Multi-scene wireless network rapid deployment method based on electromagnetic field digital twinning technology

Abstract

The invention discloses a multi-scene wireless network rapid deployment method based on an electromagnetic field digital twinning technology, and relates to the technical field of computer application, and the method comprises the following steps: S1, constructing a three-dimensional digital twinning model of a corresponding scene, and integrating a building structure, equipment layout and material attribute data, and establishing a scene special material database containing metal equipment, a concrete wall body and an electromagnetic shielding layer. According to the multi-scene wireless network rapid deployment method provided by the invention, the three-dimensional digital twin model is constructed, and the building structure, the equipment layout and the material attribute data are integrated, so that the precise simulation of the environment is realized, the wireless network deployment accuracy is improved, the deployment scheme can better adapt to the dynamic change of the environment, and the deployment efficiency is improved. The improved O-FDTD algorithm is adopted to simulate electromagnetic field propagation in a building in real time, and the problem that the calculation efficiency is low when a traditional FDTD algorithm processes the metal edge effect and a moving object is effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of computer application technology, and in particular to a method for rapid deployment of multi-scenario wireless networks based on electromagnetic field digital twin technology. Background Art

[0002] With the advancement of new infrastructure construction, wireless network deployment in various scenarios faces the dual challenges of complex electromagnetic environments and dynamic business demands. In typical scenarios such as smart manufacturing, smart logistics, and smart grids, device interconnection places stringent demands on network reliability, latency, and capacity. Metal frames, composite materials, dense obstacle layouts, and the dynamic behavior of mobile devices in different scenarios lead to significant differences in electromagnetic wave propagation characteristics. For example, in smart port scenarios, the metal structures of container yards create complex reflection and diffraction paths; dynamic adjustments to shelf layouts in smart logistics warehouses alter the spatial distribution of signal propagation; and high-voltage equipment in smart grid substations introduces unique electromagnetic interference. Traditional wireless network deployment methods rely on static propagation models and manually empirical parameters, making them difficult to adapt to the dynamic electromagnetic environments and diverse business demands of multiple scenarios. This leads to problems such as unstable network performance and low spectrum resource utilization.

[0003] Existing technologies have exposed multiple limitations when dealing with multi-scenario deployment: First, traditional electromagnetic simulation methods have high computational complexity when dealing with multi-material coupling and dynamic scenarios, making it difficult to meet the needs of rapid deployment. Second, material databases lack refined modeling of multiple physical parameters, resulting in large deviations between simulation results and actual scenarios. Third, signal acquisition methods rely on fixed sensor nodes and cannot cover the dynamic needs of mobile scenarios. In addition, multi-band coordination mechanisms do not fully utilize the waveguide effect of metal structures, limiting the efficient use of spectrum resources. In addition, network optimization algorithms often target a single performance indicator, ignoring the comprehensive balance of coverage, capacity, and interference, resulting in insufficient system robustness. To address this, we propose a multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology. Summary of the Invention

[0004] In order to solve the above technical problems, a multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology is provided. This technical solution solves the above problems.

[0005] In order to achieve the above objects, the technical solution adopted by the present invention is:

[0006] The multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology includes the following steps:

[0007] S1. Build a 3D digital twin model of the corresponding scene, integrate building structure, equipment layout, and material property data, and establish a scene-specific material database. The dedicated material database includes but is not limited to: metal equipment, concrete walls, electromagnetic shielding layers, glass, and furniture;

[0008] S2. Use the improved OFDTD algorithm to simulate electromagnetic field propagation in real time within the scene. By importing materials and using a complex dielectric constant parameterized model, dynamically simulate the signal coverage changes in multiple scenarios.

[0009] S3: Deploy distributed edge computing nodes equipped with WiFi transceiver modules, integrate UWB positioning, visual SLAM and path planning algorithms, build an autonomous mobile signal acquisition network, and generate real-time RSSI signal strength heat maps;

[0010] S4. Develop a multi-band adaptive algorithm based on OFDTD prediction results to intelligently select a frequency band combination, including 2.4 GHz, 5 GHz, and Sub-1 GHz. Dynamically select one of the following enhancement strategies through the environmental feature recognition module:

[0011] (1) When a metal structure is detected, a multipath enhanced communication corridor is constructed using the waveguide effect;

[0012] (2) When non-metallic structures are detected, penetration loss compensation and multipath diversity reception technology are used to improve signal quality;

[0013] S5. Dynamically adjust network node deployment parameters based on real-time collected signal data and simulation results. The network nodes include core nodes and edge nodes, support distributed collaborative deployment, determine the optimal location of multiple nodes through particle swarm optimization algorithm, and combine game theory model to balance the transmission power and interference between nodes to optimize the network topology.

[0014] Preferably, the method for constructing the scene-specific material database in step S1 specifically includes:

[0015] For metal equipment, the waveguide method is used to measure the surface roughness parameter δ and complex dielectric constant ε of the metal equipment r The association model of is used to establish the modified Drude model:

[0016]

[0017] Where, ε r is the complex dielectric constant of the material, ε r ′ is the real part of the complex dielectric constant, ε r″ is the imaginary part of the complex dielectric constant, ε0 is the vacuum dielectric constant, δ is the surface roughness parameter of the metal device, σ is the material conductivity, ω is the angular frequency of the electromagnetic wave, and α, β, and β are experimental fitting parameters;

[0018] For concrete walls, a layered dielectric model was constructed, and the relationship between humidity and equivalent dielectric constant was measured using time domain reflectometry:

[0019] ε eff =a+bθ-cθ 2

[0020] Where, ε eff is the equivalent dielectric constant of the concrete wall, a is the reference dielectric constant when the humidity is zero, b is the linear influence coefficient of humidity on the dielectric constant, θ is the humidity of the concrete wall, and c is the nonlinear correction coefficient of humidity on the dielectric constant;

[0021] For the electromagnetic shielding layer, the transmission line method is used to establish its surface transfer impedance:

[0022]

[0023] Where Z T is the surface transfer impedance of the shielding layer, d is the thickness of the shielding layer, and μ is the magnetic permeability;

[0024] For glass materials, the transmission line method is used to measure the signal transmission parameters of glass samples in multiple frequency bands. By analyzing the signal transmission coefficient and attenuation constant, the complex dielectric constant is inverted. For coated glass with a metal coating on the surface, it is equivalent to a layered dielectric structure combining metal layers and glass layers. The transmission matrix method, similar to the modeling of multi-layer structures of electromagnetic shielding layers, is used to calculate the complex dielectric constant to reflect its special electromagnetic properties.

[0025] For wood furniture materials, a correlation model between the equivalent dielectric constant of wood and its moisture content was established. The dielectric constant data at different moisture contents were measured using time-domain reflectometry, and the humidity sensitivity coefficient was fitted to form a humidity-dependent dielectric constant expression suitable for wood. Based on the fiber orientation characteristics of solid wood panels, the concept of tensor dielectric constant was introduced to address the coupling effect between the polarization direction of electromagnetic waves and the direction of wood grain.

[0026] For plastics and composite materials, the resonant cavity method is used to measure the complex dielectric constant at different frequencies, and a relationship model between frequency and the real and imaginary parts of the dielectric constant is established through polynomial fitting. For composite materials containing metal particles or glass fiber fillers, the Maxwell-Garnett effective medium theory is used to calculate their equivalent dielectric constant to reflect the influence of the filler on the electromagnetic properties of the material.

[0027] Preferably, the modeling method of the electromagnetic shielding layer further comprises:

[0028] Establishing Shielding Effectiveness:

[0029]

[0030] Where E0 is the incident electric field intensity, E1 is the transmitted electric field intensity, is the reflection coefficient, is the decay constant;

[0031] For multi-layer shielding structures, the transfer matrix method is used to calculate the overall shielding effectiveness:

[0032]

[0033] Where, is the propagation constant, is the characteristic impedance.

[0034] Preferably, the specific implementation of the improved OF-FDTD algorithm in step S2 includes:

[0035] A dynamic time step adjustment mechanism is introduced based on the traditional FDTD algorithm:

[0036]

[0037] Where Δt is the time step, the size of the Δx, Δy, Δz spatial grid, c is the speed of light in vacuum, σ max is the maximum conductivity in the current calculation domain;

[0038] For the edges of metal devices, a subgrid technique is implemented. An n×n×n subgrid (n=5) is inserted into the standard grid (Δx, Δy, Δz). An explicit and implicit hybrid iterative method is used to solve the discrete form of the Maxwell curl equation:

[0039]

[0040] Where H is the magnetic field strength, E is the electric field strength, and J s is the surface current density;

[0041] A dynamic boundary condition updating algorithm is established for moving objects, and the moving coordinate system transformation method is used to deal with the sudden changes in field quantities caused by the movement of objects.

[0042] Preferably, the method for constructing the autonomous mobile signal acquisition network in step S3 specifically includes:

[0043] Design a dual-frequency UWB positioning module to transmit a Chirp spread spectrum signal with a pulse width of τ = 0.5ns:

[0044]

[0045] Where, τ = 0.5ns is the pulse width of the signal, f0 = 3.5GHz is the starting frequency of the signal, and k = 20MHz / ns is the frequency sweep slope;

[0046] Three-dimensional positioning is achieved through the arrival time difference algorithm, and the solution equation is:

[0047]

[0048] Where (x, y, z) is the coordinate of the target, (x i ,y i ,z i ) is the coordinate of the i-th anchor point, Δt ij is the time difference between the arrival of the signal at anchor points i and j;

[0049] The visual SLAM system uses ORB feature extraction and GMS matching algorithm to construct the photometric error function:

[0050]

[0051] Where E is the total photometric error, I i (p) is the pixel value of the i-th image at the feature point p, I j (p) is the pixel value of the jth image at the feature point p, W(p;ξ) is the warping function, and ξ∈se(3) is the pose represented by Lie algebra;

[0052] Path planning uses the improved A* algorithm, cost function:

[0053]

[0054] Where g(n) is the actual cost from the starting point to node n, h(n) is the heuristic estimated cost from node n to the end point, λ is the weight coefficient, and RSSI k Signal strength at the path point, RSSI th is the preset signal strength threshold.

[0055] Preferably, the method for generating the real-time signal strength heat map further comprises:

[0056] The Kriging interpolation algorithm is used to estimate the RSSI value of the unknown point and construct the semivariogram:

[0057]

[0058] Where h is the spatial hysteresis distance, z(x) is the RSSI value at position x;

[0059] Solve the kriging equations:

[0060] [Γ][λ]=[γ0]

[0061] Where Γ is the semivariance matrix of the known points, and γ0 is the semivariance vector of the point to be estimated;

[0062] Design a dynamic update mechanism. When new measurement data arrives, use Cholesky decomposition to update the inverse matrix Γ -1 , the update formula is:

[0063] L new =cholupdate(L,Δγ,′+′)

[0064] Where, L new is the updated lower triangular Cholesky factor matrix, L is the original Cholesky decomposition lower triangular matrix, Δγ is the semi-variant increment vector between the new data point and the existing data point, and + indicates that an additive update is to be performed.

[0065] Preferably, the method for implementing the multi-band adaptive algorithm in step S4 includes:

[0066] Establish the band selection decision function:

[0067] F=w1·SNR+w2·BER -1 +w3·C

[0068] Where w1, w2, w3 are the weights of each indicator, SNR = 10log(P r / N0) signal-to-noise ratio, is the bit error rate, C = Blog2(1+SNR) channel capacity, P r is the received power, N0 is the noise power spectral density, and B is the channel bandwidth;

[0069] The following enhancement strategies are dynamically selected based on the environmental feature recognition results:

[0070] When a metal scene is identified, a waveguide effect enhancement algorithm based on the metal structure is designed. When the parallel metal surface spacing d is detected,

[0071] When , activate the multipath enhancement mode and adjust the transmit power distribution to:

[0072]

[0073] Where, P t is the total transmission power, N is the number of paths participating in power allocation, P i is the transmission power of the i-th path, d i is the propagation distance of the i-th path, j is the imaginary unit;

[0074] When a non-metallic scene is identified, a penetration loss compensation algorithm is developed and a wall attenuation model is established based on the Sub-1 GHz frequency band:

[0075]

[0076] Where, P comp is the power after penetration loss compensation, P t is the transmission power, α wall is the attenuation coefficient of the wall, which describes the attenuation degree of the wall to the electromagnetic wave, d is the thickness of the wall through which the electromagnetic wave passes, is the transmittance coefficient, f i is the electromagnetic wave frequency corresponding to the i-th material, is the i-th material;

[0077] Develop a time-varying channel compensation algorithm and use a Kalman filter to estimate the channel impulse response:

[0078]

[0079] Where h(t,τ) is the impulse response of the time-varying channel, α k (t) is the time-varying complex gain, τ k (t) is the time delay, δ(τ-τ k (t)) is the Dirac delta function, which represents the impulse response with a time delay of τ k Discrete impulse at (t).

[0080] Preferably, the method for implementing the multipath enhanced communication corridor in step S4 specifically includes:

[0081] The following enhancement strategies are dynamically selected based on the environmental feature recognition results:

[0082] The metal scene is: when the metal surface distance d is detected to meet When , the waveguide mode is activated and the polarization direction of the antenna is adjusted to be parallel to the metal surface;

[0083] Design the delay precompensation filter:

[0084]

[0085] Where, is the delay of the nth reflection, a n =Γ n (1-Γ 2 ) is the filter coefficient;

[0086] In non-metallic scenarios, multipath diversity reception technology is used, and the maximum ratio combining algorithm is used at the receiving end:

[0087]

[0088] Where y is the total signal after the receiver combines, h m is the channel response of the mth path, |h k | 2 is the square of the magnitude of the channel response of the kth path.

[0089] Preferably, the specific method of dynamically adjusting the network node deployment parameters in step S5 includes:

[0090] Constructing network performance indicators:

[0091] J=α·Coverage+β·Capacity-γ·Interference

[0092] In the formula, α, β, γ are the weight coefficients of each indicator. For network coverage, is the network capacity, H is the step function;

[0093] Gradient descent method is used to optimize the node position (x i ,y i ,z i ), the parameter update formula is:

[0094]

[0095] Where, learning rate η = 0.01;

[0096] Develop a topology optimization algorithm based on game theory and define the utility function:

[0097]

[0098] Where U i is the utility value of node i, γ i is the signal-to-interference-and-noise ratio of node i, p i is the transmission power of node i, h ij is the channel coefficient from node j to node i, and the optimal power allocation is solved through Nash equalization.

[0099] Preferably, the network topology optimization method in step S5 includes:

[0100] Using the particle swarm optimization algorithm, define the particle position X i =[x1,y1,z1,…,x N ,y N ,z N ], speed update formula:

[0101]

[0102] Where \(w\) is the inertia weight, \(c_1\) and \(c_2\) are acceleration constants, \(r_1\) and \(r_2\) are random numbers, and pbest i is the historical optimal position of particle \(i\), and gbest is the global optimal position;

[0103] Fitness function:

[0104]

[0105] Where is the signal-to-interference-plus-noise ratio of the \(m\)th link, is the path loss, and \(\rho\) is the power penalty factor;

[0106] Design a constraint handling mechanism. When a particle violates the minimum spacing constraint \(\vert X_i - X_j\vert < d_{min}\), apply a repulsive force:

[0107] s

[0108] Where \(d\) min is the minimum spacing between nodes, and \(K\) is the elastic coefficient.

[0109] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0110] The multi-scenario wireless network rapid deployment method proposed by the present invention realizes the accurate simulation of the environment by constructing a three-dimensional digital twin model and integrating building structure, equipment layout and material property data. This not only improves the accuracy of wireless network deployment, but also enables the deployment plan to better adapt to the dynamic changes of the environment. The improved O-FDTD algorithm is used to perform real-time simulation of electromagnetic field propagation in buildings, effectively solving the problem of low computational efficiency of the traditional FDTD algorithm in dealing with metal edge effects and moving objects. This improvement makes it possible to dynamically simulate signal coverage changes, thereby further enhancing the self-adaptability of wireless network deployment. By integrating UWB positioning, visual SLAM and path planning algorithms, an autonomous mobile signal acquisition network is constructed, realizing real-time monitoring and acquisition of signal strength, which greatly improves the coverage rate and accuracy of signal acquisition, providing strong data support for the optimization of wireless networks. Utilize the waveguide effect of metal structures to construct a multipath enhanced communication corridor, effectively improving the utilization efficiency of spectrum resources. At the same time, by dynamically adjusting network node deployment parameters and optimizing network topology, the performance and robustness of the wireless network are further enhanced. Brief Description of the Drawings

[0111] Figure 1 is the method flow chart of the present invention;

[0112] Figure 2 is the schematic diagram of the key points of the method of the present invention. DETAILED DESCRIPTION

[0113] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0114] Reference Figure 1 As shown, the multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology includes the following steps:

[0115] S1. Build a 3D digital twin model of the corresponding scene, integrate building structure, equipment layout, and material property data, and establish a scene-specific material database. The dedicated material database includes but is not limited to: metal equipment, concrete walls, electromagnetic shielding layers, glass, and furniture;

[0116] The method for constructing a scene-specific material database specifically includes:

[0117] For metal equipment, the waveguide method is used to measure the correlation model between the surface roughness parameters of metal equipment and the complex dielectric constant, and a modified Drude model is established. This model considers the influence of vacuum dielectric constant, material conductivity, electromagnetic wave angular frequency and experimental fitting parameters on the complex dielectric constant.

[0118] For concrete walls, a layered dielectric model was constructed, and the relationship between humidity and equivalent dielectric constant was measured using time-domain reflectometry. This relationship reflects the linear and nonlinear effects of humidity on dielectric constant.

[0119] For the electromagnetic shielding layer, the transmission line method is used to establish its surface transfer impedance model, which takes into account the influence of the thickness of the shielding layer, the magnetic permeability and the angular frequency of the electromagnetic wave; the modeling method of the electromagnetic shielding layer further includes: establishing a shielding effectiveness model, which is based on the ratio of the incident electric field intensity to the transmitted electric field intensity, and takes into account the influence of the reflection coefficient and the attenuation constant; for the multi-layer shielding structure, the transmission matrix method is used to calculate the overall shielding effectiveness, taking into account the propagation constant and characteristic impedance of each layer of material.

[0120] For glass materials, the transmission line method is used to measure the signal transmission parameters of glass samples in multiple frequency bands. The complex dielectric constant is obtained by analyzing the signal transmission coefficient and attenuation constant. For coated glass with a metal coating on the surface, it is equivalent to a layered dielectric structure combining metal layers and glass layers. The complex dielectric constant is calculated using the transmission matrix method similar to the multi-layer structure modeling of electromagnetic shielding layers.

[0121] For wood furniture materials, a correlation model between the equivalent dielectric constant of wood and its moisture content was established. The dielectric constant data at different moisture contents were measured using time-domain reflectometry and fitted to obtain the humidity sensitivity coefficient, forming a humidity-dependent dielectric constant expression applicable to wood. Based on the fiber orientation characteristics of solid wood panels, the concept of tensor dielectric constant was introduced to address the coupling effect between the polarization direction of electromagnetic waves and the direction of wood grain.

[0122] For plastics and composite materials, the resonant cavity method is used to measure the complex dielectric constant at different frequencies, and a relationship model between frequency and the real and imaginary parts of the dielectric constant is established through polynomial fitting. For composite materials containing metal particles or glass fiber fillers, the Maxwell-Garnett effective medium theory is used to calculate their equivalent dielectric constant to reflect the influence of the filler on the electromagnetic properties of the material.

[0123] S2. Use the improved OFDTD algorithm to simulate the electromagnetic field propagation in the scene in real time. By importing materials and using the complex dielectric constant parameterized model, dynamically simulate the signal coverage changes in multiple scenarios.

[0124] The specific implementation of the improved OFDTD algorithm includes:

[0125] A dynamic time step adjustment mechanism is introduced based on the traditional FDTD algorithm. The time step is determined according to the spatial grid size and the maximum conductivity in the current calculation domain.

[0126] Implementing subgrid technology for the edges of metal devices, inserting subgrids into the standard grid, and using an explicit and implicit hybrid iterative method to solve the discrete form of Maxwell's curl equation;

[0127] A dynamic boundary condition updating algorithm is established for moving objects, and the moving coordinate system transformation method is used to deal with the sudden changes in field quantities caused by the movement of objects.

[0128] S3: Deploy distributed edge computing nodes equipped with WiFi transceiver modules, integrate UWB positioning, visual SLAM and path planning algorithms, build an autonomous mobile signal acquisition network, and generate real-time RSSI signal strength heat maps;

[0129] The method for constructing an autonomous mobile signal acquisition network specifically includes:

[0130] Design a dual-frequency UWB positioning module that transmits a Chirp spread spectrum signal with a specific pulse width and achieves three-dimensional positioning through the time difference of arrival algorithm;

[0131] The visual SLAM system uses ORB feature extraction and GMS matching algorithm to construct a photometric error function to optimize pose estimation;

[0132] Path planning utilizes a modified A* algorithm, with a cost function that comprehensively considers the actual cost, heuristically estimated costs, and the difference between the signal strength at each path point and a preset threshold. The method for generating real-time signal strength heat maps further includes: estimating the RSSI values ​​of unknown points using a Kriging interpolation algorithm; constructing a semivariogram to describe the relationship between spatial lag distance and signal strength differences; solving the Kriging equations to determine interpolation weights; and designing a dynamic update mechanism to update the inverse matrix via Cholesky decomposition when new measurement data arrives.

[0133] S4. Develop a multi-band adaptive algorithm based on OFDTD prediction results to intelligently select a frequency band combination, including 2.4 GHz, 5 GHz, and Sub-1 GHz.

[0134] The environmental feature recognition module distinguishes metals from non-metals by integrating electromagnetic property detection, visual structure analysis, and multi-band signal features. First, based on the material database established by S1, the conductivity and complex dielectric constant characteristics of the tested material are analyzed. Metallic materials exhibit extremely high conductivity and the imaginary part of the complex dielectric constant dominates, while non-metallic materials have extremely low conductivity and the real part of the dielectric constant varies with the material or humidity. In terms of signal reflection and attenuation, the metal surface causes strong reflection, resulting in a high proportion of multipath components in the received signal and drastic phase changes. Non-metallic materials are dominated by signal penetration, and the penetration loss is positively correlated with material thickness and dielectric constant. The visual SLAM system uses the ORB feature extraction algorithm to identify the geometry and texture of the structure. Metallic structures often have regular shapes and high reflectivity, while non-metallic structures have rough textures or irregular surfaces. Combined with the GMS matching algorithm, parameters such as edge sharpness and surface smoothness in the three-dimensional point cloud map are compared with the database template. In multi-band signal analysis, the 2.4GHz / 5GHz bands form multipath fading on metal surfaces, and the sub-1GHz band has strong penetration into non-metallic environments but exhibits significant attenuation in metallic environments. Based on this, the module constructs a classification decision function that integrates the signal strength, spatial distribution pattern of the signal-to-noise ratio, and edge sharpness indicators of different frequency bands.

[0135] In specific implementation, the edge computing node transmits 2.4G / 5G / Sub-1G multi-frequency signals and collects the amplitude, phase, and arrival time of the reflected signals. Based on the OFDTD simulation results of S2, a "material type-signal feature" mapping relationship is established. Metal identification requires that the signal reflection coefficient is greater than 0.8, the multipath delay difference is less than 1ns, and a regular metal surface is visually detected. Non-metal identification requires that the penetration loss in the Sub-1GHz band is greater than 10dB (10cm thickness), the real part of the dielectric constant is less than 20, and the visual features match the non-metal texture database. The pre-trained support vector machine (SVM) classifier inputs electromagnetic characteristics such as conductivity, complex dielectric constant, and reflection coefficient; visual characteristics such as surface roughness, grayscale entropy, edge gradient, and propagation characteristics such as multipath number and Rice factor. After data normalization, the output material type probability is output. If the metal probability exceeds 0.7, it is judged as a metal structure. For mixed metal and non-metal scenarios, the local signal distribution variance is analyzed using a signal strength heat map generated by the Kriging interpolation algorithm. The metal area exhibits a significant signal strength variance greater than 3dB due to the multipath effect, while the non-metal area has a variance less than 2dB due to penetration loss. Dynamic threshold adaptive adjustment is achieved to ensure accurate classification in complex environments.

[0136] Dynamically select one of the following enhancement strategies through the environment feature recognition module:

[0137] When a metal structure is detected, a multipath enhanced communication corridor is constructed using the waveguide effect;

[0138] When non-metallic structures are detected, penetration loss compensation and multipath diversity reception technology are used to improve signal quality;

[0139] The implementation method of the multi-band adaptive algorithm includes: establishing a frequency band selection decision function, comprehensively considering indicators such as signal-to-noise ratio, bit error rate and channel capacity; when the scene is identified as metal, designing an enhancement algorithm based on the waveguide effect of the metal structure, activating the multipath enhancement mode and adjusting the transmission power distribution according to the distance between metal surfaces; when the scene is identified as non-metal, developing a penetration loss compensation algorithm, establishing a wall attenuation model based on the Sub1GHz frequency band, and using a Kalman filter to estimate the channel impulse response to achieve time-varying channel compensation.

[0140] The implementation method of the multipath enhanced communication corridor specifically includes: when it is detected that the distance between metal surfaces meets specific conditions, activating the waveguide mode and adjusting the antenna polarization direction parallel to the metal surface, and designing a delay pre-compensation filter; when it is a non-metallic scenario, multipath diversity reception technology is adopted, and the receiving end adopts the maximum ratio combining algorithm to improve signal quality.

[0141] S5. Dynamically adjust network node deployment parameters based on real-time collected signal data and simulation results. The network nodes include core nodes and edge nodes, supporting distributed collaborative deployment. The optimal locations of multiple nodes are determined using a particle swarm optimization algorithm. Game theory models are used to balance transmission power and interference between nodes, optimizing the network topology. During the "dynamic adjustment of network node deployment parameters" process, edge computing nodes on the autonomous mobile signal acquisition network can be dynamically deployed as distributed edge nodes. Coverage blind spots are detected using real-time collected signal strength heat maps. For areas with weak signal coverage or insufficient capacity, a particle swarm optimization algorithm is used to generate multiple candidate node locations near the target area. The minimum spacing constraints between nodes, signal coverage overlap, and path loss models are comprehensively evaluated to determine the optimal node layout to fill coverage gaps. A layered network architecture is employed during deployment. Core nodes utilize high-power devices to provide global signal coverage, and their positions are optimized using gradient descent to ensure effective coverage of key areas. Edge nodes utilize low-power devices, and differentiated deployment is implemented based on environmental feature identification. For example, edge nodes are deployed in areas with dense metal equipment and waveguide enhancement mode is enabled. Multipath communication corridors are constructed by adjusting antenna polarization. In areas with dense non-metal partitions or walls, edge nodes serve as signal relay nodes, utilizing a sub-1GHz band penetration loss compensation algorithm to improve signal transmission quality across obstacles. Network performance evaluation in multi-node scenarios couples coverage range, transmission capacity, interference level, and other parameters with the number of nodes and transmit power. By constructing a fitness function that incorporates node spacing constraints and a power penalty factor, and combining it with the Nash equilibrium algorithm to solve a multi-node transmit power allocation scheme, this approach balances inter-node interference while improving coverage, avoiding signal interference caused by an increase in the number of nodes. This allows for dynamic optimization of network topology and comprehensive improvement of performance indicators.

[0142] Based on the electromagnetic field propagation simulation results of the improved OFFDTD algorithm in step S2, the theoretical RSSI value of each grid point is extracted, and the same Kriging interpolation algorithm used for real-time signal processing in S3 is used to generate a simulated RSSI signal strength heat map. Specifically, the signal strength of each discrete point in the simulated three-dimensional space is used as known data, and a semivariogram is constructed to describe the spatial correlation of the simulated signal. The interpolation weights of the unknown points are obtained by solving the Kriging equations. When the simulation scene is updated (such as equipment movement or material parameter changes), the interpolation matrix is ​​dynamically updated using Cholesky decomposition to ensure that the simulated heat map is synchronized with the real-time scene changes.

[0143] When dynamically adjusting node deployment, the real-time RSSI heat map and the simulated RSSI heat map are integrated to compare the signal coverage differences between the two (such as blind spot locations and signal strength deviations) to optimize the fitness function of the particle swarm algorithm and the utility function of the game theory model. The specific adjustments are as follows:

[0144] A new matching indicator between simulated signal coverage and real-time signal coverage has been added to the fitness function of the particle swarm optimization algorithm, reducing the deployment priority of nodes in areas where there is a large difference between simulated and actual signals.

[0145] In the power allocation of the game theory model, the multi-node interference distribution predicted by the simulation heat map is introduced to compensate for the interference error caused by dynamic changes in the environment in advance and improve the robustness of the power allocation strategy.

[0146] Specific methods for dynamically adjusting network node deployment parameters include: constructing network performance indicators that comprehensively consider network coverage, capacity, and interference; optimizing node locations using gradient descent; and developing a topology optimization algorithm based on game theory, defining a utility function, and solving the optimal power allocation through Nash equilibrium.

[0147] The network topology optimization method includes: using the particle swarm optimization algorithm, defining the particle position and velocity update formula, designing the fitness function to comprehensively consider the signal-to-interference-noise ratio and power penalty; and designing a constraint processing mechanism to apply a repulsive force when particles violate the minimum spacing constraint.

[0148] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology, characterized in that: The following steps are involved: S1. Build a 3D digital twin model of the corresponding scene, integrate building structure, equipment layout, and material property data, and establish a scene-specific material database. The dedicated material database includes but is not limited to: metal equipment, concrete walls, electromagnetic shielding layers, glass, and furniture; S2. Use the improved OFDTD algorithm to simulate electromagnetic field propagation in real time within the scene. By importing materials and using a complex dielectric constant parameterized model, dynamically simulate the signal coverage changes in multiple scenarios. S3: Deploy distributed edge computing nodes equipped with WiFi transceiver modules, integrate UWB positioning, visual SLAM and path planning algorithms, build an autonomous mobile signal acquisition network, and generate real-time RSSI signal strength heat maps; S4. Develop a multi-band adaptive algorithm based on OFDTD prediction results to intelligently select a frequency band combination, including 2.4 GHz, 5 GHz, and Sub-1 GHz. Dynamically select one of the following enhancement strategies through the environmental feature recognition module: (1) When a metal structure is detected, a multipath enhanced communication corridor is constructed using the waveguide effect; (2) When non-metallic structures are detected, penetration loss compensation and multipath diversity reception technology are used to improve signal quality; S5. Dynamically adjust network node deployment parameters based on real-time collected signal data and simulation results. The network nodes include core nodes and edge nodes, support distributed collaborative deployment, determine the optimal location of multiple nodes through particle swarm optimization algorithm, and combine game theory model to balance the transmission power and interference between nodes to optimize the network topology.

2. The multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology according to claim 1 is characterized in that: The method for constructing the scene-specific material database in step S1 specifically includes: For metal equipment, the waveguide method is used to measure the surface roughness parameter δ and complex dielectric constant ε of the metal equipment r The association model of is used to establish the modified Drude model: Where, ε r is the complex dielectric constant of the material, ε r ′ is the real part of the complex dielectric constant, ε r ″ is the imaginary part of the complex dielectric constant, ε0 is the vacuum dielectric constant, δ is the surface roughness parameter of the metal device, σ is the material conductivity, ω is the angular frequency of the electromagnetic wave, α, β and γ are experimental fitting parameters; For concrete walls, a layered dielectric model was constructed, and the relationship between humidity and equivalent dielectric constant was measured using time domain reflectometry: e eff =a+bθ-cθ 2 Where, ε eff is the equivalent dielectric constant of the concrete wall, a is the reference dielectric constant when the humidity is zero, b is the linear influence coefficient of humidity on the dielectric constant, θ is the humidity of the concrete wall, and c is the nonlinear correction coefficient of humidity on the dielectric constant; For the electromagnetic shielding layer, the transmission line method is used to establish its surface transfer impedance: Where Z T is the surface transfer impedance of the shielding layer, d is the thickness of the shielding layer, and μ is the magnetic permeability; For glass materials, the transmission line method is used to measure the signal transmission parameters of glass samples in multiple frequency bands. By analyzing the signal transmission coefficient and attenuation constant, the complex dielectric constant is inverted. For coated glass with a metal coating on the surface, it is equivalent to a layered dielectric structure combining metal layers and glass layers. The transmission matrix method, similar to the modeling of multi-layer structures of electromagnetic shielding layers, is used to calculate the complex dielectric constant to reflect its special electromagnetic properties. For wood furniture materials, a correlation model between the equivalent dielectric constant of wood and its moisture content was established. The dielectric constant data at different moisture contents were measured using time-domain reflectometry, and the humidity sensitivity coefficient was fitted to form a humidity-dependent dielectric constant expression suitable for wood. Based on the fiber orientation characteristics of solid wood panels, the concept of tensor dielectric constant was introduced to address the coupling effect between the polarization direction of electromagnetic waves and the direction of wood grain. For plastics and composite materials, the resonant cavity method is used to measure the complex dielectric constant at different frequencies, and a relationship model between frequency and the real and imaginary parts of the dielectric constant is established through polynomial fitting. For composite materials containing metal particles or glass fiber fillers, the Maxwell-Garnett effective medium theory is used to calculate their equivalent dielectric constant to reflect the influence of the filler on the electromagnetic properties of the material.

3. The multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology according to claim 2 is characterized in that: The modeling method of the electromagnetic shielding layer further includes: Establish shielding effectiveness: Where E0 is the incident electric field intensity, E1 is the transmitted electric field intensity, is the reflection coefficient, is the decay constant; For a multi-layer shielding structure, the overall shielding effectiveness is calculated using the transfer matrix method: Where, is the propagation constant, is the characteristic impedance.

4. The multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology according to claim 1 is characterized in that: The specific implementation of the improved O-FDTD algorithm described in step S2 includes: Introduce a dynamic time step adjustment mechanism based on the traditional FDTD algorithm: Where Δt is the time step, the size of the Δx, Δy, Δz spatial grid, c is the speed of light in vacuum, σ max is the maximum conductivity in the current calculation domain; Implement the sub-grid technology for the edges of metal devices. Insert an n×n×n sub-grid (n = 5) within the standard grid (Δx, Δy, Δz), and use the explicit-implicit hybrid iterative method to solve the discrete form of the Maxwell curl equation: Where H is the magnetic field strength, E is the electric field strength, and J s is the surface current density; Establish a dynamic boundary condition update algorithm for moving objects, and use the moving coordinate system transformation method to handle the sudden change of field quantities caused by object movement.

5. The multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology according to claim 1 is characterized in that: The construction method of the autonomous mobile signal acquisition network described in step S3 specifically includes: Design a dual-frequency UWB positioning module to transmit a Chirp spread spectrum signal with a pulse width τ = 0.5 ns: In the formula, τ = 0.5 ns is the pulse width of the signal, f0 = 3.5 GHz is the starting frequency of the signal, and k = 20 MHz / ns is the frequency sweep slope; Implement three-dimensional positioning through the time difference of arrival algorithm, and the solution equation is: Where (x, y, z) is the coordinate of the target, (x i ,y i ,z i ) is the coordinate of the i-th anchor point, Δt ij is the time difference between the arrival of the signal at anchor points i and j; The visual SLAM system uses the ORB feature extraction and GMS matching algorithms to construct a photometric error function: Where E is the total photometric error, I i (p) is the pixel value of the i-th image at the feature point p, I j (p) is the pixel value of the j-th image at feature point p, is the distortion function, ξ∈se(3) is the pose represented by Lie algebra; The path planning uses the improved A* algorithm, and the cost function: Where g(n) is the actual cost from the starting point to node n, h(n) is the heuristic estimated cost from node n to the end point, λ is the weight coefficient, and RSSI k Signal strength at the path point, RSSI th is the preset signal strength threshold.

6. The multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology according to claim 4 is characterized in that: The method for generating the real-time signal intensity heat map further includes: Use the Kriging interpolation algorithm to estimate the RSSI value of unknown points and construct a semi-variogram: In the formula, h is the spatial lag distance, and z(x) is the RSSI value at position x; Solve the Kriging equations: [Γ][λ] = [γ0] In the formula, Γ is the semi-variogram matrix of known points, and γ0 is the semi-variogram vector of the point to be estimated; Design a dynamic update mechanism. When new measurement data arrives, use Cholesky decomposition to update the inverse matrix Γ -1 , the update formula is: L new =cholupdate(L,Δγ,′+′) Where, L new is the updated lower triangular Cholesky factor matrix, L is the original Cholesky decomposition lower triangular matrix, Δγ is the semi-variant increment vector between the new data point and the existing data point, and + indicates that an additive update is to be performed.

7. The multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology according to claim 1 is characterized in that: The implementation method of the multi-band adaptive algorithm described in step S4 includes: Establish a frequency band selection decision function: F=w1·SNR+w2·BER -1 +w3·C Where w1, w2, w3 are the weights of each indicator, SNR = 10log(P r / N0) signal-to-noise ratio, is the bit error rate, C = Blog2(1+SNR) channel capacity, P r is the received power, N0 is the noise power spectral density, and B is the channel bandwidth; Dynamically select the following enhancement strategies according to the environmental feature recognition result: When it is recognized as a metal scene, design a waveguide effect enhancement algorithm based on the metal structure. When the detected distance d between parallel metal surfaces satisfies When , activate the multipath enhancement mode and adjust the transmit power distribution to: Where, P t is the total transmission power, N is the number of paths participating in power allocation, P i is the transmission power of the i-th path, d i is the propagation distance of the i-th path, j is the imaginary unit; When it is recognized as a non-metal scene, develop a penetration loss compensation algorithm and establish a wall attenuation model based on the Sub-1GHz frequency band: Where, P comp is the power after penetration loss compensation, P t is the transmission power, α wall is the attenuation coefficient of the wall, which describes the attenuation degree of the wall to the electromagnetic wave, d is the thickness of the wall through which the electromagnetic wave passes, is the transmittance coefficient, f i is the electromagnetic wave frequency corresponding to the i-th material, is the i-th material; Develop a time-varying channel compensation algorithm and use the Kalman filter to estimate the channel impulse response: Where h(t,τ) is the impulse response of the time-varying channel, α k (t) is the time-varying complex gain, τ k (t) is the time delay, δ(τ-τ k (t)) is the Dirac delta function, which represents the impulse response with a time delay of τ k Discrete impulse at (t).

8. The multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology according to claim 1 is characterized in that: The implementation method of the multipath enhanced communication corridor described in step S4 specifically includes: Dynamically select the following enhancement strategies according to the environmental feature recognition result: The metal scene is: when the metal surface distance d is detected to meet When , the waveguide mode is activated and the polarization direction of the antenna is adjusted to be parallel to the metal surface; Design a delay pre-compensation filter: Where, is the delay of the nth reflection, a n =Γ n (1-Γ 2 ) is the filter coefficient; Among them, the non-metal scene is: adopt the multipath diversity reception technology, and the receiver adopts the maximum ratio combining algorithm: Where y is the total signal after the receiver combines, h m is the channel response of the mth path, |h k | 2 is the square of the magnitude of the channel response of the kth path.

9. The multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology according to claim 1 is characterized in that: The specific method for dynamically adjusting the network node deployment parameters described in step S5 includes: Construct network performance metrics: J = α·Coverage + β·Capacity - γ·Interference In the formula, α, β, γ are the weight coefficients of each indicator. For network coverage, is the network capacity, H is the step function; Gradient descent method is used to optimize the node position (x i ,y i ,z i ), the parameter update formula is: In the formula, the learning rate η = 0.01; Develop a topology optimization algorithm based on game theory and define a utility function: Where U i is the utility value of node i, γ i is the signal-to-interference-and-noise ratio of node i, p i is the transmission power of node i, h ij is the channel coefficient from node j to node i, and the optimal power allocation is solved through Nash equalization.

10. The multi-scenario wireless network rapid deployment method based on electromagnetic field digital twin technology according to claim 1 is characterized in that: The network topology structure optimization method described in step S5 includes: Using the particle swarm optimization algorithm, define the particle position X i =[x1,y1,z1,…,x N ,y N ,z N ], speed update formula: Where w is the inertia weight, c1, c2 are acceleration constants, r1, r2 are random numbers, and pbest i is the historical optimal position of particle i, gbest is the global optimal position; Fitness function: Where, is the signal-to-interference-and-noise ratio of the mth link, α w d m is the path loss, ρ is the power penalty factor; Design a constraint handling mechanism. When a particle violates the minimum distance constraint ||Xi - Xj|| < dmin, apply a repulsive force: Where, d min is the minimum spacing between nodes, and K is the elastic coefficient.

Images