CN118884061B

CN118884061B - Electromagnetic environment analysis method and system combining situational awareness and space modeling

Info

Publication number: CN118884061B
Application number: CN202411358178.3A
Authority: CN
Inventors: 张云飞; 李海涛; 文凡; 张贤志
Original assignee: Wuhan Haihua Communication Technology Co ltd
Current assignee: Wuhan Haihua Communication Technology Co ltd
Priority date: 2024-09-27
Filing date: 2024-09-27
Publication date: 2024-12-20
Anticipated expiration: 2044-09-27
Also published as: CN118884061A

Abstract

The invention provides an electromagnetic environment analysis method and system combining situation awareness and space modeling, wherein the method comprises the steps of obtaining regional map information of a target region, constructing a regional three-dimensional model of the target region, generating device deployment points of a mobile detection device on the periphery of the target region, configuring unmanned aerial vehicle detection tasks of electromagnetic detection unmanned aerial vehicles and equipment detection tasks of electromagnetic signal acquisition and analysis equipment, controlling the electromagnetic detection unmanned aerial vehicles to execute the unmanned aerial vehicle detection tasks in a silent communication mode, simultaneously controlling all the electromagnetic signal acquisition and analysis equipment to execute the equipment detection tasks, obtaining first regional detection results through the mobile detection device, simultaneously obtaining second regional detection results of all the electromagnetic signal acquisition and analysis equipment, and analyzing the radiation source positions and the radiation source types of radiation sources in the target region by combining the first regional detection results and the second regional detection results. The invention has the effect of efficiently completing electromagnetic environment analysis in a large-scale and complex-terrain area.

Description

Electromagnetic environment analysis method and system combining situation awareness and space modeling

Technical Field

The invention belongs to the technical field of electromagnetic environment monitoring, and particularly relates to an electromagnetic environment analysis method and system combining situation awareness and space modeling.

Background

With the widespread use of wireless communication technology, the electromagnetic environment has become increasingly complex. To ensure proper operation of various wireless communication systems while preventing electromagnetic interference from potentially affecting the safety of the device, it is desirable to fully analyze and manage the radiation sources in the electromagnetic environment of a particular area.

Conventional electromagnetic environment analysis methods present a number of drawbacks in facing these new challenges. These methods typically rely on stationary monitoring equipment such as signal receivers and analyzers that are fixed in specific locations. While these devices are capable of providing continuous monitoring data, they have limited coverage and are difficult to fully monitor radiation sources in large areas. Particularly in areas of complex terrain or remote locations, the deployment of a sufficient number of stationary monitoring equipment tends to be costly and difficult to operate. Another common approach is to use a manually operated mobile device for measurement and analysis. Although the method increases flexibility, the method still has the problems of low efficiency, high labor cost, high safety risk and the like when facing a large-scale and complex-terrain area.

Disclosure of Invention

The invention provides an electromagnetic environment analysis method and system combining situation awareness and space modeling, which are used for solving the problem that the existing electromagnetic environment analysis method is low in efficiency when facing a large-scale and complex-terrain area.

In a first aspect, the invention provides an electromagnetic environment analysis method combining situation awareness and space modeling, which is applied to an electromagnetic environment detection system, wherein the electromagnetic environment detection system comprises three mobile detection devices with the same configuration, each mobile detection device comprises a mobile unmanned aerial vehicle base station and electromagnetic signal acquisition and analysis equipment, an electromagnetic detection unmanned aerial vehicle is deployed in each mobile unmanned aerial vehicle base station, communication connection is kept between any two mobile detection devices, the electromagnetic detection unmanned aerial vehicle is in communication connection with all mobile unmanned aerial vehicle base stations, and the electromagnetic detection unmanned aerial vehicle can stay in any mobile unmanned aerial vehicle base station;

the method comprises the following steps:

Acquiring regional map information of a target region needing electromagnetic environment analysis;

Constructing a regional three-dimensional model of the target region through a GIS (geographic information system) based on the regional map information;

generating device deployment points of all the movement detection devices, which are positioned around the target area in the area map information, by combining the area map information and the area three-dimensional model, wherein all the device deployment points are positioned at the edge of the area boundary of the target area and are positioned outside the target area, and the straight line distance between any one device deployment point and the other two device deployment points is equal;

deploying all the movement detection devices according to the device deployment points;

After all the mobile detection devices are deployed to the device deployment point, configuring an unmanned aerial vehicle detection task of the electromagnetic detection unmanned aerial vehicle and an equipment detection task of the electromagnetic signal acquisition and analysis equipment by combining the device deployment point and the area range of the target area in the area map information;

Transmitting the unmanned aerial vehicle detection task to the electromagnetic detection unmanned aerial vehicle currently deployed by the mobile unmanned aerial vehicle base station by utilizing the mobile unmanned aerial vehicle base station;

controlling all electromagnetic detection unmanned aerial vehicles to execute the unmanned aerial vehicle detection task in a silent communication mode so as to enable the electromagnetic detection unmanned aerial vehicles to detect the electromagnetic environment situation in the target area, and simultaneously controlling all electromagnetic signal acquisition and analysis equipment to execute the equipment detection task so as to enable the electromagnetic signal acquisition and analysis equipment to detect the electromagnetic environment situation in the target area;

acquiring first area detection results when all the electromagnetic detection unmanned aerial vehicles complete the unmanned aerial vehicle detection task through the movement detection device, and simultaneously acquiring second area detection results when all the electromagnetic signal acquisition and analysis equipment complete the equipment detection task;

and analyzing the radiation source positions and the radiation source types of all the radiation sources in the target area by combining the first area detection result and the second area detection result.

Optionally, the generating device deployment points of all the movement detection devices located around the target area in the area map information by combining the area map information and the area three-dimensional model includes the following steps:

Identifying the region range of the target region according to the region boundary of the target region in the region map information;

Taking the center of the maximum inscribed circle in the target area as a range expansion base point, and generating a virtual area with the equal proportion expansion of the target area according to a preset expansion proportion;

Taking a region between a virtual region boundary of a virtual region and the region boundary in the region map information as a preselected deployment region;

screening out all areas which do not meet preset detection requirements in the pre-selected deployment areas by using the area three-dimensional model, and taking all reserved pre-selected deployment areas as target deployment areas;

Taking all the position coordinate points in the target deployment area as an initial deployment point set;

Randomly selecting three position coordinate points from the initial deployment point set as initial deployment points based on a preset distance constraint condition, wherein the distance constraint condition is that the straight line distance between any one initial deployment point and the other two initial deployment points is equal;

Calculating to obtain a deployment point coverage area in the target area according to the deployment point coordinates of all the initial deployment points, wherein the deployment point coverage area is an overlapping area between a complete coverage area formed by all the initial deployment points and the target area, and the complete coverage area is an area surrounded by the shortest straight line segment between any two initial deployment points;

taking the maximized coverage area of the deployment point as an optimization target, and adjusting all initial deployment points to optimal deployment points by adopting an optimization algorithm based on the distance constraint condition;

And taking the optimal deployment point as a device deployment point of the movement detection device.

Optionally, the adjusting all the initial deployment points to the optimal deployment points by using the optimization algorithm based on the distance constraint condition with the coverage area of the deployment point as an optimization target includes the following steps:

constructing an objective function by taking the coverage area of the maximized deployment point as an optimization target;

Constructing a first initial population of an optimization algorithm based on all the initial deployment points;

Iteratively updating the first initial population by adopting the optimization algorithm based on the distance constraint condition and with the approach to the objective function as an optimization direction to obtain a first optimal population under the distance constraint condition;

and if the first optimal population meets the objective function, determining an optimal deployment point according to the first optimal population.

Optionally, the method further comprises the steps of:

if the first optimal population cannot meet the objective function, adjusting the distance constraint condition to a value that the linear distance difference between any one initial deployment point and the other two initial deployment points is smaller than a preset difference threshold;

taking the first optimal population as a second initial population of the optimization algorithm;

Based on the adjusted distance constraint condition and with the approach to the objective function as an optimization direction, iteratively updating the second initial population by adopting the optimization algorithm to obtain a second optimal population under the adjusted distance constraint condition;

And determining the optimal deployment point according to the second optimal population.

Optionally, the step of screening out all areas which do not meet a preset detection requirement in the pre-selected deployment area by using the area three-dimensional model, and taking all reserved pre-selected deployment areas as target deployment areas includes the following steps:

Obtaining the model heights of all boundary models positioned at the boundary of the target region in the region three-dimensional model;

marking the boundary model with the model height exceeding the obstacle height threshold value in the preset detection requirement as an ultrahigh boundary model, and acquiring the longitude and latitude coordinates of the model at the two ends of the ultrahigh boundary model in the regional three-dimensional model;

generating a first deployment forbidden region which does not meet the preset detection requirement in the pre-selected deployment region of the regional map information based on the longitude and latitude coordinates of the model;

screening out all the first forbidden deployment areas in the preselected deployment areas, and taking all the reserved preselected deployment areas as preferential deployment areas;

And deploying geological requirements based on the device in the preset detection requirements, performing a region screening step on the preferable deployment region by utilizing a GIS system, and taking all reserved preferable deployment regions as target deployment regions.

Optionally, the deploying geological requirements based on the device in the preset detection requirements and performing a region screening step on the preferred deployment region by using a GIS system, and taking all the reserved preferred deployment region as a target deployment region includes the following steps:

acquiring regional geological information of the preferable deployment region through the GIS system;

generating a second forbidden deployment region which does not meet the preset detection requirements in the preferred deployment region by combining the regional geological information and the device deployment geological requirements in the preset detection requirements;

Screening out all the second forbidden deployment areas in the preferred deployment area, and taking all the reserved preferred deployment areas as target deployment areas.

Optionally, the configuring the unmanned aerial vehicle detection task of the electromagnetic detection unmanned aerial vehicle and the equipment detection task of the electromagnetic signal acquisition and analysis equipment in combination with the device deployment point and the area range of the target area in the area map information includes the following steps:

generating an equipment sensing range of the electromagnetic signal acquisition and analysis equipment in the regional map information based on the device deployment point and according to equipment preset parameters of the electromagnetic signal acquisition and analysis equipment;

removing the overlapping part of the area range and the equipment sensing range, and taking the rest area range as the detection range of the electromagnetic detection unmanned aerial vehicle;

respectively configuring unmanned aerial vehicle detection tasks of the electromagnetic detection unmanned aerial vehicle in each mobile unmanned aerial vehicle base station based on the detection range and the device deployment point;

and configuring the equipment detection task of the electromagnetic signal acquisition and analysis equipment by combining the equipment perception range and the unmanned aerial vehicle detection time in the unmanned aerial vehicle detection task.

Optionally, the unmanned aerial vehicle detection task of the electromagnetic detection unmanned aerial vehicle is configured in each mobile unmanned aerial vehicle base station based on the detection range and the device deployment point, respectively, and includes the following steps:

determining an initial range division point in the detection range based on the device deployment points, wherein the linear distances between the initial range division point and all the device deployment points are equal;

Generating a segmentation boundary in the detection range with a shortest straight line between the initial range segmentation point and the device deployment point;

Allocating a base station detection range for each mobile unmanned aerial vehicle base station respectively by combining the segmentation boundary and the range boundary of the detection range;

for each mobile unmanned aerial vehicle base station, generating an unmanned aerial vehicle optimal path of the electromagnetic detection unmanned aerial vehicle by adopting an optimal path algorithm based on the base station detection range and taking another mobile unmanned aerial vehicle base station as a path end point, wherein all the mobile unmanned aerial vehicle base stations can only serve as the path end point of one unmanned aerial vehicle optimal path at most in the same unmanned aerial vehicle detection task;

calculating the flight time of the electromagnetic detection unmanned aerial vehicle in the optimal path of the unmanned aerial vehicle based on unmanned aerial vehicle flight parameters preset by the mobile unmanned aerial vehicle base station;

If the flight time is the maximum flight time in the flight times calculated by all the mobile unmanned aerial vehicle base stations, configuring an unmanned aerial vehicle detection task of the electromagnetic detection unmanned aerial vehicle by combining the unmanned aerial vehicle optimal path and the unmanned aerial vehicle flight parameters;

And if the flight time is not the maximum flight time in the flight times calculated by all the mobile unmanned aerial vehicle base stations, adjusting the unmanned aerial vehicle flight parameters until the flight time is the same as the maximum flight time, and configuring unmanned aerial vehicle detection tasks of the electromagnetic detection unmanned aerial vehicle by combining the optimal path of the unmanned aerial vehicle and the adjusted unmanned aerial vehicle flight parameters.

Optionally, the analyzing the radiation source positions and the radiation source types of all the radiation sources in the target area by combining the first area detection result and the second area detection result includes the following steps:

Analyzing and obtaining an electromagnetic environment situation sensing result in the target area by combining the first area detection result and the second area detection result;

Identifying the longitude and latitude data of the radiation sources of all the radiation sources in the target area based on the electromagnetic environment situation awareness result;

and carrying out cluster analysis on all the radiation sources by utilizing a cluster classification algorithm based on the longitude and latitude data of the radiation sources to obtain the radiation source types of all the radiation sources.

In a second aspect, the present invention also provides an electromagnetic environment analysis system combining situational awareness and spatial modeling, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the electromagnetic environment analysis method combining situational awareness and spatial modeling as described in the first aspect when executing the computer program.

The beneficial effects of the invention are as follows:

The invention can more accurately describe and analyze the geographic characteristics and electromagnetic environment of the target area by combining a GIS system and a three-dimensional modeling technology, and provides a solid foundation for subsequent detection tasks. Secondly, the uniformly distributed device deployment points generated by the intelligent algorithm ensure the whole coverage of the periphery of the target area, and greatly improve the comprehensiveness and accuracy of detection. Furthermore, the invention combines the mobile detection device, the electromagnetic detection unmanned aerial vehicle and the electromagnetic signal acquisition and analysis equipment to form a multi-layer and multi-angle detection network, and can simultaneously carry out omnibearing electromagnetic environment analysis on a target area from the ground and the air. Particularly, the electromagnetic detection unmanned aerial vehicle not only expands the detection range, but also can flexibly cope with complex terrains and environmental conditions, and overcomes the limitations of the traditional fixed equipment and manual operation mobile equipment. In addition, the silent communication mode is adopted to execute the detection task, so that the influence of the detection process on the electromagnetic environment of the target area is reduced to the maximum extent, and the authenticity and reliability of the detection result are ensured. Meanwhile, the base station of the mobile unmanned aerial vehicle is used for controlling the cooperative work of a plurality of unmanned aerial vehicles, so that the detection efficiency and the coverage range are greatly improved. Finally, the invention can more comprehensively and accurately determine the position and the type of each radiation source in the target area by comprehensively analyzing the detection results from different sources.

Drawings

Fig. 1 is a schematic system configuration diagram of an electromagnetic environment detection system according to an embodiment of the present application.

Fig. 2 is a flow chart of an electromagnetic environment analysis method combining situation awareness and spatial modeling in one embodiment of the application.

Fig. 3 is a schematic diagram of a target area and a device deployment point according to an embodiment of the present application.

Fig. 4 is a schematic view of virtual areas, pre-selected deployment areas, full coverage areas, and deployment point coverage areas in one embodiment of the present application.

FIG. 5 is a schematic view of an area where deployment is prohibited in one embodiment of the present application.

Fig. 6 is a schematic diagram illustrating generation of an optimal path of the unmanned aerial vehicle in the device perception range and the target area according to one embodiment of the present application.

Detailed Description

The technical solutions of the embodiments of the present application will be clearly described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which are obtained by a person skilled in the art based on the embodiments of the present application, fall within the scope of protection of the present application.

The terms first, second and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged, as appropriate, such that embodiments of the present application may be implemented in sequences other than those illustrated or described herein, and that the objects identified by "first," "second," etc. are generally of a type, and are not limited to the number of objects, such as the first object may be one or more. Furthermore, in the description and claims, "and/or" means at least one of the connected objects, and the character "/", generally means that the associated object is an "or" relationship.

The invention discloses an electromagnetic environment analysis method combining situation awareness and space modeling, which is applied to an electromagnetic environment detection system with reference to fig. 1, wherein the electromagnetic environment detection system comprises three mobile detection devices with the same configuration, each mobile detection device comprises a carrier vehicle, a mobile unmanned aerial vehicle base station and electromagnetic signal acquisition and analysis equipment, each mobile unmanned aerial vehicle base station is provided with an electromagnetic detection unmanned aerial vehicle, communication connection is kept between any two mobile detection devices, the electromagnetic detection unmanned aerial vehicle is in communication connection with all mobile unmanned aerial vehicle base stations, and the electromagnetic detection unmanned aerial vehicle can stay at any mobile unmanned aerial vehicle base station. Electromagnetic detection unmanned aerial vehicles are equipped with mobile electromagnetic monitoring lateral systems, electromagnetic detection unmanned aerial vehicles generally have the following performance parameters:

the monitoring/direction finding frequency range is 20-8000 MHz.

Real-time intermediate frequency bandwidth 240MHZ.

Scan speed >150GHz/s (25 kHz step).

Intermediate frequency phase noise is less than or equal to-100 dBc/Hz@10kHz (f=1 GHz).

And the modulation measurement analysis is AM/FM/CW/ASK/PSK/DPSK/QAM/FSK/MSK/QPSK and the like.

DDC channel number >32.

Burst signal interception capability 20us (100% probability).

Channel scan speed >500ch/s (25 kHz channel spacing)

The weight of the system is less than or equal to 9kg.

The working temperature is minus 20 ℃ to plus 55 ℃.

The electromagnetic signal acquisition and analysis equipment consists of an electromagnetic signal acquisition module, an electromagnetic signal analysis module, a detection antenna, a display control terminal, an electromagnetic environment situation sensing module and analysis software. Electromagnetic signal acquisition and analysis devices typically have the following performance parameters:

The frequency range is 100kHz-9GHz.

Instantaneous bandwidth 80Mhz.

Sweep rate up to 200GHz/S (RBW >100 kHz).

DANL down to-161 dBm/Hz, (> frequency band above 30 MHz).

The phase noise is as low as-105 dBc/Hz@10kHz1GHz carrier wave.

Analog demodulation AM, FM, WFM, NFM, USB, LSB, CW.

The detection mode is normal, sampling, positive peak value, negative peak value and RMS.

The system has the functions of intermediate frequency analysis, panoramic spectrum scanning, real-time scanning, discrete scanning, differential spectrum analysis, analog signal analysis, digital signal demodulation, signal identification and the like.

The equipment interface comprises an intermediate frequency output, a radio frequency input, a power switch, a reference input, a data communication port and the like.

The electromagnetic signal acquisition and analysis equipment can utilize high-efficiency modeling and real-time drawing technology to calculate electromagnetic propagation and coverage of various emission sources and analyze interference effects and electromagnetic spectrum occupation situations by uniformly modeling topography and electromagnetic data in a space electromagnetic environment and combining the influence of an antenna feeder system and topography parameters. And carrying out visualization processing on the electromagnetic environment from multiple dimensions, and intuitively displaying the distribution situation of the electromagnetic environment, electromagnetic compatibility analysis and electromagnetic radiation evaluation based on an electronic map.

FIG. 2 is a flow diagram of an electromagnetic environment analysis method combining situational awareness and spatial modeling in one embodiment. It should be understood that, although the steps in the flowchart of fig. 2 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or other steps. As shown in fig. 2, the electromagnetic environment analysis method combining situation awareness and space modeling disclosed by the invention specifically comprises the following steps:

s101, obtaining regional map information of a target region needing electromagnetic environment analysis.

Where the geographical boundaries of the target area, which may be administrative divisions, natural geographical units or specific item areas, need to be defined explicitly first. Terrain data for the area is then collected, including altitude, grade, type of surface coverage, etc. Such data typically comes from an open source geographic data platform. In practice, specialized geographic information system software, such as ArcGIS or QGIS, may be used to process and integrate such data. For example, a base topography layer may be imported and then satellite images superimposed to obtain up-to-date surface information. Then, vector data layers such as roads, buildings, barriers and the like are added, and a unified coordinate system and a projection mode are ensured to be used for all data.

S102, constructing a regional three-dimensional model of the target region based on regional map information through a GIS system.

Wherein a basic three-dimensional model of the terrain is created using Digital Elevation Model (DEM) data. DEM data is typically stored in a grid format, with each pixel representing the altitude of a geographic location. These elevation data can be converted into a continuous three-dimensional surface using the three-dimensional modeling function of the GIS software. For example, in ArcGIS, a Triangular Irregular Network (TIN) may be generated from a DEM using a "create TIN" tool, and then a smoother surface may be created using a "TIN to grid" tool. The surface coverage information is then added. This includes vegetation, bodies of water, buildings, and the like. For vegetation, a normalized vegetation index (NDVI) in satellite telemetry data can be used to distinguish between different types and densities of vegetation and represented in a three-dimensional model with three-dimensional objects of different heights and textures. The body of water can be modeled by setting a specific reflectivity and smoothness.

Three-dimensional modeling of buildings is the most complex part. First, starting from two-dimensional building contour data, these data are typically stored in a vector format. Each building is then assigned height information. This may be obtained by in-the-field measurements, liDAR data, or estimates. The two-dimensional contours can be converted into three-dimensional building models using the three-dimensional modeling function of GIS software, such as the ArcGIS "stretch" tool. For complex buildings, it may be necessary to create a more detailed model using specialized three-dimensional modeling software (e.g., sketchup or CITYENGINE) and then import the GIS system.

S103, combining the regional map information and the regional three-dimensional model to generate device deployment points of all the mobile detection devices positioned at the periphery of the target region in the regional map information.

Wherein, referring to fig. 3, first, an accurate boundary of a target area needs to be determined in a GIS system. A buffer is then created around the boundary of the target area. The width of this buffer zone depends on the effective detection range of the movement detection means and the size of the target area. Then, a part of the area unsuitable for deploying the mobile detection device in the buffer area is screened out through the area three-dimensional model, then uniformly distributed candidate deployment points are generated in the buffer area, and then a spatial clustering algorithm (such as K-means) is applied to select an optimal point set (such as point A, B, C in fig. 3). The key point is to ensure that the straight line distance between any one deployment point and the other two nearest deployment points is equal. This can be achieved by the following steps:

1. Initializing, namely randomly selecting N points in the buffer area as initial deployment points.

2. Calculate distance-for each point, calculate its distance to all other points.

3. And (3) adjusting the position, namely moving each point to form an equilateral triangle with the nearest two points. An iterative method may be used to fine-tune the position of the point each time until a predetermined accuracy is reached.

4. Steps 2 and 3 are repeated until the positions of all points are stable.

The point adjustment process can be expressed specifically by the following formula, P' i=pi+α (((pj+pk)/2) -Pi)

Where Pi is the current point position, pj and Pk are the positions of the nearest two points, α is the adjustment coefficient (0 < α < 1), and P' i is the new adjusted position.

The main advantages of this deployment approach are the ability to achieve uniform coverage of the target area, minimize the detection dead zone, and provide the ideal geometry for subsequent triangulation. By ensuring equidistant distribution among deployment points, signal intensity analysis and a source positioning algorithm can be simplified, and positioning accuracy is improved.

S104, deploying all the movement detection devices according to the device deployment points.

S105, after all the mobile detection devices are deployed to the device deployment point, combining the device deployment point and the region range of the target region in the region map information, and configuring an unmanned aerial vehicle detection task of the electromagnetic detection unmanned aerial vehicle and an equipment detection task of the electromagnetic signal acquisition and analysis equipment.

And the detection task is configured by combining the specific position of the device deployment point and the range of the target area in the area map information. The probing task should consider the following key factors:

And the coverage range is that the whole target area is ensured to be in the detection range, and no monitoring blind area is left.

And detecting frequency, namely determining the frequency and interval of detection according to the complexity and the change speed of the electromagnetic environment of the target area.

Resolution-determining the spatial and temporal resolution of the probe to meet the analysis requirements.

Energy efficiency-the detection path and time are optimized to maximize battery life and detection efficiency.

And detecting parameters, namely setting proper frequency range, signal strength threshold and other parameters according to the characteristics of the target area and the detection purpose.

For electromagnetic detection unmanned aerial vehicles, the following specific tasks need to be configured:

flight path-a flight path is designed which can cover the whole target area. A grid-like or spiral path is typically employed to ensure uniform coverage. For example, the mesh path may be generated using the following algorithm:

a. the target area is divided into grids of equal size.

B. The center point coordinates of each grid are calculated.

C. a shortest path is generated (e.g., using a traveler problem algorithm) that connects all the center points.

Fly height-an optimal fly height is determined based on terrain, building height, and performance characteristics of the detection device. Typically, fly height requires a balance between signal coverage and resolution.

Flight speed-the proper flight speed is set to ensure the quality and continuity of the detected data. Too fast a speed may result in undersampling of the data, while too slow a speed may affect the detection efficiency.

The detection frequency range is set to an appropriate frequency sweep range according to the type of electromagnetic signal that may be present in the target area. For example, if mobile communication signals are of interest, it may be desirable to cover a frequency band of 700MHz to 2.6 GHz.

Signal strength threshold-a minimum strength threshold for signal detection is set to filter background noise.

Data sampling rate-the data sampling frequency is set according to the required time resolution.

And (3) battery management strategy, namely setting a battery electric quantity warning threshold value and an automatic return point.

For the electromagnetic signal acquisition and analysis equipment on the ground, the following tasks need to be configured:

detection schedule-setting the time interval for periodic detection, it may be necessary to adjust the frequency according to the electromagnetic activity level of different periods.

Frequency sweep range-similar to unmanned aerial vehicles, but may cover a wider frequency range, as ground equipment typically has more processing power.

Signal analysis parameters-algorithm parameters for signal identification and classification, such as thresholds for signal modulation type identification, are set.

Data storage and transmission policies-determining the local storage capacity of the data and the frequency of transmission to the central system.

Interference detection threshold-signal strength and frequency signature thresholds are set to identify potential sources of interference.

In the configuration process, the cooperation between devices also needs to be considered. For example:

Time synchronization-ensuring that the clocks of all devices remain synchronized is critical to subsequent data fusion and analysis.

Frequency allocation-different frequency bands are allocated to different devices, where possible, to improve overall detection efficiency.

Data cross-validation, namely setting certain overlapped detection areas for cross-validation of data among different devices.

Dynamic task adjustment, which allows the system to dynamically adjust task parameters according to the real-time detection result. For example, if an abnormal signal is detected in a certain area, the sampling frequency of the area may be increased.

After configuration is completed, comprehensive system testing is required, including:

Communication test-ensuring that all devices can communicate with the control center stably.

And (3) data consistency testing, namely verifying the data consistency of different devices in the overlapping area.

Task execution testing, namely simulating an actual detection process to ensure that all devices can work normally according to a preset task.

Abnormal situation processing test, namely simulating various possible abnormal situations (such as equipment faults, communication interruption and the like) and testing the coping capability of the system.

Through the comprehensive and detailed task configuration process, the whole electromagnetic environment detection system can be ensured to operate efficiently and reliably, and high-quality original data is provided for subsequent data analysis. The method not only improves the detection efficiency and accuracy, but also enhances the flexibility and adaptability of the system, and can meet the monitoring requirements of various complex electromagnetic environments.

S106, the unmanned aerial vehicle base station is utilized to send an unmanned aerial vehicle detection task to the electromagnetic detection unmanned aerial vehicle currently deployed by the mobile unmanned aerial vehicle base station.

The mobile unmanned aerial vehicle base station is used as a central center of the whole system, and needs to have strong computing power and communication power. It is typically equipped with a high performance processor, mass storage device, and multi-mode communication module. The main functions of the base station include:

And (3) task planning and scheduling, namely generating a specific unmanned aerial vehicle flight path and detection parameters according to the detection tasks configured in the previous steps.

And (3) real-time communication, namely, continuous two-way communication with the electromagnetic detection unmanned aerial vehicle.

And the data processing and storage is used for receiving, processing and storing the detection data returned by the unmanned aerial vehicle.

And the state monitoring is used for monitoring the flight state, the battery electric quantity, the working state of the detection equipment and the like of the unmanned aerial vehicle in real time.

Before sending the task to the drone, the mobile drone base station needs to perform the following steps:

path planning, namely generating an optimized flight path for each unmanned aerial vehicle, wherein the considerations comprise terrain, obstacles, no-fly areas and the like.

And setting task parameters, namely setting specific parameters of each task, such as a detection frequency range, a signal strength threshold value, a data sampling rate and the like, according to detection requirements.

The task delivery process typically takes the following steps:

And establishing a communication link, namely firstly establishing a safe communication link between the base station and the target unmanned aerial vehicle. This typically uses an encrypted wireless communication protocol such as AES encrypted WiFi or dedicated drone control channel.

And the identity authentication is that the base station and the unmanned aerial vehicle mutually authenticate the identity, so that the communication safety is ensured. This may involve the exchange of digital certificates or encryption keys.

Task data packing, namely packing task information into a standard format by the base station. A typical task data packet may contain information such as task ID, flight path coordinate sequence, probe parameters (frequency range, sampling rate, etc.), time window (start time, end time), emergency handling instructions.

Data transmission, namely sending the task data packet to the unmanned aerial vehicle by using a reliable transmission protocol (such as TCP/IP).

And the task activation step of sending a task activation instruction after the base station receives the confirmation, and starting to execute the task by the unmanned aerial vehicle.

The mobile unmanned aerial vehicle base station can effectively control and coordinate the work of a plurality of electromagnetic detection unmanned aerial vehicles. This not only ensures accurate execution of the probing task, but also improves the flexibility and efficiency of the overall system. The base station can dynamically adjust the tasks according to real-time conditions, for example, when an abnormal signal is found, the base station immediately sends an additional detection instruction to a nearby unmanned aerial vehicle. The method greatly enhances the adaptability and response speed of the electromagnetic environment detection system, so that the electromagnetic environment detection system can better cope with complex and changeable electromagnetic environments.

S107, controlling all electromagnetic detection unmanned aerial vehicles to execute unmanned aerial vehicle detection tasks in a silent communication mode so as to enable the electromagnetic detection unmanned aerial vehicles to detect electromagnetic environment situations in a target area, and simultaneously controlling all electromagnetic signal acquisition and analysis equipment to execute equipment detection tasks so as to enable the electromagnetic signal acquisition and analysis equipment to detect the electromagnetic environment situations in the target area.

The silent communication mode refers to that electromagnetic radiation of the unmanned aerial vehicle is reduced as much as possible when the unmanned aerial vehicle executes a detection task so as not to interfere with an electromagnetic environment of a target area. The following techniques are generally employed to implement the silent communication mode:

Low power communication, namely, using an ultra-low power communication module, and transmitting data with short distance and low bandwidth only when necessary.

Optical communication-short-range communication using laser or infrared, where possible, completely avoids radio frequency radiation.

And the passive receiving is that the unmanned aerial vehicle is mainly in a passive receiving state and returns only when a specific instruction of the base station is received.

And (3) time division multiplexing, namely, a plurality of unmanned aerial vehicles are communicated with the base station in turn according to a preset time schedule, so that the requirement of simultaneous communication is reduced.

Directional antennas-using highly directional antennas, communication is performed only in the necessary direction, reducing radiation spread.

When the unmanned aerial vehicle detection task is executed, each unmanned aerial vehicle flies according to the path generated by the base station, and electromagnetic signal detection is carried out at the same time. The probing process generally includes the steps of:

and (3) signal scanning, namely continuously scanning a preset frequency range by electromagnetic detection equipment carried by the unmanned aerial vehicle. For example, flexible frequency scanning may be implemented using Software Defined Radio (SDR) technology.

And detecting signals, namely recording the frequency, the intensity, the direction and other characteristics of the signals when the signals exceeding the preset threshold value are detected. This is typically done using Digital Signal Processing (DSP) techniques to achieve real-time analysis.

And (3) data recording, namely associating the detected signal characteristics with the current time stamp and GPS coordinates, and storing the detected signal characteristics in a local memory.

Preliminary analysis the drone may perform some preliminary signal analysis, such as modulation type identification, signal classification, etc., to reduce the amount of data that needs to be transmitted.

Exception reporting-if a particularly strong or unusual signal is detected, the drone may immediately report to the base station, even though this may temporarily break the silence state.

Meanwhile, the electromagnetic signal acquisition and analysis equipment on the ground is also executing the detection task. These devices typically have greater processing power and wider frequency coverage. Their working procedures include:

and high-precision analysis, namely performing real-time signal analysis by using a high-performance signal processor, wherein the real-time signal analysis comprises signal demodulation, feature extraction, interference identification and the like.

Directional direction finding-using a multi-antenna array for direction of arrival (DOA) estimation of signals, to assist in locating the radiation source.

And storing the acquired original data and analysis results in a mass storage device to prepare for subsequent deep analysis.

And alarming in real time, namely immediately sending an alarm to a control center when a preset abnormal signal is detected.

S108, acquiring first area detection results when all electromagnetic detection unmanned aerial vehicles complete unmanned aerial vehicle detection tasks through a mobile detection device, and simultaneously acquiring second area detection results when all electromagnetic signal acquisition and analysis equipment complete equipment detection tasks.

The unmanned aerial vehicle may transmit data back to the mobile detection device by using various communication modes, including a high-bandwidth wireless link such as a 4G/5G cellular network or a special long-distance wireless communication system, satellite communication such as laser communication suitable for long-distance or complex terrain, and optical communication such as laser communication suitable for short-distance, high-bandwidth and low-interference requirements. The security is ensured by adopting a strong encryption algorithm (such as AES-256) in the data transmission process. The amount of data transferred is reduced using efficient data compression algorithms (e.g., LZMA 2). The first region probe result data is typically in a standardized data format, such as JSON or Protocol Buffers, to facilitate data exchange between different systems. The first area detection result data may include information including a time stamp, GPS coordinates, detected signal characteristics (frequency, intensity, direction, etc.), and drone status information (battery level, altitude, speed, etc.).

The second region detection result data acquired by the electromagnetic signal acquisition and analysis device may include original signal data, a spectrum analysis result, a signal modulation type identification result, a direction finding result and the like. Since electromagnetic signal acquisition and analysis equipment is generally stationary, a wired network (e.g., optical fiber) may be used for high-speed, reliable data transmission. For sites that are remote or difficult to access a wired network, a wireless link of a high gain directional antenna may be used. The use of an accurate time synchronization mechanism (such as NTP or PTP) ensures that the time stamps of all devices are consistent. A data synchronization protocol is implemented to ensure that data from different devices can be properly aligned.

Data from the drone and the ground equipment are spatially and temporally aligned. And a data fusion algorithm (such as Kalman filtering) is used for combining the data from different sources, so that the overall accuracy is improved. And performing spatial interpolation on the discrete measurement points to generate a continuous electromagnetic field intensity distribution map. Statistical methods such as Kriging or Inverse Distance Weighting (IDW) may be used. And (3) carrying out time sequence analysis on the continuously acquired data, and identifying a time change mode of the electromagnetic environment. Fourier transform or wavelet analysis methods may be used. In combination with the data of the plurality of measurement points, a triangulation or multi-point localization algorithm is used to estimate the position of the radiation source. Taking into account terrain and building factors, it may be desirable to use advanced propagation models such as ray tracing.

S109, analyzing the detection result of the first area and the detection result of the second area to obtain the positions and the types of the radiation sources of all the radiation sources in the target area.

In one embodiment, generating device deployment points of all the movement detection devices located around the target area in the area map information by combining the area map information and the area three-dimensional model includes the steps of:

Taking the center of the maximum inscription circle in the target area as a range expansion base point, and generating a virtual area with the same-proportion expansion of the target area according to a preset expansion proportion;

Taking a region between virtual region boundaries of the virtual region and the region boundaries in the region map information as a preselected deployment region;

Screening out all areas which do not meet the preset detection requirements in the pre-selected deployment areas by using an area three-dimensional model, and taking all reserved pre-selected deployment areas as target deployment areas;

Randomly selecting three coordinate points from the initial deployment point set as initial deployment points based on a preset distance constraint condition, wherein the distance constraint condition is that the straight line distance between any one initial deployment point and other two initial deployment points is equal;

The deployment point coverage area in the target area is obtained according to the deployment point coordinates of all the initial deployment points, wherein the deployment point coverage area is an overlapping area between a complete coverage area formed by all the initial deployment points and the target area, and the complete coverage area is an area surrounded by the shortest straight line segment between any two initial deployment points;

taking the coverage area of the maximized deployment point as an optimization target, and adjusting all initial deployment points to the optimal deployment point by adopting an optimization algorithm based on a distance constraint condition;

In this embodiment, referring to fig. 4, identifying the region range of the target region from the region boundary of the target region in the region map information is a key initial step, which lays a foundation for subsequent analysis and planning. This process typically involves a variety of Geographic Information System (GIS) technologies and image processing methods. First, it is necessary to load and process regional map information, including vector data (e.g., shapefiles) and raster data (e.g., satellite images). For vector data, boundary information can be directly extracted, and for raster data, image segmentation and edge detection algorithms are required. Common edge detection algorithms include Canny algorithm, sobel algorithm, and the like. For example, the steps of using the Canny algorithm include Gaussian filter denoising, image gradient calculation, non-maximum suppression, dual-threshold detection, and edge tracking.

The identified boundaries may require further processing such as smoothing (using the Douglas-Peucker algorithm) and topology error correction. The results of boundary recognition are typically stored in the form of polygons, each vertex having precise geographic coordinates. This polygon represents the exact extent of the target area and provides a spatial reference for subsequent analysis. In practice, challenges that may be encountered include handling complex topographical features (such as coastlines), handling changes in administrative boundaries, and handling errors and inconsistencies in data sources. To address these challenges, techniques such as multi-phase data alignment, geocoding correction, and the like may be employed. In addition, the conversion between different coordinate systems needs to be considered, so that all data can be processed under the same coordinate system.

Next, the largest inscribed circle needs to be found within the identified target region. This can be achieved by 1) discretizing the target area into a grid, 2) calculating the shortest distance to the boundary for each grid point, 3) selecting the point with the largest distance as the center of the circle, and the distance to the boundary being the radius. This process may be optimized using a distance transformation algorithm (e.g., euclidean distance transformation). In practical applications, computational efficiency may need to be considered, especially for large or complex-shaped areas. One optimization method is to recursively divide the space using a quadtree structure, rapidly excluding regions that are unlikely to contain the largest inscribed circles.

After finding the maximum inscribed circle, its center coordinates (x_c, y_c) will be the base point of the expansion (point O in fig. 4). Assuming that the preset expansion ratio is k (k > 1), the expansion process can be described as that for each point (x, y) on the boundary of the target region, the corresponding point (x ', y') on the boundary of the new virtual region can be calculated by the following formula:

x'=x_c+k(x-x_c)

y'=y_c+k(y-y_c)

This transformation maintains the shape of the target area, but is scaled up in size. In practice, a large number of boundary points may need to be processed, so the use of vectorization operations may be considered to improve efficiency. In addition, care must be taken to address special situations, such as the potential for selfing of the recessed region after expansion, where additional geometric operations are required to address the topology problem. The enlarged virtual region boundaries may require further processing, such as smoothing or simplification, to reduce computational complexity. The Douglas-Peucker algorithm may be used to simplify the boundaries while maintaining the dominant shape features. In addition, it is also necessary to ensure that the enlarged virtual area does not collide with other important geographic features (such as protection areas, water bodies, etc.), and spatial analysis and adjustment are required to be performed in combination with the GIS data. The virtual region provides a larger operating space for subsequent deployment point selection while maintaining geometric similarity to the original target region. The size of the virtual area can be flexibly controlled by adjusting the expansion ratio k, so that the coverage area and the resource utilization are balanced.

The region map information is used as a pre-selected deployment region in the region map information between the virtual region boundary of the virtual region and the region boundary, and the process is actually performed by performing a spatial difference set operation, namely 'subtracting' the original target region from the enlarged virtual region to obtain an annular region. The implementation is as follows, firstly, the original target area and the virtual area need to be represented as polygonal objects. This typically involves the processing of vector data, each polygon being defined by a series of vertex coordinates. The polygon difference set operation is then performed using a spatial analysis library (e.g., a geometric processing tool such as GDAL, shape, or ArcGIS). Mathematically, this operation can be expressed as pre-selected deployment region = virtual region-target region, where "-" represents a spatial difference set operation.

In practice, if the shape of the target region is very complex or contains multiple discrete parts, the difference set operation may produce multiple separate polygons. In this case, it is necessary to decide whether to treat all these separate areas as pre-selected deployment areas or to select only the largest contiguous area. To increase computational efficiency, particularly for large-scale or high-resolution map data, spatial indexing techniques (e.g., R-trees) may be employed to accelerate geometric operations. Furthermore, if the shape of the preselected deployment region is too complex, a simplification process may be required, such as using the Douglas-Peucker algorithm to reduce the number of vertices while maintaining the dominant shape features.

And screening out all areas which do not meet the preset detection requirement in the pre-selected deployment area by using the area three-dimensional model, and taking all reserved pre-selected deployment areas as target deployment areas, wherein the aim of the step is to ensure that the finally selected deployment areas are suitable in geographic position and can meet the requirement in electromagnetic detection effect. Specifically, a regional three-dimensional model needs to be loaded and processed first. This model typically contains terrain elevation data (e.g., digital elevation model DEM) and surface characterization data (e.g., buildings, vegetation, etc.). The accuracy and detail level of the model directly affect the accuracy of the screening.

The preset detection requirements generally include the following aspects:

1. Line of sight visibility-the deployment point needs to have a good line of sight to the target area.

2. Signal coverage, namely considering the propagation characteristics of electromagnetic waves, ensuring that a detection signal can effectively cover a target area.

3. Terrain suitability-deployment points need to be at terrain-suitable locations, avoiding steep or unstable areas.

4. Access convenience, namely considering the accessibility of infrastructure such as roads.

5. Environmental factors such as avoiding sources of electromagnetic interference, taking into account climatic conditions, etc.

To evaluate these requirements, a series of spatial analyses are required:

1. Line-of-sight analysis-line-of-sight analysis algorithms (e.g., ray tracing) are used to evaluate the visual range of each potential deployment point. This requires consideration of the effects of terrain and shielding of buildings.

2. Signal coverage analysis-electromagnetic wave propagation models (such as Longley-Rice model or COST-231 model) are used to model the propagation of signals in three-dimensional space. This requires consideration of topography, building reflection and diffraction, etc.

3. And (3) terrain analysis, namely calculating terrain parameters such as gradient, slope direction and the like, and identifying a flat area suitable for equipment installation.

4. Accessibility analysis-the distance and ease of each potential deployment point to the nearest road or infrastructure is assessed using a network analysis algorithm.

5. Environmental factor analysis, namely evaluating the suitability of the environment by combining other geographic data (such as land utilization diagrams, meteorological data and the like).

These analyses may be implemented by a trellis operation or vector analysis. For example, the pre-selected deployment region may be converted to a high resolution grid, then each grid element is analyzed as described above, and finally grid elements that meet all conditions are reconverted to vector polygons. In practice, it may be necessary to set weights and thresholds to balance the different requirements. For example, a multi-criteria decision analysis (MCDA) method, such as Weighted Linear Combination (WLC) or Analytic Hierarchy Process (AHP), may be used to comprehensively evaluate the suitability of each location. The screening process may be expressed as a series of boolean operations or fuzzy logic operations:

target deployment area = preselected deployment area ∈ (line of sight visible area region Σ signal coverage area Σ topography suitable area Σ access convenient area Σ environment suitable area)

Where ". U" represents a spatial intersection operation.

The above process can obtain a strictly screened target deployment area, which not only meets the geographical location requirement, but also ensures that the equipment deployed therein can effectively execute the detection task. The method has the advantages that a plurality of influencing factors are comprehensively considered, and the success rate and effect of subsequent deployment can be greatly improved. Furthermore, this step may be computationally intensive, especially for large or complex areas. To increase efficiency, parallel computing techniques such as GPU acceleration or distributed computing may be considered. Meanwhile, a multi-resolution analysis strategy can be adopted, the low-resolution data is firstly used for coarse screening, and then the high-resolution data is used for fine analysis in a key area.

In one embodiment, with the maximized deployment point coverage area as an optimization target, adjusting all initial deployment points to an optimal deployment point based on the distance constraint condition and by adopting an optimization algorithm comprises the following steps:

constructing a first initial population of an optimization algorithm based on all initial deployment points;

based on the distance constraint condition and with the approaching objective function as an optimization direction, iteratively updating the first initial population by adopting an optimization algorithm to obtain a first optimal population under the distance constraint condition;

In this embodiment, if the first optimal population cannot meet the objective function, the distance constraint condition is adjusted to be that the difference value of the linear distance between any one initial deployment point and the other two initial deployment points is smaller than a preset difference value threshold;

Based on the adjusted distance constraint condition and with the approaching objective function as an optimization direction, iteratively updating the second initial population by adopting an optimization algorithm to obtain a second optimal population under the adjusted distance constraint condition;

and determining an optimal deployment point according to the second optimal population.

In this embodiment, constructing the objective function with the maximized deployment point coverage area as the optimization objective may translate the problem into a form that can be solved mathematically. The design of the objective function directly affects the effect and efficiency of the optimization. By adjusting the form and parameters of the objective function, different actual demands and constraint conditions can be flexibly adapted. And then constructing a first initial population of the optimization algorithm based on all initial deployment points, wherein the quality and diversity of the initial population directly relate to whether the optimization algorithm can find a global optimal solution. In this embodiment, each individual represents a set of possible deployment point locations.

One common method of constructing an initial population is to randomly perturb based on the initial deployment point. The method comprises the following specific steps:

Each initial deployment point is represented as a vector (x, y), where x and y are geographic coordinates.

For each initial deployment point, a plurality of variants are generated:

(x',y')=(x+δx,y+δy)

Where δx and δy are small perturbations randomly generated from a range, such as sampling from a normal distribution N (0, σ 2).

Ensuring that the generated new point remains within the target deployment area. If not, it may be regenerated or projected to the nearest feasible location.

The above process is repeated until a sufficient number of individuals are generated. Typical population sizes may range from 50 to 200, depending on the complexity of the problem and the computational resources.

Another approach is to use a heuristic to generate the initial population. For example, a k-means clustering algorithm may be used to generate uniformly distributed points within the target deployment area, and then randomly perturb around those points. This approach ensures that the initial population has a good spatial distribution. To increase the diversity of the population, various generation methods may be combined. For example, 70% of individuals are generated using random perturbation, 20% are generated using clustering methods, and the remaining 10% are completely randomly generated. This allows a balance to be struck between local searching and global exploration.

And then, based on the distance constraint condition and with the approaching objective function as an optimization direction, iteratively updating the first initial population by adopting an optimization algorithm to obtain a first optimal population under the distance constraint condition. This process involves multiple iterations, each of which includes operations such as selection, crossover, mutation, and evaluation. The choice of the optimization algorithm has a significant impact on the results. Common algorithms include Genetic Algorithms (GA), particle Swarm Optimization (PSO), differential Evolution (DE), and the like. Taking genetic algorithm as an example, a typical iterative process is as follows:

Selecting, selecting excellent individuals according to fitness (objective function value). A tournament selection method may be used.

Crossing, namely selecting two parent individuals to generate new offspring. For example, for two parent deployment points (x 1, y 1) and (x 2, y 2), children may be generated:

(x',y')=(αx1+(1-α)x2,αy1+(1-α)y2)

Where α is a random number between [0,1 ].

Variation, namely random variation is carried out on individuals with small probability, so that population diversity is increased. For example, small amplitude random movements are made to the selected points.

And (3) evaluating, namely calculating the fitness value of the newly generated individual.

Replacement, namely replacing the old population with the new population, and reserving elite individuals.

In this process, the distance constraint needs to be specially handled. A penalty function approach may be employed that reduces the fitness value of an individual if it violates a distance constraint. In the iterative process, the algorithm parameters need to be dynamically adjusted. For example, the variability may be gradually reduced as the number of iterations increases to achieve a transition from global exploration to local refinement. The stop condition may be set to reach a maximum number of iterations, the fitness value no longer improves significantly, or to reach a preset target value. Through multiple iterations, individuals in the population will gradually converge toward the optimal solution. The optimization process not only improves the coverage effect, but also ensures that the distance between deployment points meets the requirements, and provides a feasible scheme for actual deployment.

And calculating an objective function value of the first optimal population, and if the objective function value meets the objective function, determining an optimal deployment point according to the first optimal population. The optimal individuals in the first optimal population are directly selected as a final scheme, so that a group of optimal deployment points meeting all requirements is obtained. These points not only theoretically maximize the coverage area, but also satisfy the constraints of the actual deployment.

If the objective function value obtained by calculating the first optimal population cannot meet the objective function, adjusting the distance constraint condition to the point that the linear distance difference between any one initial deployment point and the other two initial deployment points is smaller than a preset difference threshold. This process aims to relax constraints and provide a larger search space for the optimization algorithm. Specific embodiments of adjusting the distance constraint are as follows:

a difference threshold epsilon is defined. This value should be set according to the specific case of the problem and may be set to 10% of the original distance requirement, for example.

The new distance constraint can be expressed as:

|d_ij-d_ik|<ε

where d_ij and d_ik are the distances of point i to point j and point k, respectively.

And updating the checking mode of the constraint condition in the optimization algorithm.

The implementation effect of the above steps is to create a larger search space for the optimization algorithm, increasing the likelihood of finding a satisfactory solution. By allowing for small variations in distance between deployment points, an excellent layout that is ignored under strictly equidistant conditions may be found.

Then, the first optimal population is used as a second initial population of the optimization algorithm, so that the convergence speed of a secondary optimization algorithm process can be realized, and the possibility of finding a better solution is improved. The specific implementation mode is as follows:

The first optimal population is evaluated by first re-calculating fitness values for each individual in the first optimal population. This is because changes in constraints may result in previous fitness estimates not being accurate.

And sorting and screening, namely sorting individuals in the first optimal population according to the new fitness value. A proportion (e.g., the top 50%) of the top-ranked individuals may be selected to remain as the basis for the second initial population.

Population expansion in order to maintain diversity of the population, it is necessary to expand the population on the basis of the remaining excellent individuals. The following method may be employed:

a) And carrying out small-amplitude random variation on the reserved individuals to generate new individuals.

B) Using crossover operations, the remaining individuals are paired two by two, generating new offspring.

C) A proportion (e.g., 10-20%) of new individuals are introduced that are completely randomly generated to increase population diversity.

Population balance, namely ensuring that the size of the expanded population is the same as that of the first initial population so as to keep the consistency of the algorithm.

Initializing parameters, namely, according to new constraint conditions and optimization targets, some algorithm parameters such as mutation rate, crossover rate and the like may need to be adjusted. Typically, these parameters may be slightly increased to increase the exploratory capacity of the population.

And then, based on the adjusted distance constraint condition and with the approximated objective function as an optimization direction, iteratively updating the second initial population by adopting an optimization algorithm to obtain a second optimal population under the adjusted distance constraint condition. The specific implementation process is as follows:

initializing, namely using the second initial population generated in the previous step as a starting point. This population already contains individuals of higher quality, helping to find good solutions quickly.

Fitness evaluation-fitness values are calculated for each individual in the population. The fitness function needs to take into account new distance constraints, which can be expressed as:

F=O(x)-λ*P(x)

Where O (x) is the original objective function, P (x) is the penalty function based on the new distance constraint, and λ is the weight coefficient.

Selection operations-selection of superior individuals for reproduction using appropriate selection methods, such as tournament selection. The use of elite retention strategies can be considered to ensure that optimal individuals are not lost during evolution.

And (3) intersecting, namely selecting parent individuals to intersect and generating new offspring. Various crossover methods may be used, such as single point crossover, two point crossover, or uniform crossover.

And (3) performing mutation operation, namely performing random mutation on part of individuals to maintain population diversity. The variation may be a small amplitude random adjustment of the deployment point coordinates. The variability may decrease as the number of iterations increases to achieve a transition from global exploration to local refinement.

Constraint processing, namely ensuring that individuals after crossing and mutation still meet the adjusted distance constraint condition. Individuals that do not meet the constraints may be treated using repair policies or regeneration policies.

And (5) iterative updating, namely repeating the steps 2-6 until the termination condition is met. The termination condition may be that the maximum number of iterations is reached, the fitness value is no longer significantly improved, or a preset target value is reached.

Parameter adaptation, namely, considering the use of an adaptive parameter adjustment strategy, such as dynamically adjusting the crossover rate and the mutation rate according to population diversity so as to balance exploration and utilization.

And (3) carrying out local search on the current optimal individual after a certain algebra, for example, using a simulated annealing or mountain climbing algorithm to further improve the quality of the solution.

Diversity maintenance, monitoring population diversity, if diversity falls below a certain threshold, introducing new individuals generated randomly or increasing mutation rate.

The implementation effect of the steps is to obtain a high-quality solution set under the new constraint condition. Because the starting point is better (the second initial population), the optimization process may converge faster than the first optimization. Meanwhile, due to the relaxation of constraint conditions, a better solution which cannot be achieved under the original constraint can be found.

In one embodiment, the step of screening out all areas which do not meet the preset detection requirement in the pre-selected deployment area by using the area three-dimensional model and taking all reserved pre-selected deployment areas as target deployment areas comprises the following steps:

obtaining the model height of all boundary models positioned at the boundary of the target region in the region three-dimensional model;

Marking a boundary model with the model height exceeding the obstacle height threshold value in the preset detection requirement as an ultrahigh boundary model, and acquiring longitude and latitude coordinates of the model in the region three-dimensional model at two ends of the ultrahigh boundary model;

generating a first deployment inhibition area which does not meet preset detection requirements in a pre-selected deployment area of the regional map information based on the longitude and latitude coordinates of the model;

screening out all first deployment forbidden areas in the pre-selected deployment areas, and taking all reserved pre-selected deployment areas as preferred deployment areas;

In this embodiment, the steps of deploying geological requirements based on the device in the preset detection requirements and performing region screening on the preferred deployment region by using the GIS system, and taking all the reserved preferred deployment regions as target deployment regions include the following steps:

Acquiring regional geological information of a preferable deployment region through a GIS system;

Generating a second forbidden deployment area which does not meet the preset detection requirements in the optimal deployment area by combining the regional geological information and the device deployment geological requirements in the preset detection requirements;

All second forbidden deployment areas in the preferred deployment area are screened out, and all reserved preferred deployment areas are taken as target deployment areas.

In the present embodiment, referring to fig. 5, acquiring the model height of all the boundary models located at the boundary of the target region in the region three-dimensional model involves processing and analysis of three-dimensional space data, specifically, first, it is necessary to accurately locate the boundary of the target region in the three-dimensional model. In practice this may involve sampling of Digital Elevation Model (DEM) or Triangular Irregular Network (TIN) data. For example, a height value may be extracted at fixed distances (e.g., 1 meter or 5 meters, depending on the required accuracy) along the boundary line. For complex terrain or buildings, it may be desirable to use more advanced interpolation techniques, such as bilinear interpolation or kriging, to obtain more accurate altitude estimates. In processing large-scale data, it may be desirable to use spatial indexes (e.g., R-trees or quadtrees) to optimize query efficiency. In addition, coordinate system consistency between different data sources needs to be considered, and coordinate conversion may be needed. Ultimately, this step will generate a data set containing boundary locations and corresponding heights, providing the basis for subsequent analysis. The effect of this step is to obtain an accurate boundary height model that reflects the change in elevation of the terrain and buildings surrounding the target area, providing critical information for assessing potential line of sight blockage and signal propagation.

And then marking a boundary model with the model height exceeding the obstacle height threshold value in the preset detection requirement as an ultrahigh boundary model, and acquiring the longitude and latitude coordinates of the model in the three-dimensional model of the region at the two ends of the ultrahigh boundary model. This step first requires setting a suitable height threshold, which is typically determined based on the performance characteristics of the detection device and the desired detection effect. For example, if the effective detection height of the detection device is 50 meters, the height threshold may be set to 45 meters, leaving a certain safety margin. Next, all the boundary height data obtained in the previous step need to be traversed, and the height of each point is compared with a threshold value. For points exceeding the threshold, a mark is required. This may be achieved by adding a boolean flag bit in the data structure.

It is important to note the continuous superelevation points, as they may represent a complete obstacle (e.g., a tall building or a mountain). To obtain the coordinates of both ends of the ultra-high boundary model, the start and end points (e.g., points C and D in fig. 5) in the marker sequence need to be detected. This can be achieved by checking the marking status of the neighboring points. Once the range of the ultra-high boundary model is determined, the precise longitude and latitude coordinates of its two endpoints need to be extracted. This typically involves a transformation of the coordinate system, as the three-dimensional model may use a local coordinate system, which needs to be transformed into a standard geographic coordinate system (e.g., WGS 84). The effect of this step is to obtain the exact location and extent of a range of potential line of sight obstructions or signal propagation obstructions, providing important constraints for subsequent deployment area planning.

And then, generating a first deployment forbidden region which does not meet the preset detection requirement in a pre-selected deployment region of the regional map information based on the longitude and latitude coordinates of the model. The purpose of this step is to determine those areas that are unsuitable for deployment of the detection device due to tall obstacles. First, longitude and latitude coordinates of the ultrahigh boundary model obtained in the previous step need to be imported into the GIS system. These coordinate points represent potential line of sight obstructions or signal propagation obstructions. Next, an appropriate buffer distance needs to be determined based on the performance characteristics of the detection device and the height of the obstacle. This distance can be estimated by a simple geometric model, e.g. buffer distance = (obstacle height-detection device height)/tan (minimum elevation). Where the minimum elevation angle is the minimum vertical angle at which the detection device can operate effectively. Using this formula, a specific buffer distance can be calculated for each superhigh boundary model. Then, a corresponding buffer is created centering on each ultra-high boundary model. These buffers constitute a first deployment-prohibited area. When creating the buffer, care needs to be taken to handle the overlapping situation. The intersecting buffers may be merged using a spatial merging (Dissolve) operation to obtain one contiguous exclusion zone. In addition, it is desirable to spatially intersect these exclusion areas with the preselected deployment area to ensure that only the exclusion areas within the preselected area remain.

Next, all first undeployed areas in the preselected deployment area are screened out and all reserved preselected deployment areas are taken as preferred deployment areas. This step is actually performed as a spatial difference set operation with the objective of removing those areas from the original pre-selected deployment area that are determined to be unsuitable for deployment, resulting in a more optimal deployment area. First, both the pre-selected deployment area and the first forbidden deployment area need to be represented as polygonal geometric objects. This typically involves the processing of vector data, which can be accomplished using a geospatial data processing library such as GDAL/OGR. Next, a spatial difference set operation is performed, which may be expressed as preferred deployment area = preselected deployment area-first forbidden deployment area.

In practice, this process may create complex geometries, including possible islands (small allowed areas completely surrounded by forbidden areas) or elongated connected areas. Thus, post-processing may be required, such as removing areas of too small an area or simplifying complex boundaries. This can be achieved by setting a minimum area threshold and using the Douglas-Peucker algorithm. In addition, it is necessary to check the connectivity of the preferred deployment area to ensure that it is not segmented into discrete parts, and if this happens, it may be necessary to perform subsequent analysis separately for each part.

Next, regional geological information of the preferred deployment region is obtained by the GIS system. The purpose of this step is to provide more detailed geological background information for deployment decisions, ensuring that the deployment location selected meets not only line-of-sight and signal propagation requirements, but also geological conditions for equipment installation and long-term stable operation. First, the type of geologic information that is needed needs to be determined, which may include the type of rock, the formation structure, the groundwater level, the risk of geologic hazards (e.g., landslide, subsidence), etc. These data then need to be obtained from an associated geological database or geological survey. The data may exist in different formats, such as vector layers, raster images, or attribute tables. After the data are imported into the GIS system, the coordinate system is unified and the data format is converted, so that the data can be subjected to spatial superposition analysis with the data of the preferable deployment area. And performing space intersection operation on the geological data and the optimal deployment area by using the space query and attribute extraction functions of the GIS. This process may involve resampling of raster data or clipping of vector data. For continuous geologic features (such as terrain gradients), interpolation operations may be required to obtain higher resolution data. For discrete geologic features (e.g., geotechnical types), then spatial linking operations are required to correlate attribute information to different portions of the preferred deployment area. In addition, some statistical analysis may be required, such as calculating the area occupancy for each geologic type, identifying high-lying regions of geologic risk, and so forth.

Next, a second deployment prohibited area that does not meet the preset detection requirements is generated in the preferred deployment area in combination with the device deployment geological requirements in the area geological information and the preset detection requirements. The purpose of this step is to further refine the regions available for deployment, excluding those regions that, while meeting line-of-sight and signal propagation requirements, are unsuitable due to geological conditions. First, there is a need to explicitly define the geological requirements of the deployment of the device, which may include a number of indicators of ground bearing capacity, terrain slope, groundwater level depth, geological stability, etc. For each index, a threshold or acceptable range needs to be set. For example, it may be desirable that the ground slope not exceed 15 degrees, the groundwater level depth be at least 5 meters, etc. These requirements then need to be translated into spatial query conditions operable in GIS. For continuous variables (e.g., grade), non-conforming areas may be generated directly using grid computing or surface analysis tools. For classification variables (such as geotechnical types), it is necessary to first determine which categories are unsatisfactory and then extract the areas covered by these categories. For indicators of different importance, a weighting system may be introduced, using weighted overlap analysis to comprehensively evaluate the suitability of each location. The generated unconditional area is the second deployment forbidden area, and comprehensively considers the influence of geological conditions on equipment installation and operation. The method not only can improve the success rate of deployment and the long-term stability of equipment, but also can reduce potential geological risks, and provides a more reliable space range for the subsequent selection of specific deployment positions. All second forbidden deployment areas in the preferred deployment area are then screened out and all reserved preferred deployment areas are taken as target deployment areas.

In one embodiment, the unmanned aerial vehicle detection task of the electromagnetic detection unmanned aerial vehicle and the equipment detection task of the electromagnetic signal acquisition and analysis equipment are configured by combining the device deployment point and the area range of the target area in the area map information, and the method comprises the following steps:

Removing the overlapping part of the area range and the equipment sensing range, and taking the remaining area range as the detection range of the electromagnetic detection unmanned aerial vehicle;

And configuring the equipment detection task of the electromagnetic signal acquisition and analysis equipment by combining the equipment sensing range and the unmanned aerial vehicle detection time in the unmanned aerial vehicle detection task.

In this embodiment, referring to fig. 6, it is first necessary to determine the precise geographical coordinates of the deployment point of each device, and then calculate the theoretical maximum perceived distance of the device according to preset parameters of the electromagnetic signal acquisition and analysis device, such as signal receiving sensitivity, antenna gain, operating frequency, and the like. For example, maximum perceived distance =Where Pt is the transmit power, gt and Gr are the transmit and receive antenna gains, respectively, λ is the signal wavelength, and pr_min is the receive sensitivity. Next, using a buffer analysis tool of the GIS software, creating a circular buffer with the calculated maximum perceived distance as a radius centered on each device deployment point, representing the overall perceived range of the electromagnetic signal acquisition and analysis apparatus.

The region range of the target region is identified based on the region boundary of the target region in the region map information, and specifically, the region map information needs to be loaded and parsed first, which is typically stored in a vector data format (such as shape or GeoJSON). The boundary of the target area may be a complex polygon and may even contain a plurality of discontinuous portions or internal voids. The vertex coordinate sequence of the boundary polygon can be extracted by using a geometric processing tool of GIS software. If the target area contains a plurality of discrete portions, it may be necessary to perform these calculations for each portion separately and to mark their relative positional relationship. Finally, a scoped object or data structure is generated containing all of this information, including not only the boundary coordinates, but also the derivative information of area, perimeter, bounding box, etc. The implementation effect of this step is to obtain a well-defined and quantitatively described target area range, providing a basic reference frame for subsequent spatial analysis and mission planning.

And eliminating the overlapping part of the area range and the equipment sensing range, and taking the remaining area range as the detection range of the electromagnetic detection unmanned aerial vehicle. The purpose of this step is to determine the area that needs important attention of the electromagnetic detection unmanned aerial vehicle, avoid the repetition with the perception scope of ground equipment. First, it is necessary to convert the target area range and the device perception range into polygonal objects in the same coordinate system. Then, a spatial analysis tool using GIS software performs a polygon difference set operation, which may be expressed as a detection range=target area range-device perception range. The detection range provides a clear space boundary for unmanned aerial vehicle mission planning, and is beneficial to formulating a more accurate and efficient flight path. And then, respectively configuring unmanned aerial vehicle detection tasks of the electromagnetic detection unmanned aerial vehicle in each mobile unmanned aerial vehicle base station based on the detection range and the device deployment point.

Next, the timing of the unmanned detection tasks, including the predicted detection start and end times for each zone, needs to be analyzed. Such information may be calculated from the flight path and speed of the drone. These time information are then combined with the perceived range of the electromagnetic signal acquisition and analysis device to create a schedule for each surface device. This schedule should include 1) a start time of the probe, typically set slightly before the start time of the probe of the drone in the corresponding area, 2) an end time of the probe, typically set slightly after the end time of the probe of the drone, 3) a frequency and duration of the probe, which may need to be dynamically adjusted depending on the position of the drone. For example, when the drone is near the edge of the device perception range, it may be desirable to increase the detection frequency to increase the accuracy of the data. Furthermore, the duty cycle and energy consumption of the device need to be considered. Optimization algorithms (such as genetic algorithms or simulated annealing) can be used to balance the detection effect and energy efficiency. For each time period, specific detection parameters such as frequency range, signal strength threshold, sampling rate, etc. need to be configured. These parameters may need to be dynamically adjusted depending on the environmental conditions and the detection targets. It is also necessary to set data storage and transmission policies including local storage capacity, data compression method, transmission frequency, etc. The effect of this step is to obtain an accurate, dynamic and coordinated configuration of the ground equipment detection tasks. This configuration not only ensures time synchronization of the ground equipment and the unmanned aerial vehicle detection activity, but also optimizes resource utilization, improving efficiency and reliability of the overall detection system. By the coordination configuration, seamless matching of ground equipment and an unmanned aerial vehicle system can be realized, and the detection coverage range and the data quality are maximized.

In one embodiment, the unmanned aerial vehicle detection task of the electromagnetic detection unmanned aerial vehicle is configured in each mobile unmanned aerial vehicle base station based on the detection range and the device deployment point, and the unmanned aerial vehicle detection task comprises the following steps:

Determining an initial range division point in a detection range based on the device deployment point, wherein the linear distance between the initial range division point and all the device deployment points is equal;

Generating a segmentation boundary in the detection range with a shortest straight line between an initial range segmentation point and a device deployment point;

The base station detection range is respectively allocated to each mobile unmanned aerial vehicle base station by combining the segmentation boundary and the range boundary of the detection range;

For each mobile unmanned aerial vehicle base station, generating an unmanned aerial vehicle optimal path of the electromagnetic detection unmanned aerial vehicle by adopting an optimal path algorithm based on the detection range of the base station and taking another mobile unmanned aerial vehicle base station as a path end point, wherein all mobile unmanned aerial vehicle base stations can only be used as the path end point of the optimal path of one unmanned aerial vehicle at most in the same unmanned aerial vehicle detection task;

calculating the flight time of the electromagnetic detection unmanned aerial vehicle in the optimal path of the unmanned aerial vehicle based on unmanned aerial vehicle flight parameters preset by a mobile unmanned aerial vehicle base station;

If the flight time is the maximum flight time in the flight times calculated by all the mobile unmanned aerial vehicle base stations, an unmanned aerial vehicle detection task of the electromagnetic detection unmanned aerial vehicle is configured by combining the optimal path of the unmanned aerial vehicle and the flight parameters of the unmanned aerial vehicle;

And if the flight time is not the maximum flight time in the flight times calculated by all the mobile unmanned aerial vehicle base stations, adjusting the flight parameters of the unmanned aerial vehicle until the flight time is the same as the maximum flight time, and configuring an unmanned aerial vehicle detection task of the electromagnetic detection unmanned aerial vehicle by combining the optimal path of the unmanned aerial vehicle and the adjusted flight parameters of the unmanned aerial vehicle.

In the present embodiment, as shown in fig. 6, the initial range division point is determined in the detection range based on the device deployment point, and the straight line distances between the initial range division point and all the device deployment points are equal. The purpose of this step is to find a balance point (point P in fig. 6) so that the probe range can be evenly distributed to the individual mobile drone base stations. One way to achieve this is to use a weighted centroid algorithm. First, each device deployment point is considered a particle, whose quality can be assigned according to the importance of the point or the expected workload. Then, the coordinates of the initial range division points are calculated using the following formula:

x=Σ(wi*xi)/Σwi

y=Σ(wi*yi)/Σwi

where (xi, yi) is the coordinates of the i-th device deployment point and wi is its weight.

However, this initial point may not fully satisfy the requirement of being equidistant from all deployment points. Therefore, iterative optimization is required. A gradient descent method may be used to define the objective function as the variance of the separation point to all deployment point distances, and then minimize this variance through multiple iterations. The position of the division point is slightly adjusted for each iteration until the preset precision requirement or the upper limit of the iteration times is reached. In practical applications, the shape and topographical features of the detection range need to be considered. If the detection range is non-convex or contains inaccessible areas, it may be necessary to use more complex algorithms, such as Voronoi diagram-based methods to determine the segmentation points.

The segmentation boundary is generated in the detection range with a shortest straight line between the initial range segmentation point and the device deployment point. This process is actually building a Voronoi diagram with each device deployment point as the center of a Voronoi cell. The generation of the segmentation boundary involves the following steps:

1. For each pair of adjacent device deployment points, their perpendicular bisectors are calculated. This bisector may be determined by the following method:

a) Calculate the midpoint of the two points (xm, ym) = ((x1+x2)/2, (y1+y2)/2)

B) Calculating a direction vector (- Δy, Δx) perpendicular to the connecting line, where Δx=x2-x 1, Δy=y2-y 1

2. These bisectors are extended until they intersect the boundary of the detection range.

3. At each intersection point, it is necessary to determine whether the point really belongs to the final Voronoi boundary. This may be determined by examining the distance of the point to all device deployment points. Only if the distance from this point to the nearest two device deployment points is equal and less than the distance to all other points is it a valid boundary point.

Next, a base station detection range is allocated to each mobile unmanned aerial vehicle base station in combination with the division boundary and the range boundary of the detection range, respectively. The purpose of this step is to determine the specific area for which each mobile drone base station is responsible, ensuring that the entire probe range is covered completely and efficiently. Implementing this step involves several key operations:

1. boundary intersection calculation, namely performing space intersection operation on the Voronoi segmentation boundary generated in the previous step and the outer boundary of the detection range. This may be accomplished using the geometric function of GIS software, such as the "Intersection" tool of QGIS or the "Interect" tool of ArcGIS.

2. The sub-region is generated in such a way that the result of the intersection operation is a series of polygons, each representing a preliminary detection range of one base station.

3. Attribute assignment-each sub-region is assigned a unique identifier and associated attributes such as responsible base station ID, area, perimeter, etc.

4. Connectivity check-ensuring that the probe range of each base station is continuous. If a discontinuity occurs, the boundary needs to be reassigned or adjusted.

5. Load balancing-calculating the area allocated to each base station-if the difference is too large-it may be necessary to adjust the segmentation boundary to achieve a more balanced allocation. This can be achieved by moving the generation point of the Voronoi diagram or adjusting the weights.

6. Obstacle handling-considering obstacles (e.g., buildings, bodies of water, etc.) within a detection range, it may be desirable to exclude these areas from the detection range or assign them a special detection strategy.

For each mobile unmanned aerial vehicle base station, generating an optimal unmanned aerial vehicle path of the electromagnetic detection unmanned aerial vehicle by adopting an optimal path algorithm based on the detection range of the base station and taking another mobile unmanned aerial vehicle base station as a path end point, wherein the step involves optimization consideration of various aspects including path length, coverage efficiency, energy consumption and the like. This step may be performed by the following method:

1. the path planning algorithm is selected by using an A-algorithm, a fast random search tree (RRT) or an ant colony Algorithm (ACO) and the like. Taking the example of the a algorithm, it combines the heuristic of the best-first search with the completeness of Dijkstra algorithm.

2. Map discretization, namely converting the detection range of the base station into a grid map, wherein each grid represents one possible unmanned plane position. The choice of grid size requires balancing computational complexity with path accuracy.

3. Cost function definition, namely designing a cost function comprehensively considering a plurality of factors, such as:

f(n)=g(n)+h(n)+c(n)

Where g (n) is the actual cost from the start point to the current node, h (n) is the estimated cost from the current node to the target, and c (n) is an additional constraint (e.g., obstacle avoidance, coverage requirement, etc.).

4. Endpoint selection, selecting an appropriate endpoint base station for each base station. This may be based on distance, direction, or a predetermined rotation strategy. It is necessary to ensure that each base station only serves as an end point of one path at most.

5. Path generation for each base station, a path to its destination base station is generated using a selected algorithm.

6. Path optimization, namely optimizing the initially generated path, such as smoothing, redundant point deletion and the like, so as to improve the flight efficiency.

8. Task allocation verification, which is to ensure that all base stations are allocated tasks and each base station only serves as the endpoint of one path at most.

The method not only ensures the integrity and efficiency of the detection task, but also realizes the balanced use of unmanned aerial vehicle resources by reasonably distributing the terminal base station. The generated path provides detailed guidance for subsequent task execution and resource scheduling, and is beneficial to improving the reliability and flexibility of the whole detection system.

Next, the flight time of the electromagnetic detection unmanned aerial vehicle in the optimal path of the unmanned aerial vehicle is calculated based on unmanned aerial vehicle flight parameters preset by the mobile unmanned aerial vehicle base station. This calculation process requires consideration of a number of factors to ensure accurate estimation of time of flight, providing basis for subsequent task coordination and resource allocation. Specifically, preset unmanned aerial vehicle flight parameters including maximum flight speed (Vmax), cruise speed (Vcruise), acceleration (a) and deceleration (d), turning radius (R) or maximum turning angular speed (ω), climb/descent rate, first need to be understood and interpreted. The optimal path is then broken down into a plurality of straight and curved segments. For each path segment, the change in speed over time is calculated. For the turn section, the time is calculated using the angular velocity and the angular change. If the path contains significant height variation, climbing or descending time needs to be considered, and finally, the time of all path sections is accumulated to obtain the flight time of the electromagnetic detection unmanned aerial vehicle in the optimal path of the unmanned aerial vehicle.

If the calculated flight time in a certain mobile unmanned aerial vehicle base station is the maximum flight time in the flight time calculated by all the mobile unmanned aerial vehicle base stations, the unmanned aerial vehicle base station can be directly combined with an optimal path of an unmanned aerial vehicle and flight parameters of the unmanned aerial vehicle to configure unmanned aerial vehicle detection tasks of the electromagnetic detection unmanned aerial vehicle. The unmanned aerial vehicle detection task not only comprises accurate flight paths and time arrangement, but also comprises a plurality of aspects such as detection parameters, energy management, communication strategies and the like, and high efficiency and reliability of task execution are ensured.

If the calculated flight time in a certain mobile unmanned aerial vehicle base station is not the maximum flight time in the flight times calculated by all the mobile unmanned aerial vehicle base stations, for the mobile unmanned aerial vehicle base station, unmanned aerial vehicle flight parameters need to be adjusted until the flight time is the same as the maximum flight time, and unmanned aerial vehicle detection tasks of the electromagnetic detection unmanned aerial vehicle are configured by combining the optimal path of the unmanned aerial vehicle and the adjusted unmanned aerial vehicle flight parameters. The aim of this step is to ensure that all unmanned aerial vehicles can complete tasks simultaneously, the specific implementation of the adjustment of the unmanned aerial vehicle flight parameters is that firstly, the parameters to be adjusted are determined, generally comprising cruise speed, hover time, sampling frequency, etc., these parameters are adjusted step by step using an optimization algorithm (such as a gradient descent method or a genetic algorithm), and after each adjustment, the flight time is recalculated until it matches the maximum flight time. If the speed adjustment is insufficient to reach the target time, fine tuning of the flight path is considered. A spiral or zig-zag flight can be added to extend the flight time.

After the parameters are adjusted, integrating all the adjusted parameters and plans to generate a detailed task execution script including specific instructions of each waypoint, so as to ensure that task configuration of all unmanned aerial vehicles is synchronous in time. The final task configuration is securely transferred to the respective drone. The steps realize time synchronization of all unmanned aerial vehicle tasks, and through fine adjustment of flight parameters and task details, the coordination and efficiency of the whole detection system are improved, and the flexibility and adaptability of the system are enhanced.

In one embodiment, the analyzing the radiation source positions and the radiation source types of all the radiation sources in the target area by combining the first area detection result and the second area detection result includes the following steps:

analyzing by combining the first region detection result and the second region detection result to obtain an electromagnetic environment situation sensing result in the target region;

Identifying the longitude and latitude data of the radiation sources of all the radiation sources in the target area based on the electromagnetic environment situation sensing result;

In this embodiment, the electromagnetic environment situation sensing result in the target area is obtained by combining the first area detection result and the second area detection result. Specifically, first, data preprocessing is required for the two region detection results, including noise removal, outlier detection, and data normalization. This may be achieved using techniques such as wavelet denoising, median filtering, etc. Data alignment and fusion is then performed, ensuring that the data from the different sources are consistent in time and space. This may involve timestamp correction and spatial interpolation techniques. The fused data is analyzed using advanced signal processing techniques. Including spectral analysis (e.g., fast fourier transform FFT), time-frequency analysis (e.g., short-time fourier transform STFT or wavelet transform), and modulation identification (e.g., cyclic spectrum analysis). These analyses help identify characteristics of the signal such as center frequency, bandwidth, signal strength, and modulation type. For analysis of the spatial distribution, a statistical method such as kriging or Inverse Distance Weighting (IDW) may be used to generate a spatial distribution map of the electromagnetic field strength.

To identify potential sources and disturbances, signal source localization algorithms such as time difference of arrival (TDOA) or angle of arrival (AOA) methods may be applied. In addition, time series analysis is required to identify the time-varying nature and periodicity pattern of the signal. This can be achieved by autoregressive integral moving average (ARIMA) models or long term memory networks (LSTM) and the like. And finally, integrating all the analysis results to form an electromagnetic environment situation sensing result. The method specifically comprises the steps of an overview of electromagnetic spectrum use conditions, positions and characteristics of main radiation sources, identification of potential interference sources, detection of abnormal signals and time-space change trend of an electromagnetic environment.

And identifying the longitude and latitude data of the radiation sources of all the radiation sources in the target area based on the electromagnetic environment situation awareness result. Specifically, firstly, situation awareness results need to be analyzed, and possible radiation source characteristics are extracted. Including identification of signal strength peaks, analysis of spectral features, and consideration of time correlation. For each potential radiation source, a variety of positioning techniques are employed to determine its geographic location. Common methods include triangulation, angle of arrival (AOA), time Difference (TDOA), frequency Difference (FDOA), and the like. These methods may be used in combination to improve positioning accuracy. For example, a weighted least squares method may be used to integrate the various measurements. In practical application, the influence of multipath effect, atmospheric refraction and other factors on the positioning accuracy also needs to be considered. Ray tracing or statistical models may be used to compensate for these effects. For multiple sources in dense areas, it may be desirable to use a high resolution algorithm, such as MUSIC (MultipleSignalClassification) algorithm, to separate and locate overlapping sources. After the preliminary positioning result is obtained, error analysis and precision evaluation are required. This can be achieved by monte carlo simulation or covariance matrix analysis. For results where accuracy is not satisfactory, it may be necessary to add measurement points or adjust measurement strategies. Finally, the positioning result is converted into a standard geographic coordinate system (such as WGS 84) to obtain longitude and latitude data of each radiation source. This transformation process requires consideration of transformation parameters between different coordinate systems and an earth ellipsoid model.

And carrying out cluster analysis on all the radiation sources by utilizing a cluster classification algorithm based on the longitude and latitude data of the radiation sources to obtain the radiation source types of all the radiation sources. The purpose of this step is to group radiation sources with similar characteristics and infer the type of radiation source from the characteristics of these groups. Implementing this step involves the following stages:

1. In addition to latitude and longitude data, other characteristics that may affect classification, such as signal strength, frequency range, modulation type, time pattern, etc., need to be extracted. These features may form a multi-dimensional vector representing each radiation source.

2. And (3) carrying out standardization or normalization processing on the extracted features, so as to ensure that the features with different scales can influence the clustering result fairly. Common methods include Z-score normalization or Min-Max scaling.

3. Depending on the nature of the data and the expected cluster structure, an appropriate clustering algorithm is selected, such as the K-means clustering algorithm.

4. A clustering algorithm is implemented, taking K-means as an example, and the basic steps are as follows:

a) Randomly selecting K center points

B) Assigning each point to the nearest center point

C) Recalculating the center point of each cluster

D) Repeating b) and c) until convergence or maximum number of iterations is reached

5. The optimum cluster number is determined using methods such as elbow rule, contour coefficient or gap statistic.

6. The cluster quality is evaluated using an internal evaluation index (e.g., profile factor, calinski-Harabasz index) and an external evaluation index (if there is label data).

7. The characteristics of each cluster, including geographical distribution, signal characteristics, etc., are analyzed to infer the type of radiation source that each cluster may represent.

8. Based on the clustering result, a classification model (such as a Support Vector Machine (SVM) or random forest) can be trained to automatically classify new radiation sources.

Through the steps, a radiation source classification result can be finally obtained, and each radiation source is assigned a type label. This classification not only takes into account geographical location, but also integrates signal characteristics, thus being able to more accurately reflect the essential properties of the radiation source. Such classification results help to understand the distribution pattern of different types of radiation sources within the target area, identifying potential sources of interference or abnormal signals.

The invention also discloses an electromagnetic environment analysis system combining situation awareness and space modeling, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the electromagnetic environment analysis method combining situation awareness and space modeling in any one of the embodiments when executing the computer program.

The processor may be a Central Processing Unit (CPU), or of course, according to actual use, other general purpose processors, digital Signal Processors (DSP), application Specific Integrated Circuits (ASIC), ready-made programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., and the general purpose processor may be a microprocessor or any conventional processor, etc., which is not limited in this respect.

The memory may be an internal storage unit of the computer device, for example, a hard disk or a memory of the computer device, or an external storage device of the computer device, for example, a plug-in hard disk, a Smart Memory Card (SMC), a secure digital card (SD), or a flash memory card (FC) provided on the computer device, or the like, and may be a combination of the internal storage unit of the computer device and the external storage device, where the memory is used to store a computer program and other programs and data required by the computer device, and the memory may also be used to temporarily store data that has been output or is to be output, which is not limited by the present application.

It will be appreciated by persons skilled in the art that the above discussion of any embodiment is merely exemplary and is not intended to imply that the scope of the application is limited to these examples, that combinations of technical features in the above embodiments or in different embodiments may also be implemented in any order, and that many other variations of the different aspects of one or more embodiments of the application as described above exist within the spirit of the application, which are not provided in detail for the sake of brevity.

One or more embodiments of the present application are intended to embrace all such alternatives, modifications and variations as fall within the broad scope of the present application. Accordingly, any omissions, modifications, equivalents, improvements and others which are within the spirit and principles of the one or more embodiments of the application are intended to be included within the scope of the application.

Claims

1. An electromagnetic environment analysis method combining situational awareness and space modeling, characterized in that it is applied to an electromagnetic environment detection system, wherein the electromagnetic environment detection system includes three mobile detection devices with the same configuration, each of which includes a mobile drone base station and an electromagnetic signal acquisition and analysis device, and an electromagnetic detection drone is deployed in each of the mobile drone base stations, and any two of the mobile detection devices maintain a communication connection, the electromagnetic detection drone is in communication connection with all of the mobile drone base stations, and the electromagnetic detection drone can stay at any of the mobile drone base stations;

The method comprises the following steps:

Obtain regional map information of the target area where electromagnetic environment analysis is required;

Based on the regional map information and through the GIS system, a regional three-dimensional model of the target area is constructed;

Identifying the area range of the target area according to the area boundary of the target area in the area map information;

Taking the center of the largest inscribed circle in the target area as the range expansion base point, and generating a virtual area after the target area is proportionally enlarged according to a preset enlargement ratio;

taking an area between a virtual area boundary of a virtual area and the area boundary in the area map information as a preselected deployment area;

Using the regional three-dimensional model, all areas in the pre-selected deployment area that do not meet the preset detection requirements are screened out, and all the retained pre-selected deployment areas are used as target deployment areas;

Randomly selecting three position coordinate points from the initial deployment point set as initial deployment points based on a preset distance constraint condition, wherein the distance constraint condition is that the straight-line distance between any one of the initial deployment points and the other two of the initial deployment points is equal;

The deployment point coverage area in the target area is calculated according to the deployment point coordinates of all the initial deployment points, the deployment point coverage area is the overlapping area between the complete coverage area formed by all the initial deployment points and the target area, and the complete coverage area is the area enclosed by the shortest straight line segment between any two of the initial deployment points;

Taking maximizing the coverage area of the deployment points as the optimization goal, all the initial deployment points are adjusted to the optimal deployment points based on the distance constraint condition and using an optimization algorithm;

The optimal deployment point is used as the device deployment point of the mobile detection device, all the device deployment points are located at the edge of the area boundary of the target area and outside the target area, and the straight-line distance between any one of the device deployment points and the other two device deployment points is equal;

Deploy all the mobile detection devices according to the device deployment points;

After all the mobile detection devices are deployed to the device deployment point, the drone detection task of the electromagnetic detection drone and the equipment detection task of the electromagnetic signal acquisition and analysis device are configured in combination with the device deployment point and the area range of the target area in the area map information;

Using the mobile drone base station to send the drone detection task to the electromagnetic detection drone currently deployed by the mobile drone base station;

Control all the electromagnetic detection drones to perform the drone detection task in silent communication mode, so that the electromagnetic detection drones detect the electromagnetic environment situation in the target area, and control all the electromagnetic signal acquisition and analysis equipment to perform the equipment detection task, so that the electromagnetic signal acquisition and analysis equipment detects the electromagnetic environment situation in the target area;

The mobile detection device obtains the first area detection results of all the electromagnetic detection drones when they complete the drone detection task, and simultaneously obtains the second area detection results of all the electromagnetic signal acquisition and analysis devices when they complete the device detection task;

The radiation source positions and radiation source types of all radiation sources in the target area are obtained by combining the first area detection result and the second area detection result for analysis.

2. The electromagnetic environment analysis method combining situational awareness and spatial modeling according to claim 1 is characterized in that the optimization goal of maximizing the coverage area of the deployment points is to adjust all the initial deployment points to the optimal deployment points based on the distance constraint conditions and using an optimization algorithm, including the following steps:

Constructing an objective function with maximizing the coverage area of the deployment points as the optimization goal;

Constructing a first initial population of an optimization algorithm based on all of the initial deployment points;

Based on the distance constraint condition and taking approximation of the objective function as the optimization direction, the first initial population is iteratively updated using the optimization algorithm to obtain a first optimal population under the distance constraint condition;

If the first optimal population satisfies the objective function, an optimal deployment point is determined according to the first optimal population.

3. The electromagnetic environment analysis method combining situational awareness and space modeling according to claim 2, characterized in that the method further comprises the following steps:

If the first optimal population cannot satisfy the objective function, the distance constraint condition is adjusted so that the straight-line distance difference between any one of the initial deployment points and the other two initial deployment points is less than a preset difference threshold;

Using the first optimal population as a second initial population of the optimization algorithm;

Based on the adjusted distance constraint condition and taking approximation of the objective function as the optimization direction, the second initial population is iteratively updated using the optimization algorithm to obtain a second optimal population under the adjusted distance constraint condition;

The optimal deployment point is determined according to the second optimal population.

4. The electromagnetic environment analysis method combining situational awareness and spatial modeling according to claim 1 is characterized in that the use of the regional three-dimensional model to screen out all areas in the pre-selected deployment area that do not meet the preset detection requirements and taking all the retained pre-selected deployment areas as target deployment areas comprises the following steps:

Acquire the model heights of all boundary models located at the boundary of the target area in the regional three-dimensional model;

Mark the boundary model whose model height exceeds the obstacle height threshold in the preset detection requirement as an ultra-high boundary model, and obtain the model longitude and latitude coordinates of both ends of the ultra-high boundary model in the three-dimensional model of the region;

Generating a first prohibited deployment area that does not meet the preset detection requirements in the preselected deployment area of the regional map information based on the model longitude and latitude coordinates;

Screening out all the first prohibited deployment areas in the pre-selected deployment areas, and taking all the retained pre-selected deployment areas as preferred deployment areas;

Based on the geological requirements for device deployment in the preset detection requirements and using the GIS system, a regional screening step is performed on the preferred deployment areas, and all the retained preferred deployment areas are used as target deployment areas.

5. The electromagnetic environment analysis method combining situational awareness and spatial modeling according to claim 4 is characterized in that the step of performing regional screening on the preferred deployment area based on the geological requirements for device deployment in the preset detection requirements and using a GIS system, and taking all the retained preferred deployment areas as target deployment areas comprises the following steps:

Acquiring regional geological information of the preferred deployment area through the GIS system;

Combining the regional geological information and the device deployment geological requirements in the preset detection requirements, generating a second prohibited deployment area in the preferred deployment area that does not meet the preset detection requirements;

All the second prohibited deployment areas in the preferred deployment area are screened out, and all the retained preferred deployment areas are used as target deployment areas.

6. The electromagnetic environment analysis method combining situational awareness and spatial modeling according to claim 1 is characterized in that the configuration of the drone detection task of the electromagnetic detection drone and the equipment detection task of the electromagnetic signal acquisition and analysis device in combination with the device deployment point and the area range of the target area in the area map information comprises the following steps:

Based on the device deployment point and according to the device preset parameters of the electromagnetic signal acquisition and analysis device, generating the device perception range of the electromagnetic signal acquisition and analysis device in the regional map information;

Eliminate the overlapping portion of the area range with the sensing range of the device, and use the remaining area range as the detection range of the electromagnetic detection drone;

Based on the detection range and the device deployment point, respectively configure the drone detection task of the electromagnetic detection drone in each of the mobile drone base stations;

The device detection task of the electromagnetic signal acquisition and analysis device is configured in combination with the device sensing range and the drone detection time in the drone detection task.

7. The electromagnetic environment analysis method combining situational awareness and spatial modeling according to claim 6 is characterized in that the drone detection task of configuring the electromagnetic detection drone in each of the mobile drone base stations based on the detection range and the device deployment point comprises the following steps:

Determine an initial range division point in the detection range based on the device deployment point, wherein the straight-line distances between the initial range division point and all the device deployment points are equal;

generating a segmentation boundary in the detection range by the shortest straight line between the initial range segmentation point and the device deployment point;

Assigning a base station detection range to each of the mobile drone base stations in combination with the segmentation boundary and the range boundary of the detection range;

For each of the mobile drone base stations, based on the detection range of the base station and with another mobile drone base station as the path endpoint, an optimal path algorithm is used to generate the drone optimal path of the electromagnetic detection drone, and all the mobile drone base stations can only serve as the path endpoint of one drone optimal path at most in the same drone detection mission;

Calculating the flight time of the electromagnetic detection drone in the optimal path of the drone based on the drone flight parameters preset by the mobile drone base station;

If the flight time is the maximum flight time among the flight times calculated by all the mobile drone base stations, configuring the drone detection task of the electromagnetic detection drone in combination with the drone optimal path and the drone flight parameters;

If the flight time is not the maximum flight time among the flight times calculated by all the mobile UAV base stations, the UAV flight parameters are adjusted until the flight time is the same as the maximum flight time, and the UAV detection mission of the electromagnetic detection UAV is configured in combination with the optimal UAV path and the adjusted UAV flight parameters.

8. The electromagnetic environment analysis method combining situational awareness and space modeling according to claim 1, characterized in that the step of combining the first area detection result and the second area detection result to analyze and obtain the radiation source positions and radiation source types of all radiation sources in the target area comprises the following steps:

Combining the first area detection result and the second area detection result to analyze and obtain the electromagnetic environment situation awareness result in the target area;

Identify the radiation source longitude and latitude data of all radiation sources in the target area based on the electromagnetic environment situation awareness result;

Based on the latitude and longitude data of the radiation sources and using a clustering classification algorithm, a clustering analysis is performed on all the radiation sources to obtain the radiation source types of all the radiation sources.

9. An electromagnetic environment analysis system combining situational awareness and spatial modeling, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the electromagnetic environment analysis method combining situational awareness and spatial modeling as described in any one of claims 1 to 8 when executing the computer program.