Inland river scene wireless channel lightweight modeling method based on multipath energy ratio

Abstract

The invention provides an inland river scene wireless channel lightweight modeling method based on a multipath energy ratio, and the method comprises the following steps: S1, constructing a 1: 1 three-dimensional model of an inland river scene through fusing a satellite map and a design drawing, and endowing ground features with corresponding electromagnetic parameters; s2, a dynamic convergence threshold value and an energy contribution threshold value are adopted to quantify multipath energy proportions of various ground features, and key ground features are screened; s3, dynamically adjusting the modeling range and the scene complexity based on the energy ratio; s4, counting the multi-path number proportion and energy distribution of each type of ground features, calculating single multi-path receiving power and total energy contribution, and providing data support for channel modeling; and S5, configuring simulation parameters, performing ray tracing simulation, optimizing a noise threshold value, and verifying the accuracy of multipath time delay and energy distribution through actually measured data. Through the method, classification statistics is effectively carried out on various ground feature types, so that scene settlement acceleration simulation is simplified, and the accuracy of scene modeling simulation is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ray tracing simulation, and in particular to a lightweight modeling method for wireless channels in inland river scenarios based on multipath energy proportion. Background Art

[0002] Ray tracing is a technology used to simulate the propagation and reflection of electromagnetic waves in three-dimensional scenes. In the field of wireless communications, ray tracing is widely used to establish and analyze wireless channel models to evaluate signal propagation characteristics, multipath effects, and signal strength distribution. By calculating phenomena such as reflection, refraction, and scattering of electromagnetic waves when encountering different media, it is possible to predict the signal propagation path and its changes over time and space. Ray tracing technology is particularly suitable for use in inland waterways with complex terrain and multiple obstacles. The design of inland ship communication systems relies on accurate modeling of the propagation characteristics of electromagnetic waves in structures such as water surfaces, riverbanks, bridges, and buildings. Ray tracing can provide theoretical support for system optimization by simulating these complex propagation paths.

[0003] However, in the study of multipath distribution of wireless channels in inland river scenarios, accurately obtaining the multipath quantity and energy proportion of each ground object is a major problem. In large-scale outdoor simulations, the types of ground objects are numerous and complex, which makes it difficult to quantitatively analyze the multipath generated by various ground objects. Therefore, it is urgent to propose a lightweight modeling method for wireless channels in inland river scenarios to effectively classify and count various ground object types, so as to simplify the scene sedimentation and accelerate the simulation, and increase the accuracy of scene modeling and simulation using SketchUp software. Summary of the Invention

[0004] The purpose of the present invention is to solve the technical problems in the above background and propose a lightweight modeling method for wireless channels in inland river scenarios based on multipath energy proportion, including the following steps:

[0005] S1. By integrating satellite maps and design drawings, a 1:1 3D model of the inland river scene is constructed, and corresponding electromagnetic parameters of the ground objects are assigned;

[0006] S2. Use dynamic convergence threshold and energy contribution threshold to quantify the multipath energy proportion of various types of objects and screen key objects;

[0007] S3, dynamically adjust the modeling scope and scene complexity based on energy proportion;

[0008] S4. Count the proportion and energy distribution of multipath paths of various types of objects, calculate the received power and total energy contribution of a single multipath path, and provide data support for channel modeling.

[0009] S5. Configure simulation parameters and perform ray tracing simulation, optimize the noise threshold, and verify the accuracy of multipath delay and energy distribution through measured data.

[0010] In a preferred solution, step S1 further includes the following steps:

[0011] S11. Combine Google's high-precision 3D geographic information map and design drawings to obtain geographic data of the inland river scene.

[0012] S12. Use SketchUp software to create a 1:1 3D model of a typical inland waterway of the Yangtze River, including water areas, bridges, ships, and riverside buildings.

[0013] In a preferred embodiment, step S2 includes the following steps:

[0014] S21. defining a dynamic adjustment rule for the reflection order;

[0015] S22, setting a scattering order energy ratio threshold, and calculating the electric field intensity ratio of each order scattering path;

[0016] S23. Quantify the multipath energy contribution of various types of ground objects through pre-simulation or historical data.

[0017] S24. Retain key features whose energy contribution values ​​are greater than a set threshold, and eliminate features whose energy contribution values ​​are less than the set threshold.

[0018] In a preferred solution, in step S3, dynamically adjusting the scene modeling range specifically includes:

[0019] The modeling radius is set differently according to the channel structure and multipath distribution characteristics:

[0020] Open waters without bridges: The modeling range is limited to a spherical area with a radius of 500m centered on the receiving point.

[0021] Bridge impact area: Modeling range extended to a radius of 1000m to cover high-order reflection paths (secondary reflections from bridge piers account for >35%)

[0022] In a preferred embodiment, step S4 includes the following steps:

[0023] S41. Statistics on the proportion of multipath by ground object type:

[0024]

[0025] Where: C i is the ratio of the total number of multipath components of the i-th type of ground object to the total number of multipath components; K(i) represents the number of multipath components generated by the i-th type of ground object; Q is the total number of multipath components of all ground object types;

[0026] S42. Calculate the single multipath received power based on the transmit power, antenna gain, free space path loss, and reflection point power loss:

[0027] P r =P t ·G t ·G r ·L free U;

[0028] Where: P t is the transmission power, G t is the transmitting antenna gain, G r is the receiving antenna gain, L free is the free space path loss, U is the power loss at the reflection point;

[0029] S43. Sum the received power of multipath rays for each type of ground object to obtain the total energy:

[0030]

[0031] Where: P r,k Indicates the received power value of all multipath rays at this location.

[0032] S44. Calculate the energy proportion and normalized average energy contribution rate of each type of ground feature:

[0033] Multipath energy proportion statistics: Calculate the proportion of the total multipath energy generated by the i-th type of environmental factor to the total multipath energy:

[0034]

[0035] Where: P i is the ratio of the total multipath energy of the i-th ground object type to all multipath energies; E(i) is the total multipath energy of the i-th ground object type; T is the sum of the energies of all multipaths;

[0036] Normalized average energy contribution rate statistics: Calculate the ratio of the average multipath energy of the i-th environmental factor to the total average energy of all factors:

[0037]

[0038] Where: R i is the proportion of the average multipath energy of the i-th ground object type; E i is the average multipath energy of the i-th ground object type; I is the total number of ground object types.

[0039] In the preferred solution, the power loss factor at the reflection point is calculated by introducing the vertical polarization reflection coefficient Γ to calculate U:

[0040]

[0041] U=|Γ| 2 ;

[0042] Where: Γ represents the vertical polarization reflection coefficient, θ i is the angle of incidence, θ t is the refraction angle, η1 and η2 are the intrinsic impedances of the two media.

[0043] In a preferred embodiment, step S5 includes the following steps:

[0044] S51 dynamically adjusts the reflection and scattering orders based on the scenario; filters out noise interference to ensure the accuracy of multipath energy analysis, especially in scenarios with moving ships and complex bridge structures;

[0045] S52. Verify the dominance of water surface reflection and the impact of bridges on multipath energy through ray tracing simulation snapshots and multipath energy proportion statistics;

[0046] S53. Verify the multipath delay spread error and power delay spectrum fitting by actual measurement.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] (1) For the first time, a dynamic feature screening mechanism based on multipath energy contribution weights is proposed, breaking through the redundancy bottleneck of traditional full-factor modeling. By quantitatively analyzing the energy contribution of key paths such as water surface reflection, bridge scattering, and ship scattering in inland river scenes, a "energy threshold-modeling range" mapping relationship is constructed to accurately eliminate low-contribution features, significantly reducing the complexity of the three-dimensional scene model and fundamentally solving the problem of ray tracing computational overload in large-scale water environments.

[0049] (2) The dynamic convergence threshold method and the energy contribution threshold method are innovatively integrated to achieve adaptive control of the reflection and scattering orders. The synergistic effect of the dynamic adjustment of the reflection order and the intelligent truncation of the scattering order reduces the amount of simulation calculations, which is better than the traditional fixed-order method.

[0050] (3) Establishing the relationship between ground object electromagnetic parameters (dielectric constant, roughness) and multipath energy distribution. For example, the proportion of scattering paths directly associated with the surface roughness of bridge concrete (σ = 0.32λ) increased to 28%, providing physical interpretability support for channel modeling and significantly improving the robustness of the model in complex dynamic scenarios (such as ships crossing bridges). BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 Schematic diagram for modeling an inland river scene.

[0052] Figure 2 This is a model diagram of the bridge structure.

[0053] Figure 3 This is a model picture of a yacht.

[0054] Figure 4 This is the multipath energy analysis flow chart (including dynamic threshold judgment module).

[0055] Figure 5 Comparison of ray tracing simulation snapshots (without bridge / crossing bridge / after crossing).

[0056] Figure 6 This is a statistical chart of multipath energy proportion (water surface / bridge / ship / building).

[0057] Figure 7 This is the power delay distribution diagram. DETAILED DESCRIPTION

[0058] This paper proposes a ray tracing-based method. Taking a typical inland waterway of the Yangtze River as the research object, the method combines Google's high-precision 3D geographic information map with SketchUp software to create a 3D model of the inland waterway scene, restoring the main objects that affect ray propagation. The modeled scene primarily includes the inland waterway, buildings on both sides of the river, a cross-river bridge, ship scatterers, river embankments, and greenery. The bridge structural model features a double-layer concrete deck for vehicles and pedestrians, with large, uniformly sized circular concrete piers evenly distributed beneath the deck, representing a typical cross-river bridge in inland waters.

[0059] In ray tracing simulations, the constructed scene model directly determines the physical realism of ray propagation, the reliability of rendering results, and computational efficiency. Before ray tracing simulations can be performed, an accurate scene environment model is required to support the entire simulation. Each material in the scene is assigned corresponding electromagnetic parameters, which are derived from known literature. The accuracy of the scene model not only affects the realism of multipath propagation during the simulation but is also closely related to channel characteristics. Accurate multipath propagation is the focus of this article, as it directly affects the simulation's computational efficiency and multipath energy distribution. Therefore, achieving a 1:1 reproduction of the actual Yangtze River environment model is crucial.

[0060] The core of this invention is to achieve a balance between accuracy and efficiency by dynamically optimizing the scene modeling complexity and ray propagation order based on the ground feature screening criterion of multipath energy proportion, including the following steps:

[0061] First, high-precision three-dimensional scene modeling is carried out. By integrating satellite maps and design drawings, a 1:1 three-dimensional model of inland waters, bridges, ships, and riverbank buildings is constructed; the electromagnetic parameters of the ground objects (concrete bridge materials with a dielectric constant ε≥15 and a loss tangent tanδ≤0.02) are assigned.

[0062] Furthermore, in the multipath energy contribution analysis, a dynamic convergence threshold is defined: when the path loss difference rate ΔPL of adjacent reflection orders is less than 5%, the order is stopped from increasing (Formula 1);

[0063]

[0064] Where: PL n is the path loss under n-th order reflection, PL n-1 is the path loss under the n-1th order reflection.

[0065] In ray tracing simulations, to eliminate the impact of scattering on the reflection order setting, the propagation mechanisms in the simulation process are first set to direct radiation and reflection. The path loss values ​​of orders one to five are calculated respectively, and the difference rate of path losses of adjacent orders is given. When the difference rate of path losses of adjacent orders ΔPL is less than 5%, the received power simulation error is less than 1.5dB. The 5% threshold can cover more than 98% of the effective energy, meeting the Class A equipment certification requirements specified in JT / T 1194-2023.

[0066] Furthermore, the energy contribution threshold is set: if the scattering order energy accounts for E n If <5%, the order is ignored (Formula 2).

[0067]

[0068] Where: E scat,n is the complex electric field intensity of the nth-order scattering path, E total The total electric field strength of all order scattering paths (1st order, 2nd order, nth order, etc.).

[0069] When setting the scattering order, the propagation mechanism in the simulation process is set to direct and scattering, and the received signal power values ​​of orders one to five are calculated respectively, analogous to reflection.

[0070] Then, a lightweight scenario was constructed, screening key landforms (water surface, ships, bridges, riverside buildings) with an energy share greater than 1%, and eliminating low-energy landforms (remote green space, muddy land); the modeling range was dynamically adjusted: in areas without bridges, landforms within a radius of 500m were retained, and the bridge influence area was expanded to 1000m.

[0071] To achieve coordinated optimization of multipath simulation accuracy and computational efficiency, the present invention dynamically adjusts the scene modeling range based on multipath energy contribution, including the following steps:

[0072] (1) Key feature screening and lightweight modeling

[0073] Through pre-simulation or historical data analysis, quantify the multipath energy contribution of various environmental factors. Set the energy contribution threshold α = 1% and perform ground feature screening:

[0074] Retain key features: filter energy contribution value P r Elements ≥α (including water surface, ships, main structures of bridges, and riverbank buildings);

[0075] Eliminate low-contribution features: remove energy contribution value P r Elements <α (such as remote green space, muddy land);

[0076] Technical effect: By eliminating low-energy contributing objects, the model complexity is reduced by about 40%-60%, significantly improving the efficiency of ray tracing calculations.

[0077] (2) Dynamic adjustment of spatial range

[0078] The modeling radius is set differently according to the channel structure and multipath distribution characteristics:

[0079] Open waters without bridges: The modeling range is limited to a spherical area with a radius of R1 = 500m centered at the receiving point;

[0080] Bridge influence area: The modeling range is extended to a spherical area with a radius of R2 = 1000m centered at the receiving point;

[0081] Basis: The bridge structure causes an increase in the order of multipath reflections (measured data show that the secondary reflection path of the pier accounts for >35%), and the modeling scope needs to be expanded to cover higher-order reflection paths.

[0082] (3) Modeling scope switching mechanism

[0083] Define the trigger conditions for the bridge influence area: when the Euclidean distance D between the receiving point and any bridge component is ≤ 2L (L is the maximum span of the bridge), the modeling radius R2 = 1000m is automatically enabled; otherwise, the radius R1 = 500m is used.

[0084] In order to clarify the spatial source attributes of multipath signals in the complex environment of inland rivers, the proportion of multipath generated by each environmental factor category is counted:

[0085]

[0086] Where: C i is the ratio of the total number of multipath components of the i-th type of ground object to the total number of multipath components; K(i) represents the number of multipath components generated by the i-th type of ground object; Q is the total number of multipath components of all ground object types;

[0087] For the correlated multipath data, calculate the total received energy generated by various environmental factors:

[0088] Calculation of single multipath receiving power: For each multipath ray belonging to a specific environmental factor, its receiving power P r Calculate as follows:

[0089] P r =P t ·G t ·G r ·L free U;

[0090] Where: P t is the transmission power, G t is the transmitting antenna gain, G r is the receiving antenna gain, L free is the free space path loss, and U is the power loss at the reflection point.

[0091] Calculation of power loss factor at the reflection point: Introduce the vertical polarization reflection coefficient Γ to calculate U:

[0092]

[0093] U=|Γ| 2 ;

[0094] Where: Γ represents the vertical polarization reflection coefficient, θ i is the angle of incidence, θ t is the refraction angle, η1 and η2 are the intrinsic impedances of the two media.

[0095] Calculation of total energy of environmental factors: For all N multipath rays belonging to the i-th type of environmental factors, calculate their total received energy P total :

[0096]

[0097] Where: P r,k Indicates the received power value of all multipath rays at this location.

[0098] Calculate the contribution ratio of various environmental factors to multipath energy and the normalized average energy:

[0099] Multipath energy proportion statistics: Calculate the proportion of the total multipath energy generated by the i-th type of environmental factor to the total multipath energy:

[0100]

[0101] Where: P i is the ratio of the total multipath energy of the i-th ground object type to all multipath energies; E(i) is the total multipath energy of the i-th ground object type; T is the sum of the energies of all multipaths.

[0102] Normalized average energy contribution rate statistics: Calculate the ratio of the average multipath energy of the i-th environmental factor to the total average energy of all factors:

[0103]

[0104] Where: R i is the proportion of the average multipath energy of the i-th ground object type; E i is the average multipath energy of the i-th ground object type; I is the total number of ground object types.

[0105] During the ray tracing simulation, the path loss model used is:

[0106] L river_dB =L free +L interf +L diff +X σ ;

[0107] Where, L free is the free space loss, L interf is the interference correction, L diff is the diffraction loss, X σ For shadow fading.

[0108] For free space loss L free The expression is:

[0109] L free =32.44+20log 10 d+20log 10 f c ;

[0110] in: PL dB is the spatial propagation loss; d is the propagation distance between the transmitting and receiving antennas, in km; is the carrier frequency, in MHz.

[0111] For the interference correction L interf calculate:

[0112]

[0113] Where Γ is the water surface reflection coefficient, h t is the height of the transmitting antenna from the water surface, in m, h r is the height of the receiving antenna above the water surface, in meters, λ is the wavelength, in meters, d is the critical distance, in meters, and C is the empirical calibration constant, in dB.

[0114] For the diffraction loss L diff calculate:

[0115]

[0116] Where W is the width of the river, in meters, and W ref is the reference river width, ΔL diffis the diffraction loss coefficient.

[0117] Use the Fresnel formula to calculate Γ for different antenna polarization modes. When the antenna is vertically polarized:

[0118]

[0119] When the antenna is horizontally polarized:

[0120]

[0121] Where θ i is the angle of incidence, ε r is the relative complex permittivity of the material.

[0122] This embodiment will be further described below with reference to the accompanying drawings.

[0123] exist Figure 1 Accurate modeling of the Yangtze River waters is particularly important. Realistic modeling represents accurate multipath propagation paths, which is directly relevant to studying multipath distribution. For modeling the Yangtze River basin, the waters were uniformly modeled as planes, overlaid with high-precision satellite imagery obtained from Google Earth as a texture reference, resulting in a model of the Yangtze River waters for the entire simulation area. Finally, the corresponding material electromagnetic parameters (dielectric constant, loss tangent, scattering coefficient, and effective roughness) obtained from the literature are included.

[0124] exist Figure 2 During the simulation, when a ship carrying a transmitting antenna slowly crosses a bridge, the bridge's complex concrete structure creates even more complex multipath effects. To obtain more realistic and effective multipath information, recreating the bridge's structural model becomes crucial. A complete bridge structural model was constructed based on the bridge's design drawings, and the ship model used in the simulation was recreated based on the size of Yangtze River yachts.

[0125] Inland river environments, due to the presence of large expanses of water and bridge structures, generate numerous high-latency reflection paths and weak scattering paths, posing a challenge to noise threshold setting. Setting a high noise threshold may filter out multipath reflected from distant water surfaces, affecting the settings of the reflection and scattering orders, and consequently, the analysis of channel characteristics.

[0126] When using multipath energy analysis to simplify inland river scenario modeling, the simulated multipath delay distribution values ​​are first filtered to remove the effects of noise in order to quantitatively analyze the multipath energy distribution. For each snapshot of the simulated power delay distribution, the noise threshold filtering method developed in the inland river environment analysis is applied to each snapshot to remove the noise interference and obtain accurate multipath propagation information for the inland river environment.

[0127] During the simulation, the script can be used to set the ship's travel path in the inland river and the dynamic situation of the ships coming and going. The movement of the ships will also affect the noise threshold setting. Figure 6 As can be seen in the simulation snapshots, as the transmitting antenna moves from an area without bridge influence to crossing a bridge, multipath from water reflections consistently accounts for a significant portion of the signal. Therefore, when setting the noise threshold, it is appropriate to lower the threshold to obtain more accurate multipath propagation information.

[0128] exist Figure 4 Among them, in the research method of multipath energy distribution in inland river scenes, the dynamic convergence threshold method and the energy contribution threshold method are first used in synergy to achieve the optimal selection of reflection and scattering orders, which facilitates the subsequent statistics and analysis of multipath in inland river scenes. Then, based on the ray multipath feature extraction, the total number of paths and energy distribution are obtained; the rationality of the order selection is verified by the multipath energy ratio, and the key scatterers are screened in combination with the electromagnetic characteristics of the ground objects, and finally the order configuration that takes into account both accuracy and efficiency is output. The innovation of this process lies in the deep coupling of environmental physical characteristics with electromagnetic propagation mechanisms. It is suitable for ray tracing simulation of high-frequency complex scenes such as 5.9GHz inland river bridges, while maintaining the path loss error <1.5dB, which is significantly better than the traditional random order selection method.

[0129] In open inland waterways, the main propagation mechanisms include line-of-sight propagation, transmission, 0-5th-order reflections, and 0-5th-order scattering. In inland waterways, reflection paths primarily originate from specular reflections on the water surface, reflections from iron hulls, and reflections from tall buildings along the riverbank. Energy loss at the reflection point is relatively low in these reflections. The order of the reflection path is determined using a dynamic convergence threshold method. Path losses (PL) for reflections of order n through n+1 are calculated, and the PL difference between adjacent orders is calculated.

[0130] exist Figure 5 Ray tracing simulation results show that in the area without bridge influence, the simulated reflection and scattering orders are set to 3 and 2, respectively. When crossing a bridge, the simulated reflection and scattering orders are set to 4 and 2, respectively. After crossing the bridge, the simulated reflection and scattering orders are set to 3 and 2, respectively. Based on the above simulation configuration, the results of a snapshot are shown for the area without bridge influence, when crossing a bridge, and after crossing a bridge. The figure shows that in all three cases, multipath from the water surface always dominates.

[0131] exist Figure 6In the unaffected area, the average multipath energy in the inland waters is highest. This is because both the receiving and transmitting antennas are close to the river surface, resulting in shorter paths after multipath reflection and lower losses. Therefore, the average multipath energy is highest. At this point, the average multipath energy from buildings and river embankments is lowest. Because the buildings and embankments on both sides are already 1000 meters away from direct radiation, significant path loss occurs, causing a decrease in multipath energy. Before crossing the bridge, the average multipath energy is still highest on the water surface. As the multipath contribution from the bridge increases, the proportion of multipath energy from the water surface decreases, but still accounts for the largest portion. The average multipath energy contribution from the bridge further increases, and the average multipath energy from both sides of the river is now almost negligible.

[0132] according to Figure 7 The power delay distribution simulation results shown in the figure show that to obtain more accurate multipath information, the average noise power in this scenario is analyzed as Pn. The power of the direct path in the power delay spectrum is recorded as Pmax (the peak in the figure). Based on this, the noise threshold is reduced by 35dB. This is 16.8dB higher than the average noise power (-110.66dB). This effectively filters noise and maintains a reasonable dynamic range greater than 35dB. The remaining part above the noise threshold is used for multipath energy calculation.

[0133] To effectively calculate the noise threshold, it is necessary to ensure that there are enough noise sample values ​​for calculation. Starting from the time delay greater than 6500ns, a noise value is taken every 50ns to calculate the noise threshold P. threshold First, convert each power value into logarithmic units (dBm) P dBm Convert to linear unit (mW)P mW , the calculation formula is:

[0134]

[0135] The average noise power in linear units is then calculated as:

[0136]

[0137] Where K is the number of noise samples.

[0138] Then convert it back to logarithmic units (dBm) as follows:

[0139] P n =10log 10 (μ mW );

[0140] Where n is a constant (usually 3) to cover 99.7% of noise fluctuations.

[0141] Then, the standard deviation of the noise samples is calculated as:

[0142]

[0143] The final noise threshold is: P threshold =P n +nσ.

[0144] Under equivalent hardware conditions, simulation time was shortened from 12.6 hours using traditional methods to 6.2 hours, while GPU memory usage was reduced by 18%, meeting the real-time requirements of shipboard embedded devices. Dynamic scene simplification reduced model data size by 45%, making it suitable for low-bandwidth transmission and cloud-based collaborative computing. Multipath delay spread error was ≤8.3ns, and the power delay profile fit reached 92%, supporting high-precision channel prediction in the 5.9GHz band. Results are shown in Figure 1. Field measurements have been completed and validated in a typical section of the Yangtze River waterway, adapting to various bridge structures and ship density scenarios.

[0145] Table 1 Comparison of the effects of traditional method and optimized method under the same hardware conditions

[0146]

[0147]

[0148] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A lightweight modeling method for wireless channels in inland river scenarios based on multipath energy proportion, characterized by: The following steps are involved: S1. By integrating satellite maps and design drawings, a 1:1 3D model of the inland river scene is constructed, and corresponding electromagnetic parameters of the ground objects are assigned; S2. Use dynamic convergence threshold and energy contribution threshold to quantify the multipath energy proportion of various types of objects and screen key objects; S3, dynamically adjust the modeling scope and scene complexity based on energy proportion; S4. Count the proportion and energy distribution of multipath paths of various types of objects, calculate the received power and total energy contribution of a single multipath path, and provide data support for channel modeling. S5. Configure simulation parameters and perform ray tracing simulation, optimize the noise threshold, and verify the accuracy of multipath delay and energy distribution through measured data.

2. The lightweight modeling method for wireless channels in inland river scenarios based on multipath energy proportion according to claim 1 is characterized by: Step S1 further includes the following steps: S11. Combine Google's high-precision 3D geographic information map and design drawings to obtain geographic data for the inland river scene; S12. Use SketchUp software to create a 1:1 3D model of a typical inland waterway of the Yangtze River, including water areas, bridges, ships, and riverside buildings.

3. The lightweight modeling method for wireless channels in inland river scenarios based on multipath energy proportion according to claim 1 is characterized by: Step S2 includes the following steps: S21. defining a dynamic adjustment rule for the reflection order; S22, setting a scattering order energy ratio threshold, and calculating the electric field intensity ratio of each order scattering path; S23. Quantify the multipath energy contribution of various types of ground objects through pre-simulation or historical data; S24. Retain key features whose energy contribution values ​​are greater than a set threshold, and eliminate features whose energy contribution values ​​are less than the set threshold.

4. The lightweight modeling method for wireless channels in inland river scenarios based on multipath energy proportion according to claim 1 is characterized by: In step S3, the scene modeling range is dynamically adjusted, specifically including: The modeling radius is set differently according to the channel structure and multipath distribution characteristics: Open waters without bridges: The modeling range is limited to a spherical area with a radius of 500m centered on the receiving point; Bridge influence area: The modeling range is extended to a radius of 1000m to cover high-order reflection paths.

5. The lightweight modeling method for wireless channels in inland river scenarios based on multipath energy proportion according to claim 1 is characterized by: Step S4 includes the following steps: S41. Statistics on the proportion of multipath by ground object type: Where: C i is the ratio of the total number of multipath components of the i-th type of ground object to the total number of multipath components; K(i) represents the number of multipath components generated by the i-th type of ground object; Q is the total number of multipath components of all ground object types; S42. Calculate the single multipath received power based on the transmit power, antenna gain, free space path loss, and reflection point power loss: P r =P t ·G t ·G r ·L free ·U; Where: P t is the transmission power, G t is the transmitting antenna gain, G r is the receiving antenna gain, L free is the free space path loss, U is the power loss at the reflection point; S43. Sum the received power of multipath rays for each type of ground object to obtain the total energy: Where: P r,k Indicates the received power value of all multipath rays at this location; S44. Calculate the energy proportion and normalized average energy contribution rate of each type of ground feature: Multipath energy proportion statistics: Calculate the proportion of the total multipath energy generated by the i-th type of environmental factor to the total multipath energy: Where: P i is the ratio of the total multipath energy of the i-th ground object type to all multipath energies; E(i) is the total multipath energy of the i-th ground object type; T is the sum of the energies of all multipaths; Normalized average energy contribution rate statistics: Calculate the ratio of the average multipath energy of the i-th environmental factor to the total average energy of all factors: Where: R i is the proportion of the average multipath energy of the i-th ground object type; E i is the average multipath energy of the i-th ground object type; I is the total number of ground object types.

6. The lightweight modeling method for wireless channels in inland river scenarios based on multipath energy proportion according to claim 5 is characterized by: Calculation of power loss factor at the reflection point: Introduce the vertical polarization reflection coefficient Γ to calculate U: U=|Γ| 2 ; Where: Γ represents the vertical polarization reflection coefficient, θ i is the angle of incidence, θ t is the refraction angle, η1 and η2 are the intrinsic impedances of the two media.

7. The lightweight modeling method for wireless channels in inland river scenarios based on multipath energy proportion according to claim 1 is characterized by: Step S5 includes the following steps: S51 dynamically adjusts the reflection and scattering orders based on the scenario; filters out noise interference to ensure the accuracy of multipath energy analysis, especially in scenarios with moving ships and complex bridge structures; S52. Verify the dominance of water surface reflection and the impact of bridges on multipath energy through ray tracing simulation snapshots and multipath energy proportion statistics; S53. Verify the multipath delay spread error and power delay spectrum fitting by actual measurement.

Images