Rescue emergency multi-base-station management system and method thereof

Abstract

The invention relates to the technical field of emergency communication, and discloses a rescue emergency multi-base-station management system and a method thereof, and the rescue emergency multi-base-station management system comprises an obtaining module which is used for obtaining geographic parameters of a rescue area and function parameters and position parameters of a plurality of base stations; the signal coverage calculation module is used for calculating multi-dimensional signal coverage under the influence of the plurality of base stations on the geographic parameters in the rescue area by using a composite propagation model; the resource allocation module is used for constructing a base station resource allocation optimization model; the task scheduling module is used for constructing a calculation task scheduling strategy based on elastic scaling; the environment monitoring module is used for detecting environment parameter changes in real time when the communication environment in the rescue area changes; through the technical means of combining air-ground cooperation and intelligent environment perception, the problem of communication blind areas in a complex rescue environment is solved, the communication coverage efficiency is effectively improved, and the environmental adaptability of the system under severe conditions is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency communication, and more specifically, it relates to a rescue emergency multi-base station management system and its method. Background Art

[0002] During the rescue process of large-scale natural disasters or emergencies, communication guarantee is a key factor for the success of rescue operations. The rescue environment usually has the characteristics of complexity and variability, such as structural changes caused by building collapses, decreased visibility caused by smoke, diverse terrains, etc. These factors pose severe challenges to communication coverage.

[0003] Existing emergency communication solutions are mainly based on the deployment of a single type of base station, such as ground fixed base stations or simple mobile base stations, and the system performance mainly depends on the pre-planned coverage strategy. In actual rescue operations, due to the dynamic changes in the environment, such static planning methods are difficult to adapt to complex and variable rescue scenarios and are prone to forming communication blind spots; at the same time, the performance of traditional environmental perception systems significantly decreases under harsh conditions such as smoke and dust, and cannot provide accurate environmental information for communication planning; in addition, ground base stations and aerial communication nodes (such as drone base stations) usually belong to different systems and lack a unified management mechanism, which affects the resource scheduling efficiency; and fixed resource allocation strategies are also difficult to meet the dynamically changing communication requirements during the rescue process.

[0004] These problems are manifested as frequent communication blind spots, poor environmental adaptability, low resource utilization efficiency, and slow system response in actual rescue activities, seriously restricting the development of rescue work. Therefore, it is necessary to develop a multi-base station management method that can adapt to complex rescue environments, achieve air-ground collaboration, and intelligent environmental perception to provide reliable emergency communication guarantee. Summary of the Invention

[0005] The present invention provides a rescue emergency multi-base station management system and its method to solve the technical problem of communication guarantee in complex rescue environments in related technologies.

[0006] The present invention provides a rescue emergency multi-base station management system, including: A parameter acquisition module, configured to acquire geographical parameters of the rescue area, functional parameters and location parameters of multiple base stations; A signal coverage calculation module, configured to calculate the multi-dimensional signal coverage of multiple base stations affected by geographical parameters in the rescue area by using a composite propagation model, and the composite propagation model simultaneously considers terrain information and building information; A resource allocation module, configured to construct an optimized model for base station resource allocation with signal connectivity and resource utilization efficiency as constraint conditions according to the multi-dimensional signal coverage, and the optimized model for base station resource allocation adopts an adaptive iterative algorithm to dynamically adjust the allocation weights; The task scheduling module constructs a computing task scheduling strategy based on elastic scaling based on the base station resource allocation optimization model to dynamically balance the system load and real-time response requirements; The environmental monitoring module is used to detect real-time changes in environmental parameters when the communication environment in the rescue area changes, trigger system reconstruction and task reallocation, and feed back the information to the signal coverage calculation module to re-execute signal coverage calculation, resource allocation, and task scheduling.

[0007] In a preferred embodiment, the composite propagation model in the signal coverage calculation module extracts features from historical data through deep transfer learning to adapt to the signal propagation characteristics of different rescue scenarios.

[0008] In a preferred embodiment, the base station resource allocation optimization model in the resource allocation module includes: The network topology construction unit is used to construct a network topology model based on graph theory, with base stations as nodes and signal connectivity as edges; The criticality calculation unit is used to calculate the criticality indicators of each node, including coverage, connectivity, and load capacity; The resource allocation unit calculates the resource allocation scheme using a multi-objective optimization algorithm based on the criticality indicators and resource requirements.

[0009] In a preferred embodiment, the multi-objective optimization algorithm uses an improved particle swarm optimization algorithm, where the update speed and position of the particles are dynamically adjusted according to changes in coverage and connectivity requirements.

[0010] In a preferred embodiment, the computing task scheduling strategy in the task scheduling module is based on a microservices architecture, decomposing the computing tasks into multiple independent microservices to support horizontal scaling and error isolation.

[0011] In a preferred embodiment, when detecting changes in the communication environment, the environmental monitoring module uses a prediction model to predict potential environmental change trends and trigger the resource pre-allocation mechanism in advance.

[0012] In a preferred embodiment, it further includes: The cooperative communication module is used to maintain a base station inter-cooperative communication protocol to support resource sharing and command transmission among heterogeneous network devices.

[0013] In a preferred embodiment, it further includes: The security mechanism module is used to establish a distributed security mechanism to ensure normal communication functions can still be maintained in case of communication interruption or damage to some nodes.

[0014] In a preferred embodiment, the base station function parameters obtained by the parameter acquisition module include: signal transmission power, frequency band, bandwidth, load capacity, battery capacity, and standby time.

[0015] In a preferred embodiment, a rescue emergency multi-base station management method for implementing a rescue emergency multi-base station management system includes the following steps: Obtain the geographical parameters of the rescue area, and obtain the function parameters and location parameters of multiple base stations; Use a composite propagation model to calculate the multi-dimensional signal coverage of multiple base stations under the influence of geographical parameters in the rescue area; According to the multi-dimensional signal coverage, construct a base station resource allocation optimization model with signal connectivity and resource utilization efficiency as constraints; Based on the base station resource allocation optimization model, construct a computing task scheduling strategy based on elastic scaling; When the communication environment in the rescue area changes, detect the change of environmental parameters in real time, and trigger system reconstruction and task reallocation.

[0016] The beneficial effects of the present invention are as follows: Through air-ground cooperation and intelligent reflecting surface technology, the communication coverage efficiency of the system is improved, and the communication connection inside complex buildings is more reliable, especially.

[0017] The multi-modal environment perception ability based on near-infrared structured light technology enables the system to still achieve accurate environment perception under harsh conditions such as insufficient light and smoke, and the communication coverage planning is more accurate, adapting to complex and changeable rescue scenarios.

[0018] By uniformly managing air-ground communication nodes and intelligently scheduling resources, the system can dynamically allocate resources according to actual needs, improving the resource utilization rate of the overall communication network.

[0019] The intelligent reflecting surface technology enables the system to quickly respond to environmental changes without waiting for the deployment of new base stations, greatly shortening the coverage optimization time, and the communication in key areas can be quickly restored.

[0020] Even if some nodes fail, the system can still maintain network connectivity through re-planning and resource scheduling, support the communication needs of high-speed moving nodes, and ensure that rescue personnel always maintain communication connections. Brief Description of the Drawings

[0021] Figure 1 It is a module diagram of a rescue emergency multi-base station management system of the present invention. Detailed Embodiments

[0022] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that discussing these embodiments is only to enable those skilled in the art to better understand and thus implement the subject matter described herein, and changes can be made to the functions and arrangements of the elements discussed without departing from the scope of protection of the content of this specification. Each example can omit, substitute, or add various processes or components as needed. Additionally, the features described in some examples can also be combined in other examples.

[0023] In at least one embodiment of the present invention, a rescue emergency multi-base station management system is disclosed, as Figure 1 shown, including: A parameter acquisition module, configured to acquire geographical parameters, functional parameters, and location parameters of multiple base stations in a rescue area; Specifically, it includes the following steps: Step 1.1, deploy an environment perception unit;

[0024] Deploy multiple environment perception units in the rescue area, and each unit consists of a near-infrared structured light projector and a multispectral camera. The working wavelength range of the near-infrared structured light projector is 850 - 950 nm, and it can maintain a high penetration ability in harsh environments such as smoke and dust.

[0025] The multispectral camera includes a visible light channel (400 - 700 nm) and a near-infrared channel (750 - 1000 nm), and is equipped with an optical filter to filter out environmental interference.

[0026] The environment perception units are distributed according to the size and complexity of the rescue area. 1 to 2 units are deployed per 100 square meters of the area, forming a complete coverage environment perception network.

[0027] Each unit is connected to each other through a low-power wireless communication module, sharing data and working collaboratively.

[0028] Step 1.2, implement structured light penetrability imaging; Use the deployed environment perception units to perform structured light penetrability imaging. The specific process is as follows: The near-infrared structured light projector emits a structured light pattern with a specific code (such as Gray code, phase-shifted sine pattern, etc.); The multispectral camera synchronously captures the reflected structured light pattern; For areas with severe smoke and dust interference, apply an adaptive exposure control algorithm to dynamically adjust the camera parameters to improve the image quality; Through multiple sampling and image fusion techniques, suppress environmental noise and improve the signal-to-noise ratio.

[0029] Step 1.3, apply a phase shift algorithm to calculate the surface depth; Based on the captured structured light pattern, the phase shift algorithm is applied to calculate the depth of the three-dimensional surface of the environment; This algorithm utilizes the phase information of the structured light pattern to extract depth data through mathematical transformation; The specific depth calculation formula is: ; Where, represents the depth value at the coordinate point , that is, the vertical distance from this point to the camera; is the baseline distance (i.e., the distance between the projector and the camera), representing the physical interval between the structured light projector and the camera; is the camera focal length, which determines the imaging characteristics of the camera; is the phase value of the structured light pattern at the coordinate point , obtained through structured light projection and reflection; is the phase value on the reference plane, serving as the reference point for depth calculation. This formula calculates precise depth information through the principle of triangulation and utilizes the phase difference to achieve three-dimensional environment reconstruction.

[0030] By fusing the depth data obtained from multiple environmental perception units at different perspectives, the Iterative Closest Point (ICP) algorithm is applied to eliminate registration errors and generate complete three-dimensional point cloud data of the environment.

[0031] Step 1.4, construct a voxel hash table to achieve real-time update; To support real-time capture of environmental changes and incremental update of the model, the voxel hash table technology is applied to store and manage the three-dimensional environment model. This technology has the following advantages: High spatial efficiency: Only the voxels containing actual objects are stored, saving memory space; Fast query speed: Achieved through the hash function voxel query with complexity; Flexible update: Supports efficient local updates without the need to reconstruct the entire model.

[0032] The hash function of the voxel hash table is defined as: ; Where, is the voxel coordinate, representing the position of the voxel in three-dimensional space; , , respectively represent the hash mapping transformations for the x coordinate, y coordinate, and z coordinate, used to increase the randomness of the hash function and reduce conflicts; is the exclusive OR operation, used to mix the coordinate values to generate a uniformly distributed hash value; is the size of the hash table, which determines the capacity of the hash table; is the hash value of the voxel, which is used to quickly locate voxel data in the hash table; is the modulo operation to ensure that the generated hash value is within the range of the hash table size.

[0033] When detecting environmental changes, only update the voxel data in the changed area to maintain the real-time nature of the model. The system performs an environmental scan every 30 seconds, compares the depth data before and after, marks the environmental change area, and performs a local update.

[0034] The signal coverage calculation module is used to calculate the multi-dimensional signal coverage affected by multiple base stations on the geographical parameters in the rescue area using a composite propagation model, which simultaneously considers terrain information and building information; Specifically, it includes the following steps: Step 2.1, register heterogeneous communication nodes; Unify the registration of various communication nodes in the rescue scene into the management system, including: Ground fixed base stations: including fixed communication facilities such as conventional macro base stations and micro base stations; Vehicle-mounted mobile stations: mobile communication devices installed on rescue vehicles; Drone base stations: aerial communication nodes carried on drone platforms; Balloon base stations: communication nodes suspended in the air through balloon platforms.

[0035] For each communication node, register its basic parameter information, including: location coordinates (fixed or real-time location), communication coverage range, frequency band resources, transmission power, antenna type, energy status, etc. The system assigns a unique identifier (Universally Unique Identifier, UUID) to each node and establishes a unified management database.

[0036] Step 2.2, apply the four-dimensional resource allocation algorithm to plan the node positions; Based on the environmental three-dimensional model, apply the four-dimensional (space three-dimensional + time) resource allocation algorithm to plan the optimal positions or movement trajectories of each node. This algorithm considers the three-dimensional space coordinates and the time dimension, and optimizes the node distribution on the premise of meeting the rescue coverage requirements. The core of the algorithm includes: Construct a three-dimensional coverage metric model: Divide the rescue area into grid cells, and determine the coverage requirement weight of each cell according to factors such as personnel distribution and rescue importance; Configure the four-dimensional resource allocation objective function: ; Among them, represents the four-dimensional resource allocation objective function, represents the set of positions of all nodes, including the three-dimensional space coordinates of each communication node; represents a time series, representing the planned time dimension; represents the coverage benefit function, measuring the coverage effect of communication nodes on the rescue area; represents the energy efficiency function, evaluating the energy consumption efficiency of node location configuration; represents the operation complexity function, quantifying the difficulty of node deployment and maintenance; 、 、 respectively represent the weight coefficients of coverage benefit, energy efficiency, and operation complexity in the objective function.

[0037] Apply a hybrid optimization solution algorithm: Combine the genetic algorithm and local search method to solve the node location configuration corresponding to the maximum value of the objective function.

[0038] For a fixed base station, determine its optimal working parameters; For mobile nodes (such as drone base stations), plan their optimal positions and movement trajectories.

[0039] Step 2.3, use ray tracing and atmospheric attenuation models to evaluate the link quality; To evaluate the link quality change at different altitudes, establish a connectivity evaluation index. The specific methods include: Apply ray tracing technology to simulate the propagation path of electromagnetic waves in three-dimensional space, considering effects such as reflection, refraction, and diffraction; Combine the atmospheric attenuation model to calculate the signal loss: ; Among them, is the total path loss, representing the total energy loss of the signal from the transmitter to the receiver; is the straight-line distance between the transmitter and the receiver; is the altitude of the communication node, which affects the atmospheric density and signal propagation characteristics; is the carrier frequency of the signal. Electromagnetic waves of different frequencies have different propagation characteristics; is the free space loss, representing the basic propagation loss considering only distance and frequency factors under ideal conditions; is the atmospheric absorption loss, representing the energy loss caused by the absorption of the signal by molecules such as water vapor and oxygen in the atmosphere, which is related to altitude and frequency; is the additional loss caused by environmental factors, including signal attenuation caused by terrain obstruction, building reflection, vegetation attenuation, etc.

[0040] Establish a link quality evaluation index system: including indicators such as Signal-to-Noise Ratio (SNR), channel capacity, and link reliability, forming a comprehensive scoring model.

[0041] Through the precise evaluation of link quality, the system can identify potential communication bottlenecks and provide a basis for subsequent network optimization.

[0042] A resource allocation module, which is used to construct an optimization model for base station resource allocation according to multi-dimensional signal coverage, with signal connectivity and resource utilization efficiency as constraint conditions. The resource allocation optimization model adopts an adaptive iterative algorithm to dynamically adjust the allocation weights; Specifically, it includes the following steps: Step 3.1, convert the geometric model into an electromagnetic characteristic model; In order to accurately simulate the propagation characteristics of wireless signals in complex environments, it is necessary to convert the geometric environment model into an electromagnetic characteristic model. The conversion process is as follows: Classification according to material characteristics: Classify the surfaces of different objects in the environment according to material types (such as concrete, metal, wood, glass, etc.); Allocate electromagnetic parameters: Allocate electromagnetic characteristic parameters to each type of material, including the relative permittivity 、magnetic permeability 、conductivity etc.; Construct a grid model: Convert the environment model into a grid model suitable for electromagnetic simulation, and apply adaptive grid division technology to use finer grids in areas where the electromagnetic field changes violently.

[0043] Step 3.2, apply the finite-difference time-domain method to simulate signal propagation; Based on the constructed electromagnetic characteristic model, apply the finite-difference time-domain method (FDTD) to simulate the propagation path and intensity distribution of wireless signals in complex environments. The characteristics of the FDTD algorithm include: Solve Maxwell's equations in the time domain: ; ; Among them, is the electric field intensity vector, representing the magnitude and direction of the electric field intensity at each point in space; is the magnetic field intensity vector, representing the magnitude and direction of the magnetic field intensity at each point in space; is the relative permittivity (absolute value), representing the response characteristics of the medium to the electric field; is the magnetic permeability (absolute value), representing the response characteristics of the medium to the magnetic field; is the current density vector, representing the current intensity and direction per unit area; is the curl operator, used to calculate the rotation characteristics of the vector field; is the partial derivative with respect to time, representing the rate of change of the corresponding physical quantity with time. These two equations are part of Maxwell's equations, describing the propagation laws of electromagnetic fields in time and space.

[0044] Apply the Godunov difference scheme to discretize space and time, and iteratively calculate the electromagnetic field distribution step by step in the time domain; Boundary condition processing: Apply the Perfectly Matched Layer (PML) absorbing boundary condition to avoid errors caused by simulation boundary reflections.

[0045] For wireless signals in different frequency bands, perform FDTD simulations respectively to obtain the signal intensity distribution maps of each frequency band.

[0046] Step 3.3, construct a composite propagation model for deep transfer learning; The FDTD simulation has a high computational complexity and cannot calculate the electromagnetic distribution in large and complex environments in real time. To solve this problem, according to the embodiments of the present application, a composite propagation model based on deep transfer learning is constructed, and its specific implementation method is as follows: Establish a basic model: In a simple environment, use FDTD simulation to generate a training data set, which contains input parameters such as environmental geometric features and material properties and the corresponding electromagnetic field distribution output results. Based on this data set, construct a multi-layer convolutional neural network as the basic propagation model. This network includes: Input layer: Receive voxelized environmental geometric data and material parameters; Feature extraction layer: Consists of multiple convolutional layers, using a 3×3×3 convolutional kernel to extract spatial features; Nonlinear mapping layer: Use the ReLU activation function to introduce nonlinear characteristics; Output layer: Generate the predicted electromagnetic field distribution; Domain adaptation transfer: Through the domain adaptation method, transfer the basic model to the application scenario of a complex rescue environment, and the expression is: ; Among them, is the source domain model (basic model), representing the initial propagation model trained in a simple environment; is the target domain model (model in a complex environment), representing the propagation model adapted to the complex environment of the rescue scenario; is the adaptation function, responsible for transferring the source domain knowledge to the target domain and making necessary adjustments; is a small number of samples in the target domain (measurement data in the actual environment), referring to the limited measured signal data collected at the rescue site, which is used to guide the model to adapt to the characteristics of the actual environment.

[0047] The specific implementation of domain adaptation includes: Feature alignment: Align the feature distributions of the source domain and the target domain through the Maximum Mean Discrepancy (MMD) loss function; Hierarchical fine-tuning: Freeze the parameters of the first few layers of the base model and only fine-tune the parameters of the later layers to adapt to the features of the new environment; Adversarial training: Introduce a domain discriminator to train the model to generate domain-invariant feature representations; Model optimization: When applied at the rescue site, obtain the actual signal strength data through mobile sensors, compare it with the model prediction values, and implement an incremental learning strategy to continuously optimize the model: Error backpropagation: Calculate the mean squared error between the predicted value and the measured value, and update the model parameters through gradient descent; Weight decay: Use L2 regularization to prevent overfitting, and set the weight decay coefficient to 0.001; Learning rate scheduling: Use the cosine annealing learning rate strategy, with an initial learning rate of 0.01; In an actual rescue scenario, this composite propagation model can complete the prediction of the electromagnetic field distribution of the entire rescue area (with an area of 10,000 square meters) in less than 1 second, and control the prediction accuracy error within ±3 dB, far superior to the traditional FDTD method that requires several hours of calculation time, providing a real-time basis for subsequent communication blind spot identification and signal optimization.

[0048] Step 3.4, Generate a signal coverage heat map and identify blind spots; Based on the constructed composite propagation model, combined with the configuration parameters of the actual communication nodes, perform the following operations: Calculate the signal coverage heat map of the entire area: Divide the rescue area into grids and calculate the cumulative signal strength received at each grid point; Identify communication blind spots based on the threshold method: Configure the minimum acceptable signal strength threshold , and mark the areas with signal strength lower than this threshold as communication blind spots; Blind spot priority assessment: Assign a priority score to each communication blind spot according to factors such as the urgency of communication needs, personnel density, and rescue activity frequency , and the formula is: ; Among them, is the priority score of the communication blind spot, indicating the urgency of optimization for this area; is the personnel density, indicating the number of rescue personnel per unit area; is the rescue activity frequency, indicating the intensity of rescue operations in this area; is the minimum acceptable signal strength threshold; is the actual signal strength; represents the degree of insufficient signal strength; , , respectively represent the weight coefficients of personnel density, rescue activity frequency, and signal strength deficiency.

[0049] By accurately identifying and prioritizing blind spots, it provides a clear goal for subsequent signal optimization, ensuring that communication resources are preferentially used to solve the most critical coverage problems.

[0050] The task scheduling module constructs a computing task scheduling strategy based on elastic scaling based on the base station resource allocation optimization model to dynamically balance system load and real-time response requirements; Specifically, it includes the following steps: Step 4.1, determine the optimal deployment location of the intelligent reflecting surface; The intelligent reflecting surface is an artificial structure composed of multiple controllable phase units, which can reflect incident electromagnetic waves in a controlled manner and change the signal propagation path.

[0051] Based on the electromagnetic environment analysis results, apply the geometric constraint optimization algorithm to determine the optimal deployment location of the intelligent reflecting surface: Define the candidate set of deployment locations: According to the geometric characteristics of the rescue environment, select walls, ceilings, or other planar structures where the intelligent reflecting surface can be deployed; Establish a location evaluation model: For each candidate location , calculate its coverage gain score , and the formula is: ; Among them, is the coverage gain score of candidate location , indicating the comprehensive gain effect on all communication blind spots after deploying the intelligent reflecting surface at this location; is the set of communication blind spots, including all areas that require signal enhancement; is the weight of blind spot , and the higher the weight, the more important the blind spot and the need to be solved preferentially; is the signal enhancement amount of deployment location to blind spot , indicating the degree of signal strength improvement of blind spot after deploying the intelligent reflecting surface at this location; represents the weighted sum of all blind spots, comprehensively evaluating the overall effect of candidate locations.

[0052] Consider deployment constraints: including structural stability, installation convenience, occlusion, etc., to form a constraint function ; Solve the optimization problem: ; Among them, is the optimal deployment location, is the candidate deployment location, is the location 's coverage gain score, represents finding the variable value that maximizes the objective function, represents "subject to", is the location 's constraint function value, including constraint conditions such as structural stability, installation convenience, and occlusion situation, is the constraint threshold (the upper limit value of the constraint condition).

[0053] According to the optimization result, select the top N optimal deployment locations and install the intelligent reflecting surface units.

[0054] Step 4.2, calculate the optimal phase configuration of the intelligent reflecting surface; The intelligent reflecting surface actively regulates the signal propagation path by adjusting the phases of each reflecting unit. For each deployed intelligent reflecting surface, calculate its optimal phase configuration: Establish a channel model: Assume the intelligent reflecting surface has reflecting units, then the composite channel from the source node to the target area can be expressed as: ; where, is the overall composite channel response, is the direct channel, that is, the direct propagation path from the source node to the target area, is the channel gain from the th reflecting unit to the target area, is the channel gain from the source node to the th reflecting unit, is the th reflecting unit's phase adjustment value, with a range of 0 to , represents the complex exponential form of phase adjustment, is the total number of reflecting units on the intelligent reflecting surface, is the imaginary unit, represents summing up the contributions of all reflecting units; Define the signal strength objective function: ; where, represents the phase configuration vector; is the sampling point set of the target coverage area (communication blind area), representing multiple measurement points in the area where the signal needs to be enhanced; is at the configuration under the sampling point The channel response at [location] represents the transmission characteristics of the signal from the source node to the sampling point after passing through the IRS. ; Represents the sampling point The signal power at [location]; is the signal strength objective function, representing the cumulative signal power of all sampling points in the target area under a given phase configuration, which is the objective to be maximized during the optimization process.

[0055] Apply the gradient descent algorithm to solve for the optimal phase configuration: ; where is the phase configuration at the -th iteration, representing the vector of phase adjustment values for all reflecting elements of the IRS; is the updated phase configuration after the -th iteration; is the learning rate, which controls the step size of each iteration and affects the convergence speed and stability of the algorithm; is the objective function with respect to the phase configuration . This formula continuously adjusts the phase configuration of the IRS through iteration, ultimately achieving the goal of maximizing the signal strength in the target area.

[0056] Through iterative calculation, the optimal phase configuration of each IRS is obtained to achieve directional signal enhancement in the target area.

[0057] Step 4.3, jointly optimize the parameters of the directional antenna array; To further enhance the signal coverage effect, jointly optimize the IRS and the directional antenna array of the communication node: Adjustment of the parameters of the directional antenna array: For each communication node, according to the characteristics of its antenna array, adjust the beam direction, width, and power distribution to form a collaborative enhancement effect with the IRS; Configure the joint optimization objective function: ; where represents the phase configuration vector; is the antenna array parameter, including parameters such as beam direction, width, and power distribution; is the IRS objective function, used to evaluate the enhancement effect of the IRS on signal strength; is the antenna array objective function, used to evaluate the impact of antenna array parameter adjustment on signal coverage; is the balance coefficient, used to adjust the weight ratio between IRS optimization and antenna array optimization to ensure that the two work together to achieve the best coverage effect; is the joint optimization objective function, whose maximum value corresponds to the best joint configuration scheme; Apply an alternating optimization strategy: first fix optimization , then fix optimization , and repeat alternately until convergence.

[0058] Through the joint optimization of intelligent reflective surfaces and directional antenna arrays, a synergistically enhanced coverage optimization strategy is formed to maximize the signal quality in communication blind spots.

[0059] Step 4.4, conduct real-time performance monitoring and adaptive adjustment; To adapt to the dynamic changes in the rescue environment, the intelligent reflective surface system needs to conduct real-time performance monitoring and adjustment: Deploy signal strength sensors: Deploy small signal strength sensors in key blind spots to monitor signal quality in real time. Establish a feedback adjustment mechanism: Based on monitoring data, calculate the deviation between the actual signal enhancement effect and the expected target, and trigger dynamic adjustment of the phase configuration; Apply an incremental optimization algorithm: only adjust the phase of the reflection surface corresponding to the area with larger deviation, reduce the calculation complexity and improve the response speed.

[0060] Through closed-loop feedback control, the intelligent reflective surface system can adapt to environmental changes in real time and continuously provide optimal signal coverage.

[0061] The environmental monitoring module is used to detect changes in environmental parameters in real time when the communication environment in the rescue area changes, trigger system reconstruction and task reallocation, and feed back information to the signal coverage calculation module to re-execute signal coverage calculation, resource allocation and task scheduling; The specific steps include: Step 5.1, execute the resource borrowing protocol between the open space nodes; In order to realize the cross-layer flexible allocation of spectrum and computing resources, a resource borrowing protocol between air and ground nodes is established: Resource pooling: Incorporate spectrum resources, computing resources, storage resources, etc. of each communication node into a unified resource pool, establish a resource virtualization layer, and realize resource abstraction and unified management; Borrowing request and response mechanism: Establish a priority-based resource borrowing protocol. The specific process includes: The resource demander initiates a borrowing request, specifying the resource type, quantity, and urgency; The resource scheduler selects the appropriate resource provider based on the global resource usage status; The resource provider determines the degree of response based on its own load conditions; Establish a temporary resource channel and complete resource borrowing.

[0062] Resource usage efficiency monitoring: Continuously monitor the usage efficiency of borrowed resources, automatically recycle and reallocate resources with low efficiency to ensure maximum resource utilization.

[0063] Step 5.2, optimize the real-time trajectory of mobile nodes; For mobile communication nodes such as drone base stations, optimize their moving trajectories in real time according to changes in communication requirements and network status: Establish a demand heat map: Based on factors such as communication request frequency, data traffic, and user distribution, construct a real-time communication demand heat map , where represents the three-dimensional space coordinates, corresponding to the longitude, latitude, and altitude positions in the rescue area respectively; represents the time variable, used to reflect the dynamic change of communication demand over time; represents at the coordinate position, time the communication demand intensity value at the moment, the higher the value, the more urgent the communication demand at this position; Calculate the optimal trajectory of the mobile node: Based on the current position and the demand heat map, calculate the optimal trajectory of the mobile node , meeting the following optimization objectives: ; where represents the optimal moving trajectory, represents finding the trajectory that maximizes the objective function , represents the start time, represents the end time, is the position coordinate on the moving trajectory, is the speed of the mobile node at time , is the position the coverage benefit, indicating the quality of communication services that this position can provide, is the speed the energy consumption cost at, indicating the energy consumed by the mobile node flying at this speed, is the energy consumption balance coefficient, used to balance the weight relationship between coverage benefit and energy consumption; represents a small change in time, used for integral calculation; the whole formula represents finding an optimal flight trajectory that can maximize communication coverage benefit and consider energy consumption within the time interval ; Consider flight constraints: Consider the flight constraints of mobile nodes such as drones in trajectory optimization, including maximum speed, acceleration limit, obstacle avoidance requirements, energy limit, etc.; Implement dynamic path adjustment: According to the real-time changes of rescue activities, recalculate the optimal trajectory every certain period (such as 30 seconds) and smoothly transition to the new trajectory to ensure the continuity of communication services.

[0064] Step 5.3, establish a multi-level resource allocation strategy; For communication services with different priorities, establish a multi-level resource allocation strategy: Service classification: Classify rescue communication services into multiple levels such as critical command services, emergency rescue services, and regular communication services; Hierarchical resource reservation: Reserve a certain proportion of resources for high-priority services to ensure the communication quality of critical services even in case of network congestion; Dynamic threshold adjustment: According to the network load situation, adaptively adjust the access thresholds and resource allocation ratios of each level of services, and make the most of network resources while ensuring high-priority services; Resource preemption mechanism: When the demand for high-priority services suddenly increases, allow resources to be preempted from low-priority services and implement smooth migration to reduce service interruption.

[0065] Step 5.4, implement a closed-loop feedback optimization mechanism; Establish a closed-loop feedback mechanism to continuously optimize the resource allocation strategy to cope with environmental and demand changes: Deploy distributed performance monitoring points: Deploy performance monitoring points in the rescue area to collect key indicators such as communication quality, resource utilization rate, and service satisfaction; Build a network status evaluation model: Build a network status evaluation model based on the monitoring data and calculate the communication service quality scores of each region ; Identify performance bottlenecks: Use root cause analysis algorithms to identify bottleneck factors causing service quality degradation, such as insufficient spectrum resources, excessive interference, and insufficient coverage, etc.; Adaptive strategy adjustment: According to the bottleneck analysis results, automatically adjust the resource allocation strategy, communication parameters, or network topology to form a closed-loop optimization, which is expressed by the formula: ; Among them, represents the set of policy parameters at the current time t, including key configuration parameters of the communication system such as resource allocation ratio, frequency allocation, and power control; represents the set of policy parameters at the next time t + 1, that is, the adjusted new parameter values; represents the parameter adjustment amount function, which determines the change amount of how each parameter should be adjusted; represents the set of currently actually measured service quality indicators, including multi-dimensional network performance indicators such as signal strength, delay, packet loss rate, and throughput; It represents a set of preset quality - of - service target values, representing the expected values of various network performance indicators; the entire formula represents a closed - loop control process. The system calculates the adjustment amount required for the current policy parameters by comparing the difference between the actual quality of service and the target quality, thereby obtaining the optimal policy parameters for the next moment.

[0066] Through dynamic resource scheduling and adaptive management, the system can flexibly respond to the changing communication requirements during the rescue process, ensure the communication quality in key areas, and improve the overall network resource utilization rate.

[0067] The rescue emergency multi - base - station management system and method of the present invention can be applied to various disaster rescue scenarios. The following provides a specific application example in earthquake disaster rescue.

[0068] An earthquake of magnitude 6.8 occurred in a certain area, causing extensive damage to buildings, interruption of roads, and paralysis of communication infrastructure. The rescue command center needs to quickly establish a communication network to support the positioning of rescue personnel, information transmission, and coordinated operations. The rescue area is approximately 15 square kilometers, with complex terrain, including urban areas, suburban areas, and mountainous areas.

[0069] System deployment process: Parameter acquisition: The latest geographical data of the rescue area is obtained through satellite maps and drone aerial photography, including: Terrain elevation data: The highest point is 1200 meters above sea level, and the lowest point is 800 meters above sea level; Building distribution: Approximately 40% of the buildings have collapsed or been severely damaged, and 60% of the buildings remain partially structurally intact; Vegetation coverage: Approximately 25% of the area is covered by trees; Six mobile base stations are deployed, and their functional parameters are as follows: Base stations 1 to 3: High - power base stations, with a transmission power of 80W, a coverage radius of 5km, supporting 5G / 4G dual - mode, and a bandwidth of 100Mbps; Base stations 4 to 6: Small portable base stations, with a transmission power of 40W, a coverage radius of 2km, only supporting 4G, and a bandwidth of 50Mbps; All base stations are equipped with 12 - hour backup batteries and solar charging devices; Initial position selection of base stations: Base station 1: Location of the command center (geographical coordinates: 30.12°N, 104.32°E); Base station 2: Medical rescue camp (geographical coordinates: 30.15°N, 104.35°E); Base station 3: Material storage point (geographical coordinates: 30.10°N, 104.30°E); Base Stations 4 to 6: Carried by the search and rescue team, initially located at the edge of the main search and rescue area.

[0070] Signal Coverage Calculation: Application of the Composite Propagation Model: Import the historical signal propagation data of the area as the training set; Adapt to the environmental characteristics after the earthquake through deep transfer learning; Automatically adjust the model parameters, considering the impact of collapsed buildings on signal attenuation; The system calculation results show: Base Stations 1 - 3 form the main coverage network, covering approximately 65% of the rescue area; The signal coverage in mountainous areas is weak, and the predicted average signal strength is below -100 dBm; There are multiple signal blind spots in the area with dense buildings, and the area accounts for about 20% of the total area.

[0071] Resource Allocation Optimization: Graph Theory Model Construction: Regard the 6 base stations as nodes, and the areas with signal strength greater than -95 dBm form connected edges; It is calculated that the connectivity of Base Station 1 and Base Station 2 is the highest, and the criticality index score is 0.85; Due to terrain obstruction, the connectivity of Base Station 3 is weak, and the criticality index is 0.62; Resource Allocation Optimization: Allocate 40% of the bandwidth resources of Base Station 1 for command communication; Allocate 60% of the bandwidth resources of Base Station 2 for medical data transmission; Base Stations 3 - 6 dynamically allocate resources according to the density of rescue personnel in the coverage area; After 15 iterations of the multi-objective optimization algorithm, the optimal resource allocation plan is obtained, with the average resource utilization rate increased by 28% and the signal coverage rate increased to 83%.

[0072] Computing Task Scheduling: Application of the Microservices Architecture: Decompose the rescue communication system into 10 independent microservices: Location Service: Responsible for locating and navigating rescue personnel; Message Service: Responsible for text and voice information transmission; Video Service: Responsible for live video transmission; Data Analysis Service: Responsible for disaster information processing; Resource Scheduling Service: Responsible for material and personnel allocation; Medical Service: Responsible for managing the information of the wounded; Safety Monitoring Service: Responsible for environmental risk monitoring; Network Status Service: Responsible for network health monitoring; Power Management Service: Responsible for base station power scheduling; System Management Service: Responsible for overall system control; Elastic Scaling Policy: Location Service and Message Service, as core services, prioritize resource allocation; When network congestion occurs, automatically reduce the frame rate and resolution of the video service; When the search and rescue intensity in a certain area increases, automatically expand computing resources for that area; Environment Adaptation and Reconfiguration: Environmental Change Monitoring: It is detected that new buildings have collapsed in the northwest area (about 2 square kilometers) due to aftershocks; It is real-time monitored that the power of Base Station 3 has dropped to 30%, and the remaining working time is expected to be less than 4 hours; Prediction Model Analysis: It is predicted that the communication demand in the coverage area of Base Station 3 will reach its peak within 2 hours; It is predicted that 15 search and rescue teams will enter the northwest area, and the communication demand will increase by 200%; System Reconfiguration and Adaptation: Adjust the location of Base Station 4 to the edge of the northwest area (new coordinates: 30.14°N, 104.29°E); Set the power priority for Base Station 3 to reduce the power consumption of non-core services; Reallocate 20% of the bandwidth resources from Base Station 1 and Base Station 2 to support communication in the northwest area.

[0073] Application Effect: System Reliability Index: During the test period (72 hours), the overall system availability reached 99.2%; One of the 6 base stations had a short-term failure (about 45 minutes), and the system automatically reconfigured the communication network, maintaining an 85% coverage rate; The average system response time is less than 200ms, meeting the real-time communication requirements; Resource Utilization Efficiency: The average resource utilization rate of the base stations reached 72%, which is about 35% higher than the traditional static allocation; After dynamic adjustment, the signal blind area is reduced to 40% of the original; The battery usage efficiency is increased by 25%, and all base stations have maintained the expected working time; Improvement in Rescue Efficiency: Through precise resource allocation, 42 search and rescue tasks are supported simultaneously; The real-time transmission success rate of medical data reaches 98.5%, providing reliable support for remote medical guidance; The automatically adaptable feature of the system reduces the number of technician interventions, from an average of 3 times every 8 hours in the traditional system to 1 time every 24 hours.

[0074] The above application examples show that the rescue emergency multi-base station management system and method proposed in the present invention can provide high-reliability communication guarantee in complex and ever-changing disaster environments, supporting the efficient development of rescue work. The system maximizes the use of limited communication resources through intelligent resource allocation and real-time adaptation mechanisms, ensuring the stable operation of the rescue communication network under extreme conditions.

[0075] The embodiments of the present invention have been described above, but these embodiments are not limited to the above specific implementation manners. The above specific implementation manners are merely illustrative rather than restrictive. Under the inspiration of this embodiment, those of ordinary skill in the art can also make more equivalent embodiments in various forms, all of which fall within the protection scope of this embodiment.

Claims

1. A rescue emergency multi-base station management system, characterized in that Including: A parameter acquisition module, configured to acquire the geographical parameters of the rescue area, the function parameters and location parameters of multiple base stations; A signal coverage calculation module, configured to calculate the multi-dimensional signal coverage of multiple base stations under the influence of geographical parameters in the rescue area by using a composite propagation model, and the composite propagation model simultaneously considers terrain information and building information; A resource allocation module, configured to construct an optimization model for base station resource allocation with signal connectivity and resource utilization efficiency as constraint conditions according to the multi-dimensional signal coverage. The resource allocation optimization model adopts an adaptive iterative algorithm to dynamically adjust the allocation weight; A task scheduling module, configured to construct a computing task scheduling strategy based on elastic scaling based on the base station resource allocation optimization model to dynamically balance the system load and real-time response requirements; An environment monitoring module, configured to, when the communication environment in the rescue area changes, detect the change of environmental parameters in real time, trigger system reconstruction and task reallocation, and feedback the information to the signal coverage calculation module to re-execute signal coverage calculation, resource allocation and task scheduling.

2. The rescue and emergency multi-base station management system according to claim 1, characterized in that The composite propagation model in the signal coverage calculation module extracts features from historical data through a deep transfer learning method to adapt to the signal propagation characteristics of different rescue scenarios.

3. The rescue emergency multi-base station management system according to claim 1, characterized in that, The base station resource allocation optimization model in the resource allocation module includes: A network topology construction unit, configured to construct a network topology model based on graph theory, with base stations as nodes and signal connectivity as edges; A criticality calculation unit, configured to calculate the criticality indicators of each node, including coverage range, connectivity and load capacity; A resource allocation unit, based on the criticality indicators and resource requirements, calculates a resource allocation scheme by using a multi-objective optimization algorithm.

4. The rescue and emergency multi-base station management system according to claim 3, characterized in that, The multi-objective optimization algorithm adopts an improved particle swarm optimization algorithm, in which the update speed and position of particles are dynamically adjusted according to the change of coverage range and connectivity requirements.

5. The rescue emergency multi-base station management system according to claim 1, characterized in that, The computing task scheduling strategy in the task scheduling module is based on a microservice architecture, decomposes computing tasks into multiple independent microservices, and supports horizontal expansion and error isolation.

6. The rescue and emergency multi-base station management system according to claim 1, characterized in that, When detecting a change in the communication environment, the environment monitoring module uses a prediction model to predict the potential change trend of the environment and triggers a resource pre-allocation mechanism in advance.

7. A rescue emergency multi-base station management system according to claim 1, characterized in that, Also including: A cooperative communication module, configured to maintain a cooperative communication protocol between base stations to support resource sharing and command transmission of heterogeneous network devices.

8. The rescue emergency multi-base station management system according to claim 1, characterized in that Also including: A security mechanism module, configured to establish a distributed security mechanism to ensure normal communication functions can still be maintained in case of communication interruption or damage to some nodes.

9. The rescue emergency multi-base station management system according to claim 1, characterized in that, The function parameters of the base station acquired by the parameter acquisition module include: signal transmission power, frequency band, bandwidth, load capacity, battery capacity and standby time.

10. A rescue emergency multi-base station management method, which is used to execute a rescue emergency multi-base station management system according to any one of claims 1-9, characterized in that, Including the following steps: Acquire the geographical parameters of the rescue area, and acquire the function parameters and location parameters of multiple base stations; Calculate the multi-dimensional signal coverage of multiple base stations under the influence of geographical parameters in the rescue area by using a composite propagation model; Construct an optimization model for base station resource allocation with signal connectivity and resource utilization efficiency as constraint conditions according to the multi-dimensional signal coverage; Construct a computing task scheduling strategy based on elastic scaling based on the base station resource allocation optimization model; When the communication environment in the rescue area changes, the system detects the changes in environmental parameters in real time and triggers system reconstruction and task reallocation.

Images