Deployment Methods for Mobile Electromagnetic Spectrum Monitoring Equipment Integrating Multiphysics Sensing

Abstract

This invention provides a deployment method for mobile electromagnetic spectrum monitoring equipment integrating multi-physics sensing, relating to the field of electromagnetic spectrum monitoring technology. The method includes: performing electromagnetic propagation modeling with multi-physics coupling; performing Pareto front screening to obtain H global candidate solutions; S1: performing diversity-enhanced global exploration and outputting a set of diverse potential solutions; S2: performing constraint-aware local optimization and outputting a set of compliant physical field deployment schemes; iteratively executing S1 to S2, performing two-layer physical field collaborative iterative optimization until convergence conditions are met, and outputting a multi-physics constrained deployment coordinate set; performing robustness verification, outputting an anti-interference optimized deployment scheme, and deploying the monitoring network. This solves the technical problem that existing spectrum monitoring technologies mostly rely on fixed monitoring point deployments, where the monitoring range of these devices is statically set, lacking dynamic adaptability, resulting in incomplete signal coverage and ineffective monitoring of electromagnetic interference sources.

Description

Technical Field

[0001] This invention relates to the field of electromagnetic spectrum monitoring technology, and in particular to a method for deploying mobile electromagnetic spectrum monitoring equipment that integrates multi-physics sensing. Background Technology

[0002] With the rapid development of wireless communication technology and the popularization of various electronic devices, the demand for electromagnetic spectrum usage is constantly increasing. In order to ensure the effective use of the spectrum and avoid electromagnetic interference, electromagnetic spectrum monitoring has become an important part of modern communication network management. In the civilian field, especially in densely populated urban areas, commercial areas and sensitive areas such as hospitals and schools, effective spectrum monitoring can help manage and control the use of radio spectrum, ensure communication quality, and reduce the risk of electromagnetic interference and privacy leakage.

[0003] However, traditional spectrum monitoring mostly relies on fixed monitoring point deployments. The monitoring range and coverage of these devices are usually statically set, lacking dynamic adaptability. This makes them unable to effectively cope with dynamic changes in spectrum usage within the target area, such as changes in the number of devices or advancements in communication technology. As environmental conditions, terrain, weather, and other factors change, these devices cannot adjust their deployment plans in a timely manner to meet new monitoring needs, resulting in incomplete signal coverage and an inability to effectively monitor all electromagnetic interference sources.

[0004] It should be noted that the information disclosed in this background section is intended only to enhance the understanding of the overall background of the present invention, and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] In view of the above-mentioned defects or improvement needs of the existing technology, the present invention provides a deployment method for mobile electromagnetic spectrum monitoring equipment that integrates multi-physics field sensing. This solves the technical problem that spectrum monitoring in the existing technology mostly relies on fixed monitoring point deployment. The monitoring range of these devices is statically set and lacks dynamic adaptability, resulting in incomplete signal coverage and inability to effectively monitor electromagnetic interference sources.

[0006] The specific technical solution is as follows:

[0007] This invention provides a method for deploying mobile electromagnetic spectrum monitoring equipment that integrates multi-physics sensing, the method comprising:

[0008] Electromagnetic propagation modeling with multi-physics coupling is performed in the target monitoring area to generate a signal continuous coverage probability cloud map. Based on the signal continuous coverage probability cloud map, Pareto front screening is performed on multiple equipment deployment initial schemes randomly generated in the target monitoring area to obtain H global candidate solutions. S1: In the solution space wide-area search layer, the H global candidate solutions are used as the initial population to perform a diversity-enhanced global exploration and output a diversity potential solution set. S2: In the physical field constraint refinement layer, constraint-aware local optimization based on real-time feedback from the signal continuous coverage probability cloud map is performed on the diversity potential solution set to output a physical field compliant deployment scheme set. The physical field compliant deployment scheme set is used as a new generation initial population, and S1 to S2 are iteratively executed. Two-layer physical field collaborative iterative optimization is performed in the solution space wide-area search layer and the physical field constraint refinement layer until the convergence condition is met, and a multi-physics constraint deployment coordinate set is output. The multi-physics constraint deployment coordinate set is robustly verified, and an anti-interference optimized deployment scheme is output. The monitoring network is deployed in the target monitoring area with reference to this scheme.

[0009] In one implementation, based on the signal continuous coverage probability cloud map, Pareto front screening is performed on multiple initial equipment deployment schemes randomly generated in the target monitoring area to obtain H global candidate solutions, including:

[0010] The initial deployment schemes of the multiple equipment are projected onto the signal continuous coverage probability cloud map, and an evaluation based on a preset information theory performance three-objective function is performed to obtain multiple scheme fitness vectors. The preset Pareto non-dominated ranking criterion is used to perform Pareto front screening on the fitness vectors of the multiple schemes to obtain the H global candidate solutions.

[0011] In one implementation, at the solution space wide-area search layer, using the H global candidate solutions as the initial population, a diversity-enhanced global exploration is performed, outputting a diversity potential solution set, including:

[0012] A1: Based on the physical field constraint perception of the signal continuous coverage probability cloud map, perform local optimization around the H global candidate solutions to obtain H groups of local refined candidate solutions; A2: Using the H global candidate solutions as diversity seed centers, conduct global breadth exploration based on the population cooperative diffusion mechanism to obtain H groups of wide-area diffusion candidate solutions; A3: Project the H groups of local refined candidate solutions and the H groups of wide-area diffusion candidate solutions onto the signal continuous coverage probability cloud map, evaluate them based on the preset information theory effectiveness three-objective function, obtain the corresponding scheme fitness vector, and then perform frontier solution set screening based on the Pareto non-dominated ranking criterion to obtain multiple elite-retained candidate solutions and multiple diversity supplementary candidate solutions; according to the distribution ratio of the multiple elite-retained candidate solutions and multiple diversity supplementary candidate solutions in the population, dynamically adjust the population individual allocation ratio of local optimization and global breadth exploration in the next iteration, and repeat steps A1 to A3 until the preset iteration termination condition is met to obtain the diversity potential solution set.

[0013] In one implementation, at the physical field constraint refinement layer, constraint-aware local optimization based on real-time feedback from the signal continuous coverage probability cloud map is performed on the diverse potential solution set to output a set of physical field compliant deployment schemes, including:

[0014] A first refined search space is defined within the coordinate neighborhood of the first candidate deployment scheme in the diverse potential solution set. Within the first refined search space, a first set of candidate trial solutions is generated around the first candidate deployment scheme based on deterministic sampling rules. The first candidate deployment scheme and the first set of candidate trial solutions are projected onto the signal continuous coverage probability cloud map for hard constraint compliance verification to select W compliant trial solutions. Based on the signal continuous coverage probability cloud map, W signal-to-noise ratio spatial gradients and W path loss change rates of the W compliant trial solutions relative to the first candidate deployment scheme are calculated, and the results are determined according to the preset information theory performance three-dimensional... The objective function calculates W preliminary fitness vectors; based on the W signal-to-noise ratio spatial gradients, W path loss change rates, and W preliminary fitness vectors, a local search iteration guided by physical field gradient information is performed in the first refined search space to obtain a first locally refined solution; similarly, local optimization is performed on multiple candidate deployment schemes in the diverse potential solution set to obtain multiple locally refined solutions, forming a set of locally refined solutions; the set of locally refined solutions is projected onto the signal continuous coverage probability cloud map for accurate evaluation, and after obtaining the set of accurate fitness vectors, frontier screening based on Pareto non-dominated sorting is performed to output the set of physically compliant deployment schemes.

[0015] In one implementation, multi-physics coupled electromagnetic propagation modeling is performed in the target monitoring area to generate a probability cloud map of continuous signal coverage, including:

[0016] Based on the geographic spatial boundary of the target monitoring area, a three-dimensional dynamic electromagnetic propagation field is constructed by retrieving digital elevation models and real-time meteorological data from a geographic information database and a meteorological data server, respectively. Multiple initial coordinates of distributed monitoring points are preset within this three-dimensional dynamic electromagnetic propagation field. Fresnel zone diffraction loss calculations based on ray tracing are performed on the initial coordinates of these distributed monitoring points to obtain multiple diffraction loss vectors. These vectors are then fused with the real-time meteorological data for dynamic atmospheric attenuation correction, outputting multiple signal propagation characteristic vectors for each monitoring point. Using the initial coordinates of these distributed monitoring points as input, spatial interpolation and probability statistics are performed within the three-dimensional dynamic electromagnetic propagation field based on these multiple signal propagation characteristic vectors to generate the signal continuous coverage probability cloud map.

[0017] In one implementation, the initial deployment schemes of the plurality of equipment are projected onto the signal continuous coverage probability cloud map, and an evaluation based on a preset information theory performance three-objective function is performed to obtain multiple scheme fitness vectors, including:

[0018] The coordinates of multiple monitoring equipment in the initial deployment scheme of the first equipment are projected onto multiple three-dimensional spatial locations corresponding to the signal continuous coverage probability cloud map to extract multiple signal propagation characteristic data. Multiple achievable signal-to-noise ratio values ​​are extracted from the multiple signal propagation characteristic data. After calculating multiple maximum information transmission rates supported by the multiple three-dimensional spatial locations according to Shannon's theorem, they are aggregated and summed to generate a first comprehensive information resolvability. Multiple Euclidean displacements of the coordinates of the multiple monitoring equipment relative to the initial coordinates of the multiple distributed monitoring points are calculated, and the displacements are summed to output a first total displacement cost. Based on multiple path losses and multiple terrain occlusion markers in the multiple signal propagation characteristic data, a signal leakage risk assessment is performed to obtain a first signal leakage risk degree. The first comprehensive information resolvability, the first total displacement cost, and the first signal leakage risk degree are combined in a preset order to generate a first scheme fitness vector.

[0019] In one implementation, a signal leakage risk assessment is performed based on multiple path losses and multiple terrain masking indicators in the multiple signal propagation characteristic data to obtain a first signal leakage risk level, including:

[0020] Using the multiple path losses as basic propagation loss data and the multiple terrain occlusion markers as terrain correction factors, multiple signal intensity fields generated by multiple grid points corresponding to the coordinates of the multiple monitoring equipment in the target monitoring area are calculated. After spatially superimposing the multiple signal intensity fields onto a preset sensitive area distribution map, signal intensity violations are determined for the grid points in the superimposed area based on a privacy protection threshold to count the number of violating grid points. The proportion of the number of violating grid points to the total number of grids in the sensitive area distribution map is calculated, and the first signal leakage risk level is output.

[0021] In one implementation, the hard constraint compliance verification includes communication blind spot verification, signal leakage risk verification, and deployment over-limit verification.

[0022] Beneficial effects of the embodiments of the present invention:

[0023] By modeling electromagnetic propagation through multi-physics coupling, a continuous signal coverage probability cloud map is generated based on geographical and meteorological data of the target area. This cloud map provides the propagation characteristics of electromagnetic waves within the target area, accurately predicting signal strength, coverage, and quality in different regions. Using Pareto frontier screening technology, the optimal equipment deployment scheme is selected, avoiding uneven signal coverage and over-deployment problems caused by random deployment. The selected H global candidate solutions represent the best balance among multiple objectives. Through a two-layer optimization process involving a wide-area solution space search layer and a physical constraint refinement layer, the potential of deployment schemes can be explored in depth, while ensuring that all deployment schemes conform to actual physical constraints. This enhanced global exploration ensures the diversity of schemes and avoids... By avoiding the trap of local optima, the comprehensive coverage of the monitoring network is ensured. Dual-layer physical field collaborative iterative optimization ensures that the deployment scheme can be optimized at both the global and local levels, making the scheme not only theoretically optimized but also adaptable to dynamically changing environments. Through iterative optimization, it can effectively cope with different physical conditions and changes. Robustness verification of the deployment coordinate set ensures that the scheme can maintain effective operation in the face of various real-world interferences, ensuring the stability and reliability of the system in real-world environments. Through multi-physics sensing and accurate modeling, the possibility of signal leakage can be effectively assessed, and corresponding optimization measures can be taken to reduce the risk of signal leakage, thereby avoiding compliance and security risks such as privacy leaks, excessive electromagnetic interference, or public health concerns.

[0024] Of course, implementing any product or method of the present invention does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This invention illustrates a flowchart of the deployment method for a mobile electromagnetic spectrum monitoring equipment that integrates multi-physics sensing.

[0027] Figure 2This diagram illustrates the Pareto front screening process in the deployment method of the mobile electromagnetic spectrum monitoring equipment that integrates multi-physics sensing provided by the present invention. Detailed Implementation

[0028] To facilitate understanding of the present invention, a more complete description of the invention will be given below with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein; rather, these embodiments are provided so that the disclosure of the invention will be more thorough and complete.

[0029] Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0030] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0031] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0032] The present invention provides a method for deploying mobile electromagnetic spectrum monitoring equipment that integrates multi-physics sensing, which solves the technical problem that in the prior art, spectrum monitoring mostly relies on fixed monitoring point deployment. The monitoring range of these devices is statically set and lacks dynamic adaptability, resulting in incomplete signal coverage and inability to effectively monitor electromagnetic interference sources.

[0033] See Figure 1 The present invention provides a method for deploying mobile electromagnetic spectrum monitoring equipment that integrates multi-physics sensing, the method comprising:

[0034] Y100: Perform electromagnetic propagation modeling with multi-physics coupling in the target monitoring area to generate a probability cloud map of continuous signal coverage.

[0035] The target monitoring area for electromagnetic spectrum monitoring is defined, which can be an urban area, rural area, or other specific geographical region. Geographic Information System (GIS) data is used to determine the area's boundaries and related spatial characteristics. Electromagnetic propagation is modeled using a multiphysics coupling method, considering the impact of various physical environmental factors, such as terrain, weather, and buildings, on electromagnetic wave propagation. A dynamic three-dimensional electromagnetic propagation field is constructed using a digital elevation model (DEM) and real-time meteorological data. Based on this data, initial coordinates of multiple distributed monitoring points are generated within the target monitoring area. Ray tracing is used to calculate Fresnel zone diffraction loss, and atmospheric attenuation correction is performed using real-time meteorological data. Finally, the electromagnetic propagation characteristic vector for each monitoring point is output. Based on the calculated propagation characteristics, spatial interpolation and probabilistic statistical methods are used to generate a signal continuity coverage probability cloud map. This cloud map displays the signal coverage and signal strength at various points within the target area, accurately reflecting the coverage probability of different regions.

[0036] Y200: Based on the signal continuous coverage probability cloud map, Pareto front screening is performed on multiple initial equipment deployment schemes randomly generated in the target monitoring area to obtain H global candidate solutions.

[0037] Multiple initial equipment deployment schemes are randomly generated within the target monitoring area. Each deployment point in these schemes corresponds to the installation location of a monitoring device. The objective of each scheme is to cover the target area and ensure that signal strength, stability, and other parameters meet requirements. These initial deployment schemes are projected onto a previously generated signal continuity probability cloud map. Each scheme is evaluated using a three-objective function based on information theory effectiveness, resulting in a fitness vector for each scheme. These fitness vectors describe the performance of each scheme across multiple objectives, including comprehensive information resolvability, total displacement cost, and signal leakage risk. The Pareto front ranking criterion is used to screen all initial schemes. The Pareto front is a set of solutions that cannot be simultaneously improved across all objectives. This method selects H global candidate solutions that achieve an optimal balance among different objectives.

[0038] S1: In the solution space wide-area search layer, using the H global candidate solutions as the initial population, perform a global exploration with enhanced diversity and output a set of solutions with diversity potential.

[0039] In the wide-area search layer of the solution space, H global candidate solutions are used as the initial solution set of the population. These solutions represent the preliminary optimization results. Next, a diversity-enhancing global exploration is performed, that is, by introducing different search strategies, potential optimal solutions are explored in a wider solution space. This step emphasizes increasing the diversity of the solution set to prevent getting trapped in local optima. Appropriate diversity enhancement techniques, such as population diversity mechanisms and genetic algorithms, ensure that the algorithm escapes local optima during the search process and explores more promising solutions. Through wide-area exploration, a diverse potential solution set is obtained, representing potentially excellent solutions in the solution space.

[0040] S2: In the physical field constraint refinement layer, perform constraint-aware local optimization based on the real-time feedback of the signal continuous coverage probability cloud map on the diverse potential solution set, and output a set of physical field compliant deployment schemes.

[0041] In the physical field constraint refinement layer, the diverse potential solution set is further optimized. This stage focuses on refining the solution set based on actual physical constraints, such as signal coverage and path loss. Detailed local optimization is performed on the solution set based on real-time feedback of the signal coverage probability cloud map. Local optimization is performed within the neighborhood of each potential solution. This process evaluates the performance of each solution under specific constraints based on the generated probability cloud map and performs further local searches to refine the performance of each solution. Through this local optimization, solutions that meet the physical constraints are further selected. Throughout this process, it is ensured that the solution set conforms to physical field constraints, such as signal coverage continuity, path loss minimization, and other environment-related physical conditions, such as meteorological factors and terrain. Finally, physically compliant deployment schemes are output, which have high signal coverage quality and meet the physical limitations of the target area.

[0042] Y300: The set of compliant physical field deployment schemes is used as the new generation initial population. S1 to S2 are executed iteratively. Two-layer physical field collaborative iterative optimization is performed in the solution space wide-area search layer and the physical field constraint refinement layer until the convergence condition is met, and the multi-physical field constraint deployment coordinate set is output.

[0043] The scheme is further optimized to ensure that the optimal deployment point is found in the solution space. During the optimization process, S1 to S2 are iteratively executed. Each iteration generates a new potential solution set. After refinement and adjustment, the iteration continues until a preset convergence condition is met, such as the change in the optimization result falling below a certain threshold or reaching the maximum number of iterations. Once the convergence condition is met, the optimized deployment coordinates are finally obtained, i.e., the multiphysics-constrained deployment coordinate set. These coordinate sets satisfy the multiphysics constraints, ensuring the optimal deployment of the electromagnetic spectrum monitoring system within the target area.

[0044] Y400: Perform robustness verification on the multiphysics constraint deployment coordinate set, output an anti-interference optimized deployment scheme, and deploy the monitoring network in the target monitoring area.

[0045] Robustness verification is performed on the multiphysics-constrained deployment coordinate set to ensure that these deployment schemes can operate stably under different environmental conditions, such as considering different meteorological conditions and terrain changes in the target area. The purpose of robustness verification is to evaluate the adaptability and stability of the deployment schemes in the face of external interference (such as electromagnetic interference, equipment failure, environmental changes, etc.), ensuring that the deployed monitoring network can effectively cope with complex real-world environments. During the robustness verification process, anti-interference capability is also optimized. This includes evaluating the response of each deployment scheme to electromagnetic interference and adjusting the deployment schemes through optimization algorithms to improve anti-interference capability. For example, the deployment scheme may prioritize areas with less interference or adjust the deployment position to ensure that the monitoring equipment can maintain good performance under interference environments. After robustness verification and anti-interference optimization, the final anti-interference optimized deployment scheme is output. This scheme can effectively cope with external interference within the target monitoring area, ensuring the stability and reliability of the electromagnetic spectrum monitoring system.

[0046] In one implementation, based on the signal continuous coverage probability cloud map, Pareto front screening is performed on multiple initial equipment deployment schemes randomly generated in the target monitoring area to obtain H global candidate solutions, including:

[0047] Y210: Project the multiple initial equipment deployment schemes onto the signal continuous coverage probability cloud map, and evaluate them based on a preset information theory performance three-objective function to obtain multiple scheme fitness vectors; Y220: Use a preset Pareto non-dominated ranking criterion to perform Pareto front screening on the multiple scheme fitness vectors to obtain the H global candidate solutions.

[0048] The generated initial equipment deployment schemes are projected onto a signal continuous coverage probability cloud map. Through projection, the deployment schemes are mapped to the signal coverage situation, and the signal coverage capability of different schemes in the actual environment is evaluated. Each scheme is evaluated using a pre-defined three-objective function of information theory effectiveness. Specifically, after projection, signal propagation characteristic data is extracted from each three-dimensional spatial location. Using the extracted signal propagation characteristic data, the maximum information rate that the signal can transmit is obtained according to Shannon's theorem. By aggregating and summing the maximum information transmission rates of multiple three-dimensional spatial locations, the comprehensive information resolvability of the deployment scheme is obtained, which is the upper limit of the deployment scheme's ability to extract information from electromagnetic signals. The total displacement cost required for the monitoring equipment to move from the initial position to the target deployment position is calculated. The total displacement cost refers to the distance the equipment needs to move when adjusting its deployment position, expressed in the form of Euclidean displacement, i.e., measuring the distance between the initial position and the target position and summing the distances to obtain the total movement cost. Based on the path loss and terrain masking markers extracted from the signal propagation characteristic data, a signal leakage risk assessment is performed. That is, the signal strength field generated by the coordinates of each monitoring equipment is calculated, and multiple signal strength fields are superimposed on a predetermined sensitive area, such as civilian areas and sensitive facilities, to assess whether the signal will leak into these areas, determine the risk of signal leakage, and calculate the signal leakage risk level, i.e., the number of non-compliant grid points and their proportion of the total number of grid points. Based on the evaluation results, a fitness vector is generated for each deployment scheme. The fitness vector contains the score of each scheme under multiple evaluation criteria, including comprehensive information resolvability, total displacement cost, and signal leakage risk, which is a quantitative representation of the scheme performance.

[0049] The Pareto non-dominated ranking method is used to rank the fitness vectors of all deployment schemes. The Pareto front refers to the set of solutions that cannot simultaneously improve all objectives under multiple optimization goals. In the Pareto ranking process, fitness vectors are first grouped according to front level. Solutions with higher front levels indicate that they perform well on all objectives and cannot be improved by other schemes. Within the same front level, the crowding distance of each scheme is further calculated. This metric reflects the density of the relative distribution of solutions in the objective space. A larger crowding distance indicates higher diversity in the solution set, meaning that it occupies a more dispersed area in the objective space. Finally, based on the Pareto front ranking and crowding distance, H global candidate solutions are selected in descending order of crowding. These solutions represent the optimal balance points across multiple objectives and are candidates for the optimal deployment scheme.

[0050] In one implementation, in the solution space wide-area search layer, using the H global candidate solutions as the initial population, a diversity-enhanced global exploration is performed, outputting a diversity potential solution set, including:

[0051] A1: Based on the physical field constraint perception of the signal continuous coverage probability cloud map, perform local optimization around the H global candidate solutions to obtain H groups of local refined candidate solutions; A2: Using the H global candidate solutions as diversity seed centers, conduct global breadth exploration based on the population cooperative diffusion mechanism to obtain H groups of wide-area diffusion candidate solutions; A3: Project the H groups of local refined candidate solutions and the H groups of wide-area diffusion candidate solutions onto the signal continuous coverage probability cloud map, evaluate them based on the preset information theory effectiveness three-objective function, obtain the corresponding scheme fitness vector, and then perform frontier solution set screening based on the Pareto non-dominated ranking criterion to obtain multiple elite-retained candidate solutions and multiple diversity supplementary candidate solutions; S11: According to the distribution ratio of the multiple elite-retained candidate solutions and multiple diversity supplementary candidate solutions in the population, dynamically adjust the population individual allocation ratio of local optimization and global breadth exploration in the next iteration, and repeat steps A1 to A3 until the preset iteration termination condition is met to obtain the diversity potential solution set.

[0052] The signal continuous coverage probability cloud map illustrates the signal coverage at various locations within the target monitoring area. Physical constraints refer to the impact of environmental factors on electromagnetic wave propagation, including signal attenuation, path loss, and terrain masking. For the H global candidate solutions selected from the Pareto front, local optimization is performed in the solution space. This local optimization process employs a refined search strategy, refining the position and configuration of each solution to ensure optimization under specific constraints. For example, adjusting physical parameters such as position, angle, and direction further improves signal coverage and interference control. Through local optimization, H sets of locally refined candidate solutions are obtained, representing the local optimization results of each candidate deployment scheme.

[0053] H global candidate solutions are used as seeds, serving as the basis for further exploration. These seed solutions exhibit significant diversity in the solution space, covering potential solutions in different regions, thus promoting extensive exploration of the solution space. A cooperative diffusion mechanism is employed for wide-area exploration. Specifically, this involves a broad search within the solution space, utilizing the existing population to expand the exploration scope. This mechanism avoids getting trapped in local optima while enhancing the diversity of exploration. Strategies such as crossover and mutation from genetic algorithms are employed during the exploration process, allowing the population to expand more diversely within the solution space. Based on the global candidate solutions, global expansion and exploration are performed to obtain H sets of wide-area diffused candidate solutions. These solutions represent potential solutions obtained through broad exploration; they may not be local optima, but they cover different parts of the solution space, providing more options for subsequent optimization.

[0054] H groups of locally refined candidate solutions and H groups of widely diffused candidate solutions are projected onto the signal continuous coverage probability cloud map. This is to evaluate the performance of each solution in the real environment. After projection, information such as signal strength and path loss of each solution within the coverage area is obtained. Each projected solution is evaluated based on a three-objective function of information theory performance. The fitness vectors of all solutions are sorted using the Pareto non-dominated ranking criterion, and all solutions are divided according to the Pareto front, forming multiple front levels. Solutions with higher front levels outperform other solutions in all objectives and perform better. For solutions within the same front level, crowding distance is used for further ranking. The greater the crowding, the sparser the distribution of the solution in the solution space, indicating better diversity. In this way, multiple elite-preserving candidate solutions and multiple diversity-supplementing candidate solutions are selected, ensuring a solution set that is both excellent and diverse.

[0055] In each iteration, the distribution ratio of elite candidate solutions and diversity-supplemented candidate solutions is tracked. This ratio reflects the balance between high-quality and diverse solutions in the current population. Based on the performance of the solution set in the previous iteration, the individual allocation ratio of local optimization and global breadth exploration in the population is dynamically adjusted. If the current solution set has high diversity, the proportion of local optimization is increased to refine the optimization of the solution set; if the current solution set has too many elite solutions, the proportion of global breadth exploration is increased to ensure the diversity of the solution set and avoid over-convergence. After adjusting the ratio, steps A1 to A3 are executed to perform a new round of optimization until the preset iteration termination condition is met. Finally, after multiple rounds of iterative optimization, a solution set containing multiple diverse solutions is obtained. These solutions have good signal coverage, low path loss, and high anti-interference ability, forming the final diverse potential solution set.

[0056] In one implementation, at the physical field constraint refinement layer, constraint-aware local optimization based on real-time feedback from the signal continuous coverage probability cloud map is performed on the diverse potential solution set, outputting a set of physical field compliant deployment schemes, including:

[0057] S21: Define a first refined search space within the coordinate neighborhood of the first candidate deployment scheme in the diverse potential solution set; S22: Within the first refined search space, generate a first set of candidate trial solutions around the first candidate deployment scheme based on deterministic sampling rules; S23: Project the first candidate deployment scheme as the initial solution and the first set of candidate trial solutions onto the signal continuous coverage probability cloud map, perform hard constraint compliance verification, and select W compliant trial solutions; S24: Calculate the W signal-to-noise ratio spatial gradients and W path loss change rates of the W compliant trial solutions relative to the first candidate deployment scheme based on the signal continuous coverage probability cloud map, and evaluate their effectiveness according to the preset information. S25: Calculate W preliminary fitness vectors using the three objective functions; S26: Based on the W signal-to-noise ratio spatial gradients, W path loss change rates, and W preliminary fitness vectors, perform a local search iteration guided by physical field gradient information within the first refined search space to obtain a first locally refined solution; S27: Similarly, perform local optimization on multiple candidate deployment schemes in the diverse potential solution set to obtain multiple locally refined solutions, forming a set of locally refined solutions; S28: Project the set of locally refined solutions onto the signal continuous coverage probability cloud map for accurate evaluation, obtain an accurate fitness vector set, and then perform frontier screening based on Pareto non-dominated sorting to output the set of physically compliant deployment schemes.

[0058] Starting with the first candidate deployment scheme in the diversity potential solution set, its neighborhood space is defined. The neighborhood space refers to a certain range around the current deployment scheme, representing the possible solution set near the current solution. This neighborhood space is defined based on the coordinates of the current deployment scheme and sets a certain search range, for example, considering minor adjustments to the position or reasonable offsets in the nearby region. Based on the defined neighborhood space, a fine-grained search space is further defined within this space. This is a relatively small area used for searching within a more detailed range. Through fine-grained searching, the scheme is further optimized to ensure its optimal performance under physical constraints. For example, it explores minute changes such as local position adjustments and device orientation to obtain better signal coverage or reduce path loss.

[0059] Deterministic sampling rules mean that within a fine-grained search space, points are not randomly selected for searching. Instead, sampling is performed according to certain rules or strategies. Deterministic sampling ensures the predictability of the sampling process, avoids the influence of randomness on the search process, and thus improves search efficiency and accuracy. Deterministic sampling rules include uniform sampling, which involves selecting multiple points evenly within the search space, or selecting preferred locations based on the preferences of a certain algorithm, such as signal coverage strength. Within the first fine-grained search space, a new set of alternative tentative solutions is generated based on the deterministic sampling rules. These solutions are generated by fine-tuning the original first candidate deployment scheme. Each alternative solution represents a potential deployment scheme that may be better than the current solution. The main objectives of these tentative solutions are to optimize signal coverage, reduce path loss, or reduce interference. By evaluating these solutions, the potential improvements in the neighborhood space can be better understood, thus helping to select the optimal solution.

[0060] The first candidate deployment scheme and the first set of alternative trial solutions are projected onto a pre-generated probability cloud map of continuous signal coverage. The purpose is to compare the location of each solution with the actual signal coverage and evaluate their effectiveness in the actual deployment environment. This matches the theoretical solutions with the actual environmental conditions. Hard constraint compliance verification is a necessary check performed on each projected solution to ensure that these solutions meet all physical field constraints, including communication blind zone verification, signal leakage risk verification, and deployment over-limit verification. After hard constraint compliance verification, W compliant trial solutions that meet all constraints are selected. These solutions meet the actual deployment requirements and perform well in terms of signal coverage and interference control.

[0061] The signal-to-noise ratio (SNR) spatial gradient of each compliance trial solution relative to the first candidate deployment scheme is calculated. The SNR gradient refers to the rate of change of signal quality in space. Specifically, this gradient represents the change in signal strength within a certain area. A large gradient indicates that the signal changes drastically, indicating unstable signal quality. Path loss is the degree of signal attenuation during electromagnetic wave propagation. The path loss change rate represents the rate of signal attenuation change between different deployment locations. By calculating the path loss change rates of W compliance trial solutions, the impact of different solution locations on signal attenuation can be understood, further identifying areas with significant attenuation. Based on these two metrics, and combined with a pre-defined three-objective function, the preliminary fitness vector of each compliance trial solution is calculated. These fitness vectors represent the comprehensive performance of each solution across multiple objectives.

[0062] Using the previously calculated spatial gradient of signal-to-noise ratio and the rate of change of path loss, a local search is performed within the first refined search space. Here, the search is guided by physical field gradient information, meaning the search direction is determined based on the spatial distribution of signal quality and the variation in path loss, prioritizing regions with smaller signal quality variations or lower path loss for further exploration. This approach avoids wasting computational resources in regions with poor signal quality or high path loss, allowing for focused optimization on promising areas. After local search and optimization, the first locally refined solution is obtained, which shows further optimization in signal coverage and path loss, representing a significant improvement over the original solution.

[0063] For each candidate deployment scheme in the diverse potential solution set, the local optimization process in S26 is repeated. Each solution undergoes a physical constraint-guided local search within its corresponding refined search space. Through this process, starting from each candidate solution, fine-tuning and optimization are performed step by step to ensure that physical constraints such as signal coverage and path loss are sufficiently improved for each solution. Each candidate solution generates a locally refined solution, which represents the optimal result on different optimization paths, forming a set of locally refined solutions. These solutions are of relatively high quality and satisfy the physical constraints.

[0064] The obtained locally refined solution set is projected onto the previously generated signal continuous coverage probability cloud map for precise evaluation. This method allows assessment of each solution's performance in the real-world environment, further confirming whether the signal coverage, path loss, and other parameters of these solutions have reached optimal levels. Each locally refined solution is assigned a precise fitness vector. Based on these vectors, Pareto non-dominated sorting is performed, and the solution set undergoes front screening. Finally, through Pareto front screening, a set of physics-compliant deployment schemes is output. This set represents the final optimization result, ensuring optimal electromagnetic spectrum monitoring performance within the target area.

[0065] In one implementation, electromagnetic propagation modeling with multi-physics coupling is performed in the target monitoring area to generate a probability cloud map of continuous signal coverage, including:

[0066] Y110: Based on the geographic spatial boundary of the target monitoring area, retrieve the digital elevation model and real-time meteorological data from the geographic information database and meteorological data server respectively to construct a three-dimensional dynamic electromagnetic propagation field; Y120: Preset the initial coordinates of multiple distributed monitoring points in the three-dimensional dynamic electromagnetic propagation field; Y130: Perform Fresnel zone diffraction loss calculation based on ray tracing on the initial coordinates of the multiple distributed monitoring points to obtain the diffraction loss vectors of multiple monitoring points, then fuse the real-time meteorological data to perform dynamic atmospheric attenuation correction, and output the signal propagation characteristic vectors of multiple monitoring points; Y140: Using the initial coordinates of the multiple distributed monitoring points as input, perform spatial interpolation and probability statistics based on the signal propagation characteristic vectors of the multiple monitoring points in the three-dimensional dynamic electromagnetic propagation field to generate the signal continuous coverage probability cloud map.

[0067] The target monitoring area refers to the specific region where the electromagnetic spectrum needs to be monitored. This can be a city, a rural area, or other specific geographical region. The specific scope of the monitoring area is determined by obtaining the geographic spatial boundaries of this region. A digital elevation model (DEM) is a digital representation of the Earth's surface elevation, including the height of the ground and other objects such as buildings and mountains. Obtaining DEM data from a geographic information database allows for a detailed description of the topography of the target area. This data is fundamental to electromagnetic wave propagation modeling because electromagnetic wave propagation is affected by terrain, especially complex terrain such as mountainous or urban areas, which significantly impacts the signal propagation path. Real-time meteorological data, including information on temperature, humidity, wind speed, and air pressure, is crucial for electromagnetic wave propagation. Meteorological factors affect air density, thus influencing the attenuation of electromagnetic waves. Combining the DEM and real-time meteorological data, a three-dimensional dynamic electromagnetic propagation field is constructed. This field dynamically simulates the propagation process of electromagnetic waves within the target area, providing fundamental data for subsequent signal coverage assessment.

[0068] Based on the constructed three-dimensional dynamic electromagnetic propagation field, the locations of multiple monitoring points are preset. Distributed monitoring points refer to multiple monitoring devices distributed within the target area to collect electromagnetic signal data. The initial coordinates of these monitoring points are determined based on factors such as the geographical characteristics of the target area, signal coverage requirements, and equipment layout scheme. They are distributed in key locations within the area, such as city centers, peripheral areas, or complex terrain areas, to optimize signal coverage.

[0069] Ray tracing is a method for simulating electromagnetic wave propagation. It calculates signal attenuation and reflection by tracing the path of the electromagnetic wave. The Fresnel zone is an important concept in electromagnetic wave propagation, indicating that diffraction may occur in certain areas along the propagation path, affecting signal strength. For each preset distributed monitoring point, the ray tracing algorithm is used to calculate the diffraction loss between that point and other areas, i.e., the signal loss when passing through obstacles (such as buildings, hills, etc.). This helps to estimate the signal attenuation at each monitoring point.

[0070] Atmospheric attenuation refers to the decrease in signal strength of electromagnetic waves during propagation due to changes in atmospheric conditions such as humidity, temperature, and wind speed. Real-time meteorological data has a direct impact on atmospheric attenuation. After calculating diffraction loss, atmospheric attenuation is corrected based on real-time meteorological data because meteorological changes can lead to variations in diffraction loss; for example, high humidity can increase signal attenuation. By calculating diffraction loss through ray tracing and correcting it with meteorological data, the final output is a signal propagation characteristic vector for each monitoring point. These vectors contain information about the signal propagation characteristics of each monitoring point, including diffraction loss, atmospheric attenuation, and other influencing factors.

[0071] Spatial interpolation estimates signal propagation between known monitoring points or at other unknown points by utilizing their known propagation characteristics. This interpolation method provides a smooth transition of signal propagation across the entire monitoring area. Probabilistic statistics analyze the signal coverage probability at different locations, i.e., the likelihood of signal strength in each area under different times or conditions. Combining spatial interpolation and probabilistic statistics generates a continuous signal coverage probability cloud map. This cloud map describes the signal coverage probability at different locations within the target area, displaying the spatial distribution of signal strength, signal attenuation, and coverage quality.

[0072] In one implementation, the initial deployment schemes of the multiple equipment are projected onto the signal continuous coverage probability cloud map, and an evaluation based on a preset information-theoretic three-objective function is performed to obtain multiple scheme fitness vectors, including:

[0073] Y211: Project the coordinates of multiple monitoring equipment in the initial deployment scheme of the first equipment onto multiple three-dimensional spatial locations corresponding to the signal continuous coverage probability cloud map to extract multiple signal propagation characteristic data; Y212: Extract multiple achievable signal-to-noise ratio values ​​from the multiple signal propagation characteristic data, calculate multiple maximum information transmission rates supported by the multiple three-dimensional spatial locations according to Shannon's theorem, and then aggregate and sum them to generate a first comprehensive information resolvability; Y213: Calculate multiple Euclidean displacements of the coordinates of the multiple monitoring equipment relative to the initial coordinates of the multiple distributed monitoring points, and then sum the displacements to output a first total displacement cost; Y214: Based on multiple path losses and multiple terrain occlusion markers in the multiple signal propagation characteristic data, perform a signal leakage risk assessment to obtain a first signal leakage risk degree; Y215: Combine the first comprehensive information resolvability, the first total displacement cost, and the first signal leakage risk degree in a preset order to generate a first scheme fitness vector.

[0074] The coordinates of multiple monitoring devices in the initial deployment plan are projected onto a signal continuous coverage probability cloud map. This projection allows for comparison between the actual deployed monitoring points and their corresponding three-dimensional spatial locations in the cloud map, enabling analysis of their signal propagation in the real-world scenario. By aligning the monitoring device coordinates with their three-dimensional spatial locations in the signal coverage probability cloud map, signal propagation characteristic data related to these locations are extracted. This includes: achievable signal-to-noise ratio (SNR), which reflects the ratio between signal strength and noise; a higher SNR indicates better signal quality; path loss, the attenuation of electromagnetic waves during propagation; lower path loss indicates better signal quality; and terrain occlusion indicators, which reflect the potential obstruction of electromagnetic waves during propagation due to terrain features that may block or reflect them.

[0075] From the extracted signal propagation characteristic data, multiple achievable signal-to-noise ratio (SNR) values ​​are extracted. These reflect the signal quality of each monitoring device relative to its corresponding location. A higher SNR value indicates better signal quality at that location, while a lower SNR value may indicate a weaker signal or higher noise levels. Shannon's theorem provides a method for calculating the maximum information transmission rate of a channel, i.e., calculating the maximum data transmission rate that the channel can support under a given SNR condition. For the three-dimensional spatial location of each monitoring device, the corresponding maximum information transmission rate is calculated using Shannon's theorem, which represents the maximum data rate that signal propagation and reception can support at that location. The maximum information transmission rates of multiple locations are aggregated and summed to obtain the first comprehensive information resolvability of the entire scheme. This index represents the comprehensive information transmission capability of the entire deployment scheme; the higher the information transmission rate, the stronger the overall effectiveness of the deployment scheme.

[0076] The Euclidean displacement of the monitoring equipment relative to the initial coordinates of the distributed monitoring points is calculated. Euclidean displacement is the straight-line distance between two coordinates, representing the spatial displacement of the equipment from one location to another. For each monitoring equipment, its displacement distance to each distributed monitoring point is calculated. The displacement values ​​between all monitoring equipment and distributed monitoring points are summed to obtain the first total displacement cost. This cost reflects the spatial distribution of the monitoring equipment in the deployment plan. A higher displacement cost indicates that the equipment is too far apart, resulting in higher deployment costs or greater coordination difficulties.

[0077] Based on the previously extracted path loss and terrain occlusion indicators, a signal leakage risk assessment is conducted. Path loss reflects the attenuation of electromagnetic waves during propagation; higher path loss may lead to more severe signal attenuation, thus increasing the risk of signal leakage. Terrain occlusion indicators provide information on the impact of terrain on electromagnetic wave propagation; if strong terrain occlusion exists, the signal may not be able to propagate normally to the expected location, resulting in signal leakage risk. Combining path loss and terrain occlusion, a first signal leakage risk level is derived for each deployment scheme. This indicator reflects the potential signal leakage risk that may arise when deploying the scheme within the target area.

[0078] The first comprehensive information resolvability, the first total displacement cost, and the first signal leakage risk are combined in a preset order. This combination process is to balance multiple objectives and obtain a comprehensive evaluation index. Finally, a first scheme fitness vector is generated, which represents the performance of the deployment scheme in multiple aspects. The better the fitness vector, the better the overall effect of the deployment scheme.

[0079] In one implementation, a signal leakage risk assessment is performed based on multiple path losses and multiple terrain masking indicators in the multiple signal propagation characteristic data to obtain a first signal leakage risk level, including:

[0080] Y2141: Using the multiple path losses as basic propagation loss data and the multiple terrain occlusion markers as terrain correction factors, calculate the multiple signal intensity fields generated by the multiple grid points corresponding to the coordinates of the multiple monitoring equipment in the target monitoring area; Y2142: After spatially superimposing the multiple signal intensity fields onto a preset sensitive area distribution map, determine the signal intensity violation of the grid points in the superimposed area according to the privacy protection threshold, and count the number of violating grid points; Y2143: Calculate the proportion of the number of violating grid points to the total number of grids in the sensitive area distribution map, and output the first signal leakage risk level.

[0081] By combining multiple path loss data with terrain masking indicators, the propagation of electromagnetic waves is analyzed. Path loss represents the attenuation of electromagnetic waves during propagation, while terrain masking indicators reflect the obstruction and reflection effects of terrain on electromagnetic wave propagation. The combination of these two methods provides a more realistic representation of signal propagation in the actual environment. The coordinates and location of each monitoring device affect the signal strength it generates. Based on the location of each monitoring device, its path loss, and terrain masking indicators, the signal strength field at that point is calculated. The signal strength field refers to the intensity distribution of the signal as it propagates in space. The calculation results in multiple signal strength fields, representing the signal strength in different areas. These signal strength fields will be used for subsequent risk assessment.

[0082] Multiple signal strength fields are superimposed onto a pre-defined sensitive area distribution map. Sensitive areas refer to areas requiring special protection, such as residential areas, commercial areas, hospitals, and schools, where signal leakage risks must be strictly controlled. A privacy protection threshold is a set signal strength threshold, indicating that signal strength within the sensitive area should not exceed this threshold. This threshold is set based on privacy protection requirements and electromagnetic radiation safety standards. For each grid point in the superimposed area—that is, each small area in space—signal strength violation judgment is performed. If the signal strength of a grid point exceeds the privacy protection threshold, then that point is considered a violation. Through signal strength violation judgment, the number of violating grid points is counted. These violation points indicate excessive electromagnetic signal leakage within the sensitive area, which may affect the surrounding environment or privacy.

[0083] Based on the number of non-compliant grid points, the percentage of these non-compliant points within the entire sensitive area is calculated. A higher percentage indicates a greater risk of signal leakage and a higher likelihood of privacy breaches. Finally, the percentage of non-compliant grid points yields the first signal leakage risk level. This value represents an assessment of the current deployment's privacy protection capabilities. A high signal leakage risk level means additional measures are needed to ensure that electromagnetic signal leakage or interference does not occur within the sensitive area.

[0084] In one implementation, the hard constraint compliance verification includes communication blind spot verification, signal leakage risk verification, and deployment over-limit verification.

[0085] A communication dead zone refers to a situation where certain areas within the deployment area cannot receive a signal. This typically occurs due to building obstruction, terrain cover, or improper equipment deployment. Verifying communication dead zones ensures that all critical areas can effectively receive a signal. Signal leakage risk verification further assesses the signal leakage risk level obtained in the previous steps, ensuring that the deployment plan does not cause signal leakage beyond acceptable limits. This verification process helps ensure that the signal does not exceed the designated area, avoiding unnecessary interference and privacy leaks. Deployment over-limit verification checks whether the deployment plan violates certain hard restrictions, such as the number of deployed devices, power limits, and geographical restrictions. This ensures that the deployment plan complies with all regulations while maximizing signal coverage quality and interference control.

[0086] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0087] The foregoing description of specific exemplary embodiments of the invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. The scope of the invention is intended to be defined by the claims and their equivalents.

Claims

1. A deployment method for mobile electromagnetic spectrum monitoring equipment integrating multi-physics sensing, characterized in that, The method includes: Electromagnetic propagation modeling with multi-physics coupling is performed in the target monitoring area to generate a probability cloud map of continuous signal coverage; Based on the signal continuous coverage probability cloud map, Pareto front screening is performed on multiple initial equipment deployment schemes randomly generated in the target monitoring area to obtain H global candidate solutions. S1: In the solution space wide-area search layer, using the H global candidate solutions as the initial population, perform a global exploration with enhanced diversity and output a set of solutions with diversity potential; S2: In the physical field constraint refinement layer, perform constraint-aware local optimization based on the real-time feedback of the signal continuous coverage probability cloud map on the diverse potential solution set, and output a set of physical field compliant deployment schemes. The set of compliant physical field deployment schemes is used as the new generation initial population. S1 to S2 are executed iteratively. Two-layer physical field collaborative iterative optimization is performed in the solution space wide-area search layer and the physical field constraint refinement layer until the convergence condition is met, and the multi-physical field constraint deployment coordinate set is output. The robustness of the multiphysics-constrained deployment coordinate set is verified, and an anti-interference optimized deployment scheme is output. The monitoring network is then deployed in the target monitoring area.

2. The deployment method for mobile electromagnetic spectrum monitoring equipment integrating multi-physics sensing as described in claim 1, characterized in that, Based on the signal continuous coverage probability cloud map, Pareto front screening is performed on multiple initial equipment deployment schemes randomly generated in the target monitoring area to obtain H global candidate solutions, including: The initial deployment schemes of the multiple equipment are projected onto the signal continuous coverage probability cloud map, and an evaluation based on a preset information theory effectiveness three-objective function is performed to obtain fitness vectors of multiple schemes. The fitness vectors of the multiple solutions are subjected to Pareto front screening using a preset Pareto non-dominated ranking criterion to obtain the H global candidate solutions.

3. The deployment method for mobile electromagnetic spectrum monitoring equipment integrating multi-physics sensing as described in claim 2, characterized in that, In the wide-area search layer of the solution space, using the H global candidate solutions as the initial population, a diversity-enhanced global exploration is performed, outputting a set of diverse potential solutions, including: A1: Based on the physical field constraint perception of the signal continuous coverage probability cloud map, local optimization is performed around the H global candidate solutions to obtain H groups of local fine candidate solutions; A2: Using the H global candidate solutions as diversity seed centers, a global breadth exploration based on the population cooperative diffusion mechanism is carried out to obtain H groups of wide-area diffusion candidate solutions; A3: Project the H groups of local fine candidate solutions and H groups of wide-area diffusion candidate solutions onto the signal continuous coverage probability cloud map, respectively, and evaluate them based on the preset information theory effectiveness three objective function. After obtaining the corresponding scheme fitness vector, perform frontier solution set screening based on the Pareto non-dominated ranking criterion to obtain multiple elite-retained candidate solutions and multiple diversity supplementary candidate solutions. Based on the distribution ratio of the multiple elite-retained candidate solutions and the multiple diversity-supplemented candidate solutions in the population, the population individual allocation ratio for local optimization and global breadth exploration in the next iteration is dynamically adjusted, and steps A1 to A3 are repeated until the preset iteration termination condition is met to obtain the diversity potential solution set.

4. The deployment method for mobile electromagnetic spectrum monitoring equipment integrating multi-physics sensing as described in claim 2, characterized in that, In the physical field constraint refinement layer, constraint-aware local optimization based on real-time feedback from the signal continuous coverage probability cloud map is performed on the diverse potential solution set, outputting a set of physical field compliant deployment schemes, including: Define a first fine-grained search space within the coordinate neighborhood space of the first candidate deployment scheme in the set of diverse potential solutions; Within the first refined search space, a first set of alternative trial solutions is generated around the first candidate deployment scheme based on deterministic sampling rules; The first candidate deployment scheme is projected as the initial solution and the first group of alternative trial solutions onto the signal continuous coverage probability cloud map for hard constraint compliance verification, so as to select W compliant trial solutions. Based on the signal continuous coverage probability cloud map, calculate the W signal-to-noise ratio spatial gradients and W path loss change rates of the W compliant trial solutions relative to the first candidate deployment scheme, and calculate the W preliminary fitness vectors according to the preset information theory performance three objective function. Based on the W signal-to-noise ratio spatial gradients, W path loss change rates, and W preliminary fitness vectors, a local search iteration guided by physical field gradient information is performed in the first refined search space to obtain the first local refined solution. Similarly, local optimization is performed on multiple candidate deployment schemes in the diverse potential solution set to obtain multiple locally refined solutions, which constitute a set of locally refined solutions; The local refined solution set is projected onto the signal continuous coverage probability cloud map for accurate evaluation. After obtaining the accurate fitness vector set, front screening based on Pareto non-dominated sorting is performed to output the physical field compliant deployment scheme set.

5. The deployment method for mobile electromagnetic spectrum monitoring equipment integrating multi-physics sensing as described in claim 1, characterized in that, Electromagnetic propagation modeling with multi-physics coupling is performed in the target monitoring area to generate a probability cloud map of continuous signal coverage, including: Based on the geographic spatial boundary of the target monitoring area, digital elevation models and real-time meteorological data are retrieved from the geographic information database and meteorological data server, respectively, to construct a three-dimensional dynamic electromagnetic propagation field. Multiple distributed monitoring points are pre-set with initial coordinates in the three-dimensional dynamic electromagnetic propagation field; The initial coordinates of multiple distributed monitoring points are calculated based on ray tracing in the Fresnel zone to obtain the diffraction loss vectors of multiple monitoring points. Then, the real-time meteorological data is fused to perform dynamic atmospheric attenuation correction, and the signal propagation characteristic vectors of multiple monitoring points are output. Using the initial coordinates of the multiple distributed monitoring points as input, spatial interpolation and probability statistics are performed on the signal propagation characteristic vectors of the multiple monitoring points in the three-dimensional dynamic electromagnetic propagation field to generate the signal continuous coverage probability cloud map.

6. The deployment method for mobile electromagnetic spectrum monitoring equipment integrating multi-physics sensing as described in claim 5, characterized in that, The initial deployment schemes of the multiple equipment are projected onto the signal continuous coverage probability cloud map, and an evaluation based on a preset information theory performance three-objective function is performed to obtain fitness vectors for multiple schemes, including: The coordinates of multiple monitoring equipment in the initial deployment scheme of the first equipment are projected onto multiple three-dimensional spatial locations corresponding to the signal continuous coverage probability cloud map in order to extract multiple signal propagation characteristic data. Multiple achievable signal-to-noise ratio values ​​are extracted from the multiple signal propagation characteristic data. After calculating the multiple maximum information transmission rates supported by the multiple three-dimensional spatial locations based on Shannon's theorem, the values ​​are aggregated and summed to generate the first comprehensive information resolvability. After calculating multiple Euclidean displacements of the coordinates of the multiple monitoring equipment relative to the initial coordinates of the multiple distributed monitoring points, the displacements are summed, and the first total displacement cost is output. Based on multiple path losses and multiple terrain masking indicators in the multiple signal propagation characteristic data, a signal leakage risk assessment is performed to obtain a first signal leakage risk level. The first comprehensive information parseability, the first total displacement cost, and the first signal leakage risk are combined in a preset order to generate a first scheme fitness vector.

7. The deployment method for mobile electromagnetic spectrum monitoring equipment integrating multi-physics sensing as described in claim 6, characterized in that, Based on multiple path losses and multiple terrain masking indicators in the aforementioned multiple signal propagation characteristic data, a signal leakage risk assessment is performed to obtain a first signal leakage risk level, including: Using the multiple path losses as basic propagation loss data and the multiple terrain occlusion markers as terrain correction factors, calculate the multiple signal intensity fields generated by the multiple grid points corresponding to the coordinates of the multiple monitoring equipment in the target monitoring area. After spatially superimposing the multiple signal intensity fields onto a preset sensitive area distribution map, the signal intensity violation of the grid points in the superimposed area is determined according to the privacy protection threshold, so as to count the number of violation grid points. Calculate the percentage of the number of the illegal grid points in the total number of grids in the sensitive area distribution map, and output the first signal leakage risk level.

8. The deployment method for mobile electromagnetic spectrum monitoring equipment integrating multi-physics sensing as described in claim 4, characterized in that, The hard constraint compliance verification includes communication blind spot verification, signal leakage risk verification, and deployment exceeding limits verification.

Images