondence between the microscopic physical quantities and macroscopic parameters of ray tracing.

[0078] For strategy C, the parameters in the channel impulse response model are generated using the standard ABG path loss model and exponential delay distribution, utilizing the aforementioned... The classification process extracts room volume and the total volume of the object right Perform set constraint correction, specifically as follows: .

[0079] in, This provides the basic root mean square time delay extension given by the standard statistical model. Therefore, this invention will use the real-time sensed object volume occupancy rate... As a correction factor, the time delay spread automatically increases when objects are densely packed in the room (high volume occupancy), reflecting richer physical laws of multipath reflection; when the room is empty, the time delay spread tends to the baseline value. This correction makes the output of the statistical model dynamically change with the density of the real environment, making it more consistent with the actual propagation environment than a purely statistical model. Comparison with measured data shows that the accuracy can be improved by about 20%.

[0080] In summary, the aforementioned channel impulse response combines historical buffer complex coefficients with local incremental recalculation, performing ray tracing calculations only on the changing regions in a local dynamic environment, thereby reducing channel update time from minutes to milliseconds.

[0081] The channel impulse response is formatted, encapsulated, and output, including the following steps: The channel impulse response is converted into matrix form using the following formula:

[0082] in: Angle of arrival; Departure angle; For Doppler frequency shift; Perform a Fourier transform to output the frequency domain channel response:

[0083] in: is the frequency domain independent variable in the Fourier transform. {·} is the Fourier transform operator.

[0084] In summary, this invention achieves Wi-Fi channel modeling through a hybrid architecture based on environment awareness and policy configurability: when the environment is static, it provides accuracy comparable to purely deterministic modeling; when the environment changes, it quickly refreshes the model with a computational cost far lower than re-performing global ray tracing, significantly improving the real-time performance and accuracy of channel modeling in dynamic scenarios.

[0085] Example 2: An environment-aware hybrid Wi-Fi channel modeling system is provided, applying the method described in Example 1. See details... Figure 2 , Figure 2 This is the architecture diagram of the system. It includes: environment perception module, scene parsing and classification module, hybrid modeling strategy library, channel parameter calculation engine, and model output module; The environment perception module is used to acquire real-time data of the current communication environment; The scene analysis and classification module is used to classify the communication environment based on real-time data and convert the classification results into control signals to be output to the hybrid modeling strategy library; A hybrid modeling strategy library is used to select strategies based on control signals and output the selection results to the channel parameter calculation engine. The channel parameter calculation engine is used to perform channel impulse response calculations according to the strategy. The model output module is used to convert the calculation results of the channel impulse response into a standard format output for simulation software or hardware simulation.

[0086] It is understood that the system provided in the embodiments of the present invention corresponds to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant content can be referred to the corresponding parts of the above method.

[0087] Furthermore, it should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0088] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0089] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory (ROM) / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0090] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

[0091] It is understood that the system provided in the embodiments of the present invention corresponds to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant content can be referred to the corresponding parts of the above method.

Claims

1. A hybrid Wi-Fi channel modeling method based on environment awareness, characterized in that, include: Acquire real-time data of the current communication environment, including point cloud geometric data and coarse-grained radio data, and obtain multimodal fusion output based on the real-time data; Based on the multimodal fusion output, objects within the communication environment are identified to classify the communication environment. The classification categories include: quasi-static environment, local dynamic environment, and global dynamic environment. Select the corresponding strategy from the strategy library according to the classification category. The strategy includes high-precision mode, fast update mode, and statistical auxiliary mode. Channel parameters are calculated based on the selected strategy to obtain the channel impulse response; The channel impulse response is formatted, encapsulated, and output.

2. The environment-aware hybrid Wi-Fi channel modeling method as described in claim 1, characterized in that, The point cloud geometric data is as follows: in: Let i be the spatial coordinates of the i-th point in the global coordinate system. The total number of point clouds, For timestamps; It is a three-dimensional real number space; The coarse-grained radio data is: in: For receiving signal strength indication, This refers to the signal arrival time; The point cloud geometric data and the coarse-grained radio data are fused to obtain the multimodal fusion output as follows: 。 3. The hybrid Wi-Fi channel modeling method based on environment awareness as described in claim 1, characterized in that, The process of identifying objects within the communication environment based on the multimodal fusion output includes the following steps: The point cloud density of the point cloud geometric data is reduced by voxel filtering, and the maximum plane is extracted iteratively using a random sample consensus algorithm. The plane model formula of the maximum plane is: in: It is a plane normal vector; For any point in the point cloud geometric data; The distance between the plane and the origin; A KD-tree spatial index is constructed for the foreground residual point cloud, and Euclidean clustering is performed. When the determination criteria are met, the objects are determined to belong to the same object cluster. The determination criteria are: in: This is the clustering distance threshold; Points satisfying connectivity are aggregated into K independent clusters using a breadth-first traversal, using the following formula: in: For the foreground residual point cloud; each cluster It corresponds to a physical object; For each of the clusters Calculate the bounding box size and centroid height, and then divide the physical object into a static background set based on the bounding box size and centroid height. Dynamic object collection .

4. The environment-aware hybrid Wi-Fi channel modeling method as described in claim 3, characterized in that, The classification of the communication environment includes the following steps: Calculate the set of dynamic objects The displacement of the centroid of the k-th dynamic object is given by the following formula: in, For the first The centroid coordinates of a dynamic object; The total environmental change is calculated based on the centroid displacement using the following formula: in: As weight; The set of dynamic objects The number of elements in the middle; By comparing the total amount of environmental changes with a threshold, the communication environment is classified based on the comparison results to obtain the classification category, using the following formula: 。 5. The environment-aware hybrid Wi-Fi channel modeling method as described in claim 1, characterized in that, The step of selecting the appropriate strategy from the strategy library based on the classification category includes the following steps: The selection mapping relationship of the strategy is defined by the following formula: in: For the selection of the mapping relationship, ； The target signal-to-noise ratio; For calculating resource indicators; Based on the selected mapping relationship, the maximum number of reflections is configured as follows: Based on the aforementioned selection mapping relationship, the maximum number of diffraction attempts is configured as follows: 。 6. The environment-aware hybrid Wi-Fi channel modeling method as described in claim 1, characterized in that, The process of calculating channel parameters according to the selected strategy includes the following steps: The general channel impulse response model is constructed as follows: in: For the number of multipaths; Let the amplitude be the i-th diameter; The phase of the i-th path; The time delay of the i-th path; (·) represents the Dirac function; j is the imaginary unit; The formula for calculating a deterministic electric field is as follows: in: This serves as the launch reference field. Wave number; This refers to the direct range; Let q be the Fresnel reflection coefficient of the q-th reflection; This represents the total path of the m-th reflection path; Here is the diffraction coefficient matrix; The diffraction diffusion factor; The length of the nth diffraction path; The scalar complex attenuation coefficient is calculated based on the deterministic electric field using the following formula: in: Let be the magnitude of the complex electric field vector of the i-th path; The magnitude of the reference electric field at the transmitting end; To obtain the phase angle of the vertical polarization component; The scalar complex attenuation coefficient is decomposed into amplitude and phase using the following formula: The channel impulse response is calculated as follows: in: This refers to the previous moment.

7. The environment-aware hybrid Wi-Fi channel modeling method as described in claim 6, characterized in that, The channel impulse response is formatted, encapsulated, and output, including the following steps: The channel impulse response is converted into matrix form using the following formula: in: Angle of arrival; Departure angle; For Doppler frequency shift; Perform a Fourier transform to output the frequency domain channel response: in: For the frequency domain independent variable in the Fourier transform; {·} is the Fourier transform operator.

8. A hybrid Wi-Fi channel modeling system based on environment awareness, characterized in that, The method described in any one of claims 1 to 7 includes: an environment perception module, a scene parsing and classification module, a hybrid modeling strategy library, a channel parameter calculation engine, and a model output module; The environment perception module is used to acquire real-time data of the current communication environment; The scene analysis and classification module is used to classify the communication environment based on the real-time data and convert the classification results into control signals to be output to the hybrid modeling strategy library; The hybrid modeling strategy library is used to select strategies based on the control signals and output the selection results to the channel parameter calculation engine. The channel parameter calculation engine is used to perform channel impulse response calculation according to the strategy; The model output module is used to convert the calculation results of the channel impulse response into a standard format output of simulation software or hardware simulation.

Images