An airport broadband cluster network optimization method and system for maintenance support

Abstract

This invention discloses an airport broadband trunking network optimization method and system for maintenance support, relating to the fields of equipment maintenance and wireless communication technology. The method includes: constructing airport functional area division results, a maintenance support business communication demand index table, and a scenario-based parameter library; obtaining measured airport wireless signal data; constructing a standardized wireless propagation model; constructing scenario-specific wireless propagation models for each area of ​​the airport and integrating them to obtain a model parameter library; constructing a constrained multi-objective optimization model, decision variables, and fitness function; and solving the constrained multi-objective optimization model, decision variables, and fitness function using a GA-PSO-SA hybrid optimization strategy to obtain the optimal base station deployment scheme for the airport. This invention employs a hybrid intelligent optimization strategy to quickly respond to sudden scenarios such as extreme weather, base station failures, and surges in service traffic, achieving second-level network reconfiguration and ensuring uninterrupted maintenance support communication.

Description

Technical Field

[0001] This invention relates to the fields of equipment maintenance and wireless communication technology, and in particular to an airport broadband trunking network optimization method and system for maintenance support. Background Technology

[0002] With the rapid development of the aviation industry, flight operation efficiency and the quality of aviation equipment maintenance and support have become core aspects of airport safety operations and service improvement. Aviation equipment maintenance and support covers the entire process, including pre-flight inspections, emergency repairs, scheduled maintenance, apron scheduling, data acquisition, and AR-assisted maintenance. Its communication systems need to achieve high mobility, high bandwidth, low latency, and high reliability in information transmission, directly determining maintenance efficiency and flight turnaround speed. Currently, airport maintenance and support scenarios are characterized by diversified services, intelligent terminals, and mobile operations, making traditional communication methods insufficient to meet the demands of modern maintenance and support.

[0003] Currently, the communication system for aviation equipment maintenance and support mainly relies on wired local area networks combined with conventional VHF radios, which has significant shortcomings: wired networks are fixed in deployment and have poor mobility, making them unsuitable for outdoor and indoor mobile operation scenarios such as aprons, hangars, and maintenance workshops; conventional VHF radios have narrow bandwidth and low speed, only supporting basic voice dispatching and unable to meet the needs of new business requirements such as high-definition video transmission, AR remote maintenance, real-time acquisition of equipment status data, and high-speed download of maintenance data; the system has weak compatibility and poor anti-interference capabilities, and is prone to signal jamming and interruption in the complex electromagnetic environment of airports, making it difficult to ensure the continuity and safety of maintenance operations.

[0004] Broadband trunking network standards offer advantages such as high-speed data transmission, professional trunking scheduling, secure encrypted communication, and strong anti-interference capabilities. Their frequency band planning is compatible with equipment such as airport navigation radar and surface surveillance radar, making them an ideal dedicated network communication technology for airport aviation equipment maintenance and support. Currently, airport broadband trunking network planning largely follows general commercial network design methods, without customized optimization for the specific scenarios of airport maintenance and support. This leads to three major technical challenges in actual deployment and operation: First, there is a lack of wireless propagation models for ultra-closed metal cavity environments. Airport maintenance and support areas include special scenarios such as underground hangars, all-metal maintenance workshops, and semi-underground reinforced bunkers. These environments are enclosed metal structures, resulting in significant multipath effects and severe signal attenuation. Existing indoor propagation models such as ITU-RP.1238 are only designed for ordinary commercial buildings and have not been corrected for scenarios such as airport metal enclosed cavities and thick-walled bunkers. This leads to large errors in the prediction of path loss and penetration loss, resulting in weak or no coverage in core areas such as underground hangars and maintenance workshops.

[0005] Second, there is a significant trade-off between electromagnetically sensitive areas and coverage performance. Airports contain electromagnetically sensitive areas such as fuel storage areas, fuel refueling areas, navigation equipment areas, and radar stations. Communication base station radiation must strictly meet explosion-proof safety and electromagnetic compatibility requirements to avoid electromagnetic interference with airport navigation, surveillance, and fuel safety equipment. Existing network planning methods only treat electromagnetic constraints as soft penalties, lacking an explicit constraint mechanism based on airport safety regulations for electromagnetic exclusion zones. This makes it difficult to accurately control the electromagnetic radiation intensity in sensitive areas while ensuring full coverage of maintenance areas, resulting in a trade-off between safety and coverage.

[0006] Third, the network's ability to rapidly reconfigure under emergency scenarios is insufficient. Airports are prone to base station failures, sudden changes in electromagnetic environment, and surges in traffic during extreme weather, equipment malfunctions, large-scale flight delays, and emergency repairs, requiring rapid adjustments to the communication network deployment plan. Traditional network planning relies on human experience or a single intelligent algorithm. Genetic algorithms (GA) are prone to premature convergence, particle swarm optimization (PSO) is prone to getting trapped in local optima, and simulated annealing (SA) has a slow convergence speed. A single algorithm cannot balance global optimization accuracy with fast solution speed, making it difficult to meet the rapid reconfiguration and high survivability requirements of airport maintenance and support communications. Summary of the Invention

[0007] This invention provides an airport broadband trunking network optimization method and system for maintenance support, which addresses the problems of inaccurate coverage prediction in enclosed metal cavities, constraints in electromagnetically sensitive areas, and rapid replanning in emergency scenarios in existing airport broadband trunking communication network planning methods.

[0008] On the one hand, embodiments of the present invention provide an airport broadband trunking network optimization method for maintenance support, including: constructing airport functional area division results, maintenance support business communication demand index table and scenario basic parameter library through airport geographic mapping data, aviation equipment maintenance support full business list and airport electromagnetic safety specifications; The measured wireless signal data for each area of ​​the airport were obtained through actual measurements based on the airport functional zone division results. A standardized wireless propagation model for the airport is constructed using the basic parameters of the broadband trunking network equipment and the results of the airport's functional area division. Based on the standardized wireless propagation model and the measured wireless signal data, a scenario-specific wireless propagation model for each area of ​​the airport is constructed, and all scenario-specific wireless propagation models are integrated to obtain a model parameter library. Based on the airport's base station deployment restrictions, the maintenance support business communication demand index table, the scenario basic parameter library, and the model parameter library, a constrained multi-objective optimization model, decision variables, and fitness function are constructed. The optimal base station deployment scheme for the airport is obtained by solving the constrained multi-objective optimization model, decision variables, and fitness function using a hybrid optimization strategy of GA-PSO-SA, which includes genetic algorithm, particle swarm optimization, and simulated annealing algorithm.

[0009] In one possible implementation, after obtaining the optimal base station deployment scheme for the airport by solving the constrained multi-objective optimization model, decision variables, and fitness function using a GA-PSO-SA hybrid optimization strategy that includes genetic algorithms, particle swarm optimization, and simulated annealing, the following steps are also included: The optimal base station deployment scheme was verified using an airport simulation scenario and simulation hardware environment to obtain a network performance report and the final deployment scheme.

[0010] In one possible implementation, the airport geographic mapping data includes the location and dimensions of runways, hangars, aprons, fuel storage areas, and command centers; the full business list of aviation equipment maintenance and support includes lists of emergency command, AR maintenance, video surveillance, and data acquisition; and the airport electromagnetic safety specifications include requirements for fuel storage areas, explosion-proof requirements for refueling areas, and electromagnetic radiation limits.

[0011] In one possible implementation, a standardized wireless propagation model for the airport is constructed using the basic parameters of the broadband trunking network equipment and the airport's functional area division results. The construction of scenario-specific wireless propagation models for each area of ​​the airport based on the standardized wireless propagation model and the measured wireless signal data includes: By constructing a standardized wireless propagation model using a logarithmic distance path loss model and an altitude correction factor, the formula for the logarithmic distance path loss model is as follows: ; in, It is the distance from the base station Path loss at a distance of meters; It is a region-related basic loss; It is the operating frequency; It is the propagation distance in kilometers; It is the path loss index; It is penetration loss; It is a zero-mean normal shaded fading random variable; It is the standard deviation of shadow fading based on the scene value; High correction factor The calculation formula is as follows: ; Among them, terminal height The typical value is 1.5m; The final path loss formula is as follows: ; Using measured wireless signal data from open runways, hangar maintenance areas, command centers, fuel storage areas, fuel refueling areas, parking areas, and special equipment areas, along with the final path loss formula, scenario-specific wireless propagation models for each area of ​​the airport are obtained.

[0012] In one possible implementation, the decision variables include binary variables of the base station deployment status, transmit power, and antenna parameters such as altitude, antenna downtilt angle, and antenna azimuth angle.

[0013] In one possible implementation, the multi-objective optimization model is constructed using four core optimization objectives, three hard constraints, three soft constraints and a penalty mechanism, and a comprehensive fitness function. The four core optimization objectives include maximizing coverage reliability, minimizing electromagnetic exposure in oil-sensitive areas, maximizing system capacity, and minimizing construction costs. The three hard constraints include base station coordinate and altitude range constraints, base station no-deployment zone constraints, and base station quantity constraints. The three soft constraints and penalty mechanisms include penalties for insufficient coverage, penalties for minimum spacing between base stations on the same frequency, and penalties for failing to meet service capacity standards.

[0014] In one possible implementation, the GA-PSO-SA hybrid optimization strategy employs a genetic algorithm (GA) for global search, a particle swarm optimization (PSO) algorithm for continuous parameter fine-tuning, and uses the optimal solutions of the GA and PSO algorithms as the initial points for simulated annealing (SA) for local refinement.

[0015] On the other hand, embodiments of the present invention provide an airport broadband trunking network optimization system for maintenance support, the system being used to implement any of the above methods, including: The data acquisition module is used to construct airport functional area division results, maintenance and support business communication requirement index table, and scenario basic parameter library by using airport geographic mapping data, aviation equipment maintenance and support full business list, and airport electromagnetic safety specifications; and to obtain measured wireless signal data of each area of ​​the airport through the airport functional area division results. The model parameter library construction module is used to construct a standardized wireless propagation model for the airport based on the basic parameters of the broadband trunking network equipment and the airport functional area division results; construct scene-specific wireless propagation models for each area of ​​the airport based on the standardized wireless propagation model and the measured wireless signal data; and integrate all the scene-specific wireless propagation models to obtain the model parameter library. The model optimization module is used to construct a constrained multi-objective optimization model, decision variables, and fitness function based on the base station deployment restrictions of the airport, the communication demand index table for maintenance support services, the scenario basic parameter library, and the model parameter library. The solution-solving module is used to solve the constrained multi-objective optimization model, decision variables, and fitness function using a GA-PSO-SA hybrid optimization strategy that includes genetic algorithm, particle swarm optimization, and simulated annealing algorithm to obtain the optimal base station deployment scheme for the airport.

[0016] The airport broadband trunking network optimization method and system for maintenance support described in this invention have the following advantages: (1) Based on the measured data of the airport, the path loss, penetration loss and shadow fading parameters of the enclosed cavity such as underground hangar and all-metal maintenance workshop are corrected, and an airport-specific multi-scenario wireless propagation model is constructed to replace the general indoor model, which greatly reduces the coverage prediction error.

[0017] (2) Balance electromagnetic safety and coverage requirements, and accurately meet the explosion protection and compatibility requirements of sensitive areas of airports.

[0018] (3) Adopting a hybrid intelligent optimization strategy of GA-PSO-SA, GA completes global base station site selection, PSO realizes continuous parameter fine-tuning, and SA jumps out of local optima, quickly responds to sudden scenarios such as extreme weather, base station failure, and surge in service traffic, realizes network reconfiguration in seconds, and ensures uninterrupted communication during maintenance. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating an airport broadband trunking network optimization method for maintenance support, provided as an embodiment of this application; Figure 2 A schematic diagram of the structure of an airport broadband trunking network optimization system for maintenance support is provided in an embodiment of this application; Figure 3 An airport scenario layout and base station deployment diagram for an airport broadband trunking network optimization method for maintenance support provided in this application embodiment; Figure 4 RSRP coverage heatmap comparison diagram for an airport broadband trunking network optimization method for maintenance support provided in this application embodiment; Figure 5 A comparison of SINR distribution heatmaps for an airport broadband trunking network optimization method for maintenance support, provided in an embodiment of this application. Figure 6 The convergence curve of the optimization algorithm for an airport broadband trunking network optimization method for maintenance support provided in this application embodiment; Figure 7 A capacity analysis result diagram of an airport broadband trunking network optimization method for maintenance support provided in an embodiment of this application; Figure 8 An interference analysis result diagram of an airport broadband trunking network optimization method for maintenance support provided in an embodiment of this application; Figure 9 A radar chart illustrating the overall performance of an airport broadband trunking network optimization method for maintenance support, as provided in this application embodiment. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Figure 1 This is a flowchart illustrating an airport broadband trunking network optimization method for maintenance support, provided by an embodiment of the present invention. The embodiment of the present invention provides an airport broadband trunking network optimization method for maintenance support, comprising: The airport functional zone division results, maintenance and support business communication requirement index table and scenario basic parameter library are constructed by using airport geographic mapping data, aviation equipment maintenance and support full business list and airport electromagnetic safety specifications. The measured wireless signal data for each area of ​​the airport were obtained through actual measurements based on the airport functional zone division results. A standardized wireless propagation model for the airport is constructed using the basic parameters of the broadband trunking network equipment and the results of the airport's functional area division. Based on the standardized wireless propagation model and the measured wireless signal data, a scenario-specific wireless propagation model for each area of ​​the airport is constructed, and all scenario-specific wireless propagation models are integrated to obtain a model parameter library. Based on the airport's base station deployment restrictions, the maintenance support business communication demand index table, the scenario basic parameter library, and the model parameter library, a constrained multi-objective optimization model, decision variables, and fitness function are constructed. The optimal base station deployment scheme for the airport is obtained by solving the constrained multi-objective optimization model, decision variables, and fitness function using a hybrid optimization strategy of GA-PSO-SA, which includes genetic algorithm, particle swarm optimization, and simulated annealing algorithm.

[0023] After obtaining the optimal base station deployment scheme for the airport by solving the constrained multi-objective optimization model, decision variables, and fitness function using a GA-PSO-SA hybrid optimization strategy that includes genetic algorithm, particle swarm optimization, and simulated annealing algorithm, the following steps are also included: The optimal base station deployment scheme was verified using an airport simulation scenario and simulation hardware environment to obtain a network performance report and the final deployment scheme.

[0024] The airport geographic mapping data includes the location and dimensions of runways, hangars, aprons, fuel storage areas, and command centers; the comprehensive aviation equipment maintenance and support business list includes emergency command, AR maintenance, video surveillance, and data acquisition lists; and the airport electromagnetic safety specifications include requirements for fuel storage areas, explosion-proof requirements for refueling areas, and electromagnetic radiation limits.

[0025] A standardized wireless propagation model for the airport is constructed using the basic parameters of the broadband trunking network equipment and the airport's functional area division results. Based on the standardized wireless propagation model and the measured wireless signal data, scenario-specific wireless propagation models for each area of ​​the airport are constructed, including: By constructing a standardized wireless propagation model using a logarithmic distance path loss model and an altitude correction factor, the formula for the logarithmic distance path loss model is as follows: ; in, It is the distance from the base station Path loss at a distance of meters; It is a region-related basic loss; It is the operating frequency; It is the propagation distance in kilometers; It is the path loss index; It is penetration loss; It is a zero-mean normal shaded fading random variable; It is the standard deviation of shadow fading based on the scene value; High correction factor The calculation formula is as follows: ; Among them, terminal height The typical value is 1.5m; The final path loss formula is as follows: ; Using measured wireless signal data from open runways, hangar maintenance areas, command centers, fuel storage areas, fuel refueling areas, parking areas, and special equipment areas, along with the final path loss formula, scenario-specific wireless propagation models for each area of ​​the airport are obtained.

[0026] The decision variables include binary variables representing the base station deployment status, transmit power, and antenna parameters such as altitude, antenna downtilt angle, and antenna azimuth angle.

[0027] The multi-objective optimization model is constructed by four core optimization objectives, three hard constraints, three soft constraints and a penalty mechanism, and a comprehensive fitness function. The four core optimization objectives include maximizing coverage reliability, minimizing electromagnetic exposure in oil-sensitive areas, maximizing system capacity, and minimizing construction costs. The three hard constraints include base station coordinate and altitude range constraints, base station no-deployment zone constraints, and base station quantity constraints. The three soft constraints and penalty mechanisms include penalties for insufficient coverage, penalties for minimum spacing between base stations on the same frequency, and penalties for failing to meet service capacity standards.

[0028] In the GA-PSO-SA hybrid optimization strategy, the genetic algorithm GA is used for global search; the particle swarm optimization algorithm PSO is used for continuous parameter fine-tuning; and the optimal solution of the genetic algorithm GA and the particle swarm optimization algorithm PSO is used as the initial point for simulated annealing SA for local refinement.

[0029] For example, following the maintenance and support operation process, the airport is divided into four core areas, with clearly defined physical parameters, including: Flight area: main runway (2000-3500m long, 45-60m wide), taxiway, landing area; Aircraft maintenance area: open-air apron, underground hangar (15-30m deep, all-metal enclosed), reinforced shelters, and maintenance workshops; Fuel Supply Area: Fuel storage area (explosion-proof electromagnetic sensitive area), aircraft fuel refueling area; Command and Living Area: Flight control tower, airfield command center, personnel living area; The entire maintenance process was statistically analyzed, and five core indicators were identified: speed, latency, reliability, concurrent users, and priority. A standardized requirement table was then generated, as shown in Table 1.

[0030] Table 1 Communication Requirements for Maintenance and Support Services

[0031] Then, the mandatory limits for electromagnetic radiation in the fuel storage / refueling area are clearly defined, and the communication system must be compatible with airport radar and electronic equipment, with no risk of electromagnetic interference. Functional areas are divided according to the entire maintenance and support operation process to achieve a precise match between business needs and physical scenarios, providing the most suitable basic input for subsequent models.

[0032] By employing a general model, scenario-specific corrections, and altitude compensation, accurate calculations of wireless signal propagation across the entire airport area are achieved. The logarithmic distance path loss model is used to uniformly describe the signal attenuation pattern across all regions, as shown in the following formula: ; in, It is the distance from the base station Path loss at a distance of meters; It is a region-related basic loss; It is the operating frequency; It is the propagation distance in kilometers; It is the path loss index; It is penetration loss; It is a zero-mean normal shaded fading random variable; It is the standard deviation of shadow fading based on the scene value; Based on path loss, the actual received signal strength at the target point, i.e., the received signal power RSRP, is calculated using the following formula: ; in, It is a base station t The transmission power; Antenna gain; This refers to feeder loss; The signal-to-interference-plus-noise ratio (SNR) is a core criterion for capacity calculation, measuring the signal's ability to resist interference. The formula is shown below: ; Where SINR(x,y) is at position The signal-to-interference-plus-noise ratio (SIR) is measured in dB (usually calculated by first taking the linear value and then the logarithm). In position The received signal power (linear value, in W or mW) of the main serving base station (i.e., the base station providing the strongest signal). In position Place, No. Interfering base stations (excluding the main serving base station) The received signal power (linear value) of base stations other than those in the base station; To cover all base stations connected to the main serving base station Different other base stations ( The total interference power is obtained by summing the received power of the two sides. It is noise power; The formula for calculating noise power is as follows: ; in, It refers to bandwidth; noise figure. The typical value is 5dB. Subscript I : is an abbreviation for Interference, used to identify the "received power of interfering base stations." Therefore Indicates the first Interfering base stations are located at The received power at that location; Subscript i : Represents the index of the primary serving base station. Summation condition. This means that the main serving base station itself is excluded, and only the summation is performed on other base stations.

[0033] High correction factor The calculation formula is as follows: ; Among them, terminal height The typical value is 1.5m; The final path loss formula is as follows: ; Actual measurements were conducted on the seven functional areas of the airport to determine the parameters of the specific model, resulting in a parameter table, as shown in Table 2: Table 2 Parameter Table of Wireless Propagation Model for Each Functional Area

[0034] By using measured wireless signal data from seven functional areas of the airport—open runway, hangar maintenance area, command center, fuel storage area, fuel refueling area, parking area, and special equipment area—and the final path loss formula, a scenario-specific wireless propagation model for each area of ​​the airport is obtained. Based on actual airport measurements, the path loss and penetration loss of enclosed metal cavities such as underground hangars are corrected to solve the problem that the general model is not applicable to the special scenarios of the airport.

[0035] Taking into account the specific requirements of the scenario, the following four optimization objectives are established: 1. Maximize coverage reliability; Different functional areas have varying degrees of dependence on communication. A weighted coverage ratio is introduced, as shown in the following formula: ; in, The weights of business needs for each functional area are shown in Table 1. This is the coverage threshold, set to -105dBm.

[0036] 2. Minimize electromagnetic exposure; To meet electromagnetic compatibility (EMC) requirements, the radiation intensity in sensitive areas needs to be reduced. Electromagnetic exposure is defined as the weighted sum of the inverse squares of the distances from each point in the target area to the base station, as shown in the following formula: ; in, It is a collection of sensitive areas such as oil fields; Grid points to base station The Euclidean distance; Binary decision variables; It is a base station i The transmission power.

[0037] 3. Maximize system capacity; Capacity is represented by the product of the average spectral efficiency and bandwidth in each region, and compared with the service demand rate. The calculation formula is as follows: ; in, For the number of functional areas, For the region average capacity, This represents the rate of business demand in the region. The calculation formula is as follows: ; in, To maximize the number of base stations that can be deployed, For binary decision variables, Indicates base station Deployed This indicates that it has not been deployed. The mathematical expectation symbol represents the expectation of the functional area. All grid points Take the average value. In functional areas Within the signal-to-interference-plus-noise ratio (linear value) received by the user, the channel quality is reflected. The spectral efficiency (unit: bit / s / Hz) given by Shannon's formula represents the rate of bits that can be transmitted per hertz of bandwidth.

[0038] 4. Minimize construction costs; Including fixed costs and site-related costs, the calculation formula is as follows:

[0039] in, For fixed costs; For base stations i The site-related costs; to facilitate weighting with other targets, Normalization to Range, normalization factor is

[0040] The comprehensive fitness function transforms the above objectives into a single-objective fitness function through a weighted sum, as shown in the following equation:

[0041] in ， Weight ， ， ， It can be set based on experience or dynamically adjusted according to tactical missions, but must meet the following requirements. .

[0042] The model constraints are as follows: The optimization process needs to satisfy the following hard constraints and soft penalties. Hard constraints are enforced by the constraint function, while soft constraints are fully reflected in the fitness through the penalty term.

[0043] 1. Hard constraints, including; (1) Coordinates and height range; ; (2) Restricted Areas: Base stations must not be located within oil storage areas.

[0044] (3) Base station quantity constraints: ; 2. Soft constraints, including: (1) Coverage constraint; The coverage rate in each area must not be lower than the minimum requirement. Otherwise, a penalty will be imposed: ; in For the region The measured coverage rate This is the penalty coefficient.

[0045] (2) Station spacing constraints; Inter-frequency base station spacing It must be no less than : ; In the formula, This is the penalty coefficient.

[0046] (3) Capacity constraints; Capacity of each region The rate must not be lower than the demand rate. : ; In the formula, This is the penalty coefficient.

[0047] 3. Final fitness, the formula is as follows: ; The hybrid intelligent optimization algorithm is as follows: 1. Algorithm Design Ideas To address the mixed characteristics of decision variables, which include both discrete (base station activation flag) and continuous (coordinates, altitude) variables, a genetic algorithm (GA) is used for global search, and a particle swarm optimization (PSO) algorithm is used for continuous parameter fine-tuning. The better solution of the two algorithms is used as the initial point for simulated annealing (SA) for local refinement, thus forming a GA-PSO-SA hybrid optimization strategy.

[0048] 2. Genetic Algorithm (GA): (1) Encoding: Each individual consists of two parts—a length of 10 ... binary vector (Base station activation indicator) and coordinate matrix (Only when) Time (Valid). Population size Maximum number of iterations .

[0049] (2) Initialization: Add Gaussian perturbation to the initial layout. And call the function to ensure that the coordinates and height are valid.

[0050] (3) Fitness assessment: The fitness of each individual is calculated according to formula (12) and the penalty term.

[0051] (4) Selection: Tournament selection (scale 2), winners enter the mating pool.

[0052] (5) Crossover: based on probability Perform uniform crossover. For two parent individuals, generate random mask vectors. The enable flag and coordinates of child generation 1 are inherited from parent generation 2 at the position with mask 1, and the rest are inherited from parent generation 1; the opposite is true for child generation 2.

[0053] (6) Mutation: based on probability Perform mutation on an individual, randomly select a base station, and add perturbation to its coordinates. With a probability of 0.05, randomly activate an inactive base station (if the upper limit is not reached) or randomly deactivate an active base station (if the lower limit is not lower than the lower limit), and reinitialize the coordinates.

[0054] (7) Elite retention: The current best individual is directly copied into the next generation of the population.

[0055] 3. Particle Swarm Optimization (PSO): (1) Particle encoding: Each particle is also enabled by an activation flag. and coordinate matrix Composition, population size Set the number to 30, which represents the maximum number of iterations. Take 50.

[0056] (2) Initialization: Particle positions are generated by the initial layout plus random perturbation, and velocity initialization can be set to .

[0057] (3) Speed ​​and location update: Updates are performed only on base stations with the enable flag set to 1, as shown in the following formula: ; Inertia weight The learning factor decreases linearly from 0.9 to 0.4. , is a uniformly random number. In the formula, It is the index of the particles (each particle represents a base station deployment scheme). This is the current iteration number; It is the first The particle in the first The velocity vector at the next iteration determines the direction and step size of the particle's next movement; It is the first The particle in the first The position vector at the next iteration is the candidate solution to the problem (including decision variables such as base station activation flag, coordinates, and transmit power). It is the inertial weight, which controls the tendency of particles to maintain their previous velocity. In this paper... The value is linearly decreased from 0.9 to 0.4 to balance global search and local exploitation; This refers to the individual learning factor (cognitive coefficient), which represents the weight by which a particle learns from its own historical best position. This paper takes... ; This is the social learning factor (social coefficient), representing the weight by which a particle learns to the global optimal position of the group. This paper takes... ; These are two in Random numbers that are independently and uniformly distributed within an interval are used to increase the randomness of the search and avoid getting trapped in local optima. It refers to the first The best position in the history of a particle (individual extreme value), that is, the best solution found by the particle so far; It is the first The globally optimal position perceived by each particle. Typically, the entire population shares a single globally optimal position, denoted here as... This represents the global optimum referenced by each particle; and This refers to the updated velocity and position vectors.

[0058] (4) Individual and global optimality: Real-time updates and global optimal .

[0059] 4. Hybrid Optimization (SA); The solution with higher fitness between the GA and PSO optimization results is used as the initial solution for SA.

[0060] (1) Neighborhood generation: For each base station, a perturbation is applied with a probability of 0.3, and the perturbation amplitude varies with temperature. It decreases linearly. The initial solution is: ; in, It is a planar coordinate perturbation. It is a base station height disturbance. It follows a normal distribution.

[0061] (2) Acceptance Criterion: Metropolis Criterion, if the fitness of the new solution is better than that of the current solution, or If so, then accept the new solution.

[0062] (3) Annealing scheduling: initial temperature value Cooling coefficient Iterated 40 times.

[0063] 5. Complexity Analysis Each fitness evaluation requires traversing all [items / items]. Given a grid point, calculate the distance to all base stations and the path loss, with a complexity of [time complexity missing]. Total complexity of GA PSO total complexity The complexity of the hybrid optimization stage is relatively low. Parallel computation of grid points can significantly reduce runtime. In the formula, This represents the total number of grid points, which is the number of location points after discretizing a continuous geographical area. Each grid point represents a potential user location and is used to calculate received signal power and interference. This refers to the maximum number of deployable base stations (a dimension of the decision variable), i.e., the total number of candidate base station locations; It is the maximum number of iterations (generations) of the genetic algorithm (GA). Refers to the population size (number of individuals) in a genetic algorithm. This refers to the maximum number of iterations in the Particle Swarm Optimization (PSO) algorithm. It is the number of particles (population size) in the particle swarm optimization algorithm. This indicates the time complexity of the algorithm (the asymptotic upper bound in the worst case). In the subscripts, "grid" represents a grid point; "GA" represents a Genetic Algorithm; "PSO" represents a Particle Swarm Optimization algorithm; "pop" represents a population; and "particle" represents a particle. Then, simulation verification and result analysis are performed: 1. Simulation scenario and parameter settings: A typical maintenance support scenario is constructed. The airport space is 3000m × 2000m, with a grid resolution of 20m, generating a total of 150 × 100 = 15000 grid points. Each grid point represents a potential user location. The functional areas are divided as follows: Figure 1 As shown. The broadband trunking network system parameters are set based on actual engineering conditions: operating frequency 1800MHz, system bandwidth 20MHz, maximum base station transmit power 43dBm, antenna gain 15dBi, feeder loss 2dB, receiver noise figure 5dB, thermal noise power spectral density -174dBm / Hz, and shadow fading standard deviation 8dB. The coverage threshold is set to... =105dBm, =3dB, corresponding to the minimum signal strength and signal-to-interference-plus-noise ratio required for service demodulation, respectively. To avoid unrealistically high RSRP values ​​during optimization, a physical upper limit of -60 dBm is applied to RSRP in the program; at the same time, for electromagnetically sensitive areas such as oil depots, additional penalties are applied if RSRP exceeds -90 dBm to meet the requirement of GJB 786A for electric field strength ≤10 V / m. The remaining parameters are consistent with Table 2. All simulations were run in the MATLAB R2023a environment, with an Intel Core i7-12700H CPU @ 2.30 GHz and 16GB of memory.

[0064] Figure 3Different colored rectangles represent different functional areas: gray runway, brown hangar, blue command center, red fuel storage area, yellow fuel refueling area, green apron, and purple special equipment area. Triangular symbols represent the distribution of base stations. The initial empirical layout includes 6 base stations, mainly distributed along the runway centerline.

[0065] Figure 3 In the diagram, "Underground Bunker" represents the underground hangar; "Command Center" represents the command center; "Special Equipment Area" represents the special equipment area; "Main Runway" represents the main runway; "Apron" represents the apron; "Oil Storage Area" represents the fuel storage area; and "Aircraft Refueling Zone" represents the fuel refueling area. The triangular symbols represent the distribution locations of the base stations, and the horizontal and vertical coordinates represent the locations of each base station and area within the airport. The initial empirical layout includes 6 base stations, mainly distributed along the runway centerline.

[0066] Comparison of optimization results: 1. Base station layout optimization: In the initial layout, the underground hangar and hazardous chemical storage facilities had significantly weak coverage. After GA, PSO, and hybrid optimization, the final hybrid optimization result was selected, as follows: Figure 1 As indicated by the red triangle, the number of base stations was reduced from the initial 6 to 2 (deployed at (1110, 370, 30) and (1200, 1560, 30) respectively). Effective coverage of the oil storage area was achieved through optimization of surrounding base stations. All base stations met the constraint of a minimum spacing of 300m.

[0067] 2. Improved coverage performance: Figure 4 The image shows a comparison of RSRP distribution heatmaps before and after optimization. The legend uses different shades of color to represent the RSRP distribution in different areas of the airport, that is, the coverage improvement in each area of ​​the airport. For example, after optimization, the average RSRP of the underground hangar increased from -35.9dBm to -65.8dBm, and the average RSRP of the fuel storage area increased from -60.1dBm to -78.7dBm. Areas with RSRP below -90dBm accounted for approximately 7.9%. Figure 5 The SINR distribution is shown in the figure. The legend uses different shades of color to represent the SINR distribution in different areas of the airport. The figure shows that after optimization, the average SINR increased from the initial 0.9dB to 17.2dB, and the weighted overall coverage increased from 36.6% to 98.6%, with the coverage of the hazardous materials storage area increasing from 65.8% to 89.8%. Table 3 details the changes in coverage of each functional area.

[0068] Table 3 Comparison of Coverage Rates of Different Functional Zones (%)

[0069] Note: The oil storage area is not 100% covered due to electromagnetic radiation limits, but it meets the relevant standards (electric field strength ≤10 V / m).

[0070] 3. Algorithm convergence and running time: Figure 6 Convergence curves for GA (Genetic Algorithm), PSO (Particle Swarm Optimization), and hybrid optimization (SA) are presented. The vertical axis represents the objective variable of the optimization model, i.e., fitness, and the horizontal axis represents the number of iterations for the three optimization methods. GA stabilizes at the 60th generation with an optimal fitness of 0.5434. PSO optimization, based on this, converges even faster, stabilizing at 0.9516 by the 10th generation. Hybrid optimization starts with the optimal solution of GA (0.9516), and after local search using SA, achieves a final fitness of 0.9585, an improvement of approximately 8% compared to the single algorithm.

[0071] The average runtime (10 independent runs) for each stage on an Intel Core i7-12700H @ 2.30 GHz, 16GB memory is: GA: 40.1±4.1 seconds, PSO: 37.9±3.2 seconds, SA: 28.5±2.5 seconds, and the total runtime for hybrid optimization is 136.5±7.8 seconds. This runtime meets the requirements for rapid reconfiguration during wartime (typically allowing for reconfiguration within minutes). If GPU parallel acceleration of grid point calculations is used, it can be further reduced to less than 30 seconds.

[0072] 4. Comparison of ablation experiments with benchmarks: To verify the contributions of each component in the hybrid strategy, ablation experiments were conducted. Table 4 shows the optimization results of different algorithm combinations. GA-PSO achieved the highest fitness of 0.9613, significantly outperforming GA-only and PSO-only. Notably, GA+SA showed an improvement over GA-only, indicating that SA possesses some local search capability; however, the fitness of GA-PSO-SA was lower than that of GA-PSO, suggesting that perturbations near the optimum in SA disrupted the obtained optimal structure. Therefore, this application ultimately adopts GA-PSO as the optimization method.

[0073] Table 4 Ablation Experiment Results

[0074] Compared with random search (200 samples, optimal fitness 0.52) and simplified NSGA-II (optimal fitness 0.90), the method in this application improves performance by 80.8% and 4.4%, respectively.

[0075] 5. Capacity and Interference Analysis: Capacity analysis results are as follows Figure 7 As shown, after optimization, the average system capacity increased from 26.0 Mbps to 70.3 Mbps, an increase of 170.4%, and the edge user capacity (capacity corresponding to the SINR of the 10th percentile) increased from 11.1 Mbps to 135.5 Mbps, an increase of 1118.5%, meeting the needs of a large number of people accessing the system simultaneously during missions.

[0076] Interference analysis results are as follows Figure 8 As shown, the optimized average interference level is -42.8 dBm, a decrease of 15.1 dB compared to the initial layout. The proportion of edge interference (the proportion of grid points with interference levels exceeding -90 dBm among grid points where RSRP is below the threshold of +5 dB) is 8.7% after optimization, a significant reduction from the initial 21.4%. This verifies that the optimized scheme effectively controls interference within the system while meeting electromagnetic compatibility requirements.

[0077] 6. Sensitivity analysis: The key parameters were perturbed by ±20%, and the changes in the fitness of the optimization scheme were observed. The results are shown in Table 5. The maximum relative change in fitness under the perturbation of each parameter did not exceed ±6.2%. Among them, the oil storage path loss index n was the most sensitive (fitness decreased by 5.8% when -20%), but the overall performance remained stable, indicating that the proposed method has good robustness.

[0078] Table 5 Sensitivity Analysis Results

[0079] 7. Overall Performance Radar Chart: Figure 9 The radar chart visually compares the performance of the initial and optimized schemes across six key metrics: RSRP coverage, SINR coverage, edge RSRP, edge SINR, average capacity, cost, and coverage uniformity. After optimization, all metrics were significantly improved, and the radar chart area increased by approximately 4.2 times, indicating that the optimization scheme comprehensively improved network performance.

[0080] Simulation results demonstrate that the proposed hybrid optimization method effectively improves the coverage reliability and anti-interference capability of military airport broadband trunking networks while satisfying electromagnetic compatibility constraints. Specifically, the optimized weighted coverage rate reaches 98.6%, the anti-interference margin is improved by 19.7 dB, and the system capacity is increased by 170.4%, with all indicators outperforming the initial empirical layout. Compared with existing planning methods based on a single algorithm, the two-stage GA-PSO-SA hybrid strategy in this application has significant advantages in solution quality: GA provides the foundation for global search, PSO accelerates continuous parameter optimization, and SA further escapes local optima, with the three complementing each other. Furthermore, the introduction of a regional propagation model makes coverage prediction more accurate, and the setting of weight coefficients reflects the differentiated needs of different service areas, closely aligning with actual engineering scenarios.

[0081] Figure 2 This invention provides a schematic diagram of an airport broadband trunking network optimization system for maintenance support, as shown in an embodiment of the invention. The system includes: The data acquisition module is used to construct airport functional area division results, maintenance and support business communication requirement index table, and scenario basic parameter library by using airport geographic mapping data, aviation equipment maintenance and support full business list, and airport electromagnetic safety specifications; and to obtain measured wireless signal data of each area of ​​the airport through the airport functional area division results. The model parameter library construction module is used to construct a standardized wireless propagation model for the airport based on the basic parameters of the broadband trunking network equipment and the airport functional area division results; construct scene-specific wireless propagation models for each area of ​​the airport based on the standardized wireless propagation model and the measured wireless signal data; and integrate all the scene-specific wireless propagation models to obtain the model parameter library. The model optimization module is used to construct a constrained multi-objective optimization model, decision variables, and fitness function based on the base station deployment restrictions of the airport, the communication demand index table for maintenance support services, the scenario basic parameter library, and the model parameter library. The solution-solving module is used to solve the constrained multi-objective optimization model, decision variables, and fitness function using a GA-PSO-SA hybrid optimization strategy that includes genetic algorithm, particle swarm optimization, and simulated annealing algorithm to obtain the optimal base station deployment scheme for the airport.

[0082] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0083] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for optimizing airport broadband trunking networks for maintenance support, characterized in that, include: The airport functional zone division results, maintenance and support business communication requirement index table and scenario basic parameter library are constructed by using airport geographic mapping data, aviation equipment maintenance and support full business list and airport electromagnetic safety specifications. The measured wireless signal data for each area of ​​the airport were obtained through actual measurements based on the airport functional zone division results. A standardized wireless propagation model for the airport is constructed using the basic parameters of the broadband trunking network equipment and the results of the airport's functional area division. Based on the standardized wireless propagation model and the measured wireless signal data, a scenario-specific wireless propagation model for each area of ​​the airport is constructed, and all scenario-specific wireless propagation models are integrated to obtain a model parameter library. Based on the airport's base station deployment restrictions, the maintenance support business communication demand index table, the scenario basic parameter library, and the model parameter library, a constrained multi-objective optimization model, decision variables, and fitness function are constructed. The optimal base station deployment scheme for the airport is obtained by solving the constrained multi-objective optimization model, decision variables, and fitness function using a hybrid optimization strategy of GA-PSO-SA, which includes genetic algorithm, particle swarm optimization, and simulated annealing algorithm.

2. The airport broadband trunking network optimization method for maintenance support according to claim 1, characterized in that, After obtaining the optimal base station deployment scheme for the airport by solving the constrained multi-objective optimization model, decision variables, and fitness function using a GA-PSO-SA hybrid optimization strategy that includes genetic algorithm, particle swarm optimization, and simulated annealing algorithm, the following steps are also included: The optimal base station deployment scheme was verified using an airport simulation scenario and simulation hardware environment to obtain a network performance report and the final deployment scheme.

3. The airport broadband trunking network optimization method for maintenance support according to claim 1, characterized in that, The airport geographic mapping data includes the location and dimensions of runways, hangars, aprons, fuel storage areas, and command centers; the comprehensive aviation equipment maintenance and support business list includes emergency command, AR maintenance, video surveillance, and data acquisition lists; and the airport electromagnetic safety specifications include requirements for fuel storage areas, explosion-proof requirements for refueling areas, and electromagnetic radiation limits.

4. The airport broadband trunking network optimization method for maintenance support according to claim 1, characterized in that, A standardized wireless propagation model for the airport is constructed using the basic parameters of the broadband trunking network equipment and the airport's functional area division results. Based on the standardized wireless propagation model and the measured wireless signal data, scenario-specific wireless propagation models for each area of ​​the airport are constructed, including: By constructing a standardized wireless propagation model using a logarithmic distance path loss model and an altitude correction factor, the formula for the logarithmic distance path loss model is as follows: ； in, It is the distance from the base station Path loss at a distance of meters; It is a region-related basic loss; It is the operating frequency; It is the propagation distance in km; It is the path loss index; It is penetration loss; It is a zero-mean normal shaded fading random variable; It is the standard deviation of shadow fading based on the scene value; High correction factor The calculation formula is as follows: ； Among them, terminal height The typical value is 1.5m; The final path loss formula is as follows: ； Using measured wireless signal data from open runways, hangar maintenance areas, command centers, fuel storage areas, fuel refueling areas, parking areas, and special equipment areas, along with the final path loss formula, scenario-specific wireless propagation models for each area of ​​the airport are obtained.

5. The airport broadband trunking network optimization method for maintenance support according to claim 1, characterized in that, The decision variables include binary variables representing the base station deployment status, transmit power, and antenna parameters such as altitude, antenna downtilt angle, and antenna azimuth angle.

6. The airport broadband trunking network optimization method for maintenance support according to claim 1, characterized in that, The multi-objective optimization model is constructed by four core optimization objectives, three hard constraints, three soft constraints and a penalty mechanism, and a comprehensive fitness function. The four core optimization objectives include maximizing coverage reliability, minimizing electromagnetic exposure in oil-sensitive areas, maximizing system capacity, and minimizing construction costs. The three hard constraints include base station coordinate and altitude range constraints, base station no-deployment zone constraints, and base station quantity constraints. The three soft constraints and penalty mechanisms include penalties for insufficient coverage, penalties for minimum spacing between base stations on the same frequency, and penalties for failing to meet service capacity standards.

7. The airport broadband trunking network optimization method for maintenance support according to claim 1, characterized in that, In the GA-PSO-SA hybrid optimization strategy, the genetic algorithm GA is used for global search; the particle swarm optimization algorithm PSO is used for continuous parameter fine-tuning; and the optimal solution of the genetic algorithm GA and the particle swarm optimization algorithm PSO is used as the initial point for simulated annealing SA for local refinement.

8. An airport broadband trunking network optimization system for maintenance support, the system being used to implement the method according to any one of claims 1-7, characterized in that, include: The data acquisition module is used to construct airport functional area division results, maintenance and support business communication requirement index table, and scenario basic parameter library by using airport geographic mapping data, aviation equipment maintenance and support full business list, and airport electromagnetic safety specifications; and to obtain measured wireless signal data of each area of ​​the airport through the airport functional area division results. The model parameter library construction module is used to construct a standardized wireless propagation model for the airport based on the basic parameters of the broadband trunking network equipment and the airport functional area division results; construct scene-specific wireless propagation models for each area of ​​the airport based on the standardized wireless propagation model and the measured wireless signal data; and integrate all the scene-specific wireless propagation models to obtain the model parameter library. The model optimization module is used to construct a constrained multi-objective optimization model, decision variables, and fitness function based on the base station deployment restrictions of the airport, the communication demand index table for maintenance support services, the scenario basic parameter library, and the model parameter library. The solution-solving module is used to solve the constrained multi-objective optimization model, decision variables, and fitness function using a GA-PSO-SA hybrid optimization strategy that includes genetic algorithm, particle swarm optimization, and simulated annealing algorithm to obtain the optimal base station deployment scheme for the airport.

Images