A heterogeneous base station cooperative deployment optimization method and system for air-ground fusion communication network and storage medium

Abstract

The application relates to a kind of heterogeneous base station cooperative deployment optimization methods for air-ground fusion communication network, comprising: obtaining the spatial information and communication demand of candidate heterogeneous base station, ground user and low-altitude user, forming deployment state;According to spatial information and enabled candidate heterogeneous base station, determine communication quality and interference state, establish service association and configure communication resource, form resource configuration state;According to resource configuration state, evaluate deployment state, and according to deployment evaluation result and iteration number, judge whether to meet iteration stop condition;When not satisfied, adjust the enabled state of candidate heterogeneous base station to form new deployment state, and redetermine communication quality and interference state, reestablish service association and configure communication resource, iterate until meeting iteration stop condition, determine final deployment state and to be constructed base station set from the deployment state obtained by iteration.The application can improve the rationality, engineering implementability and comprehensive optimization quality of heterogeneous base station deployment result.

Description

Technical Field

[0001] This invention relates to the field of wireless communication network planning and optimization technology, and in particular to a method, system and storage medium for the collaborative deployment optimization of heterogeneous base stations for air-ground converged communication networks. Background Technology

[0002] With the development of 5G mobile communication enhancement technologies and low-altitude economic applications, services such as drone logistics, low-altitude inspection, emergency communication, and intelligent transportation are gradually increasing. The service targets of communication networks are gradually expanding from traditional ground terminals to low-altitude terminals with different flight altitudes and spatial distribution characteristics. Due to the differences in the spatial environment and business scenarios of different types of terminals, traditional network planning methods mainly designed for ground communication scenarios have certain limitations when applied to air-ground converged communication scenarios.

[0003] Existing base station planning methods typically determine the construction sites based on the location of candidate sites, preset coverage areas, or coverage evaluation results. However, in air-ground converged communication scenarios, due to the three-dimensional spatial characteristics of the service area and the complexity of the actual network environment, planning results obtained based on relatively fixed spatial coverage relationships may differ from actual network operation results, thus affecting the accuracy of the planning results.

[0004] In heterogeneous communication networks, different types of base stations differ in coverage, service capacity, construction investment, and operational characteristics. Macro base stations typically serve larger areas, while micro base stations are suitable for localized coverage or capacity supplementation. As the number and types of candidate base stations increase, the available combinations also increase, making heterogeneous base station site planning highly complex.

[0005] Furthermore, base station planning results are influenced by various factors in actual application, including network traffic load, communication resource conditions, and equipment capabilities. Some deployment schemes obtained under idealized conditions may fail to achieve the expected service effects or meet engineering implementation requirements after further consideration of actual network operating conditions. Therefore, using only simplified planning conditions to screen candidate sites may not fully reflect the actual feasibility of different deployment schemes.

[0006] Meanwhile, base station deployment typically involves multiple requirements, including coverage capability, service effectiveness, network performance, and construction investment, and these requirements may be mutually restrictive. As the number of candidate sites increases, the number of possible deployment combinations grows rapidly, making it computationally complex to obtain a deployment result with good overall performance within a large solution space.

[0007] Therefore, in air-ground converged communication networks, how to improve the adaptability of heterogeneous base station deployment schemes to complex network environments and engineering constraints, and obtain deployment results with better comprehensive performance and engineering feasibility within a large pool of candidate schemes, remains a technical problem that needs to be further solved in this field.

[0008] It should be noted that the information disclosed in the background section above is only for understanding the background of this application, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0009] The technical problem to be solved by this application is: how to improve the matching degree between heterogeneous base station deployment schemes and communication environment and resource constraints in complex air-ground converged communication networks, and obtain base station deployment results with high comprehensive performance and engineering feasibility.

[0010] The technical solution adopted in this application to solve the above-mentioned technical problems is as follows.

[0011] This application provides an optimization method for the cooperative deployment of heterogeneous base stations in an air-to-ground converged communication network, including: S1. Obtain spatial information of candidate heterogeneous base stations, ground users and low-altitude users, as well as the communication needs of ground users and low-altitude users, and form a deployment status that represents the activation status of base stations based on candidate heterogeneous base stations. S2. Based on the spatial information and the deployed candidate heterogeneous base stations, determine the communication quality between the candidate heterogeneous base stations and ground users and low-altitude users, as well as the interference status formed by multiple deployed candidate heterogeneous base stations. When determining the interference status, the residual interference coordination coefficient is used to correct co-channel interference. The residual interference coordination coefficient is used to characterize the comprehensive interference suppression effect produced by the interference coordination measures. S3. Based on communication quality, interference status and communication requirements, establish service associations between enabled candidate heterogeneous base stations and ground users and low-altitude users, and configure communication resources according to service associations to form a resource configuration status corresponding to the deployment status. S4. Evaluate the deployment status based on the resource configuration status, obtain the deployment evaluation result, and determine whether the preset iteration stop condition is met based on the deployment evaluation result and the number of iterations. If the iteration stop condition is met, determine the final deployment status from the deployment status obtained by iteration based on the deployment evaluation result corresponding to each deployment status. If the iteration stop condition is not met, adjust the activation status of the candidate heterogeneous base stations based on the deployment evaluation result to form a new deployment status. S5. For the new deployment status, redetermine the corresponding communication quality and interference status, and re-establish service associations and configure communication resources to form a new resource configuration status corresponding to the new deployment status. S6. If the iteration stop condition is not met, based on the new deployment state formed in step S4 and the new resource configuration state formed in step S5, iteratively execute steps S4 and S5 until the iteration stop condition is met, and determine the set of base stations to be built according to the final deployment state.

[0012] In some embodiments, candidate heterogeneous base stations include candidate macro base stations and candidate micro base stations; spatial information includes the three-dimensional location information of candidate macro base stations, candidate micro base stations, ground users, and low-altitude users; Step S1 also includes: obtaining the transmit power and antenna gain of the candidate heterogeneous base stations, as well as the minimum received power and minimum signal-to-interference-plus-noise ratio of ground users and low-altitude users; Step S2 further includes: determining the three-dimensional propagation loss between the candidate heterogeneous base station and the corresponding user based on the three-dimensional location information of the candidate heterogeneous base station and the ground user and the low-altitude user; The received power between the candidate heterogeneous base station and the corresponding user is determined based on the three-dimensional propagation loss, transmit power, and antenna gain. The dynamic signal-to-interference-plus-noise ratio is determined based on the received power of the activated candidate heterogeneous base stations in the deployment status and the co-channel interference caused by other activated candidate heterogeneous base stations to the corresponding users. Based on the comparison results of received power and minimum received power, and the comparison results of dynamic signal-to-interference-plus-noise ratio and minimum signal-to-interference-plus-noise ratio, the communication quality is determined, and the interference status is determined based on co-channel interference.

[0013] In some embodiments, step S1 further includes obtaining the available bandwidth, maximum service capacity, and maximum number of concurrent users of the candidate heterogeneous base stations; In step S3, the configuration of service association and communication resources satisfies resource configuration constraints, which include: each ground user or low-altitude user can establish a service association with at most one enabled candidate heterogeneous base station; Communication resources are configured only for ground users or low-altitude users with established service relationships; Communication resources include bandwidth resources. The total amount of bandwidth resources allocated by each enabled candidate heterogeneous base station to ground users and low-altitude users that have established service associations with it shall not exceed its available bandwidth. The sum of service rates provided by each enabled candidate heterogeneous base station through each service association shall not exceed its maximum service capacity, and the number of users that have established service associations with the enabled candidate heterogeneous base station shall not exceed its maximum number of concurrent users.

[0014] In some embodiments, step S1 further includes obtaining the priorities of ground users and low-altitude users; In step S3, under the condition of satisfying resource allocation constraints, for the enabled candidate heterogeneous base stations that can establish service associations with ground users and low-altitude users in various regions, the minimum bandwidth resources required to meet the communication requirements through the corresponding enabled candidate heterogeneous base stations are determined according to the communication requirements and the communication quality between the corresponding enabled candidate heterogeneous base stations and ground users or low-altitude users. The order of service association is determined based on the number of activated candidate heterogeneous base stations for which service associations can be established for ground users and low-altitude users in each region, as well as their communication needs and priorities. According to the service association order, for ground users or low-altitude users to be associated with services, from the enabled candidate heterogeneous base stations that have remaining bandwidth resources not less than the corresponding minimum bandwidth resources, remaining service capacity to meet communication needs, and remaining number of concurrent users to meet association requirements, the comprehensive load is determined based on bandwidth load, capacity load, concurrent user load, and interference load. The enabled candidate heterogeneous base station with the smallest comprehensive load is selected to establish service association, and the corresponding minimum bandwidth resources are configured for the ground users or low-altitude users.

[0015] In some embodiments, in step S1, the deployment state is formed in the following manner: For the candidate heterogeneous base stations that can be activated, the marginal revenue per unit cost of the candidate heterogeneous base stations is determined based on the coverage gain and communication demand satisfaction gain they generate for ground users and low-altitude users who have not yet established service associations, as well as the corresponding construction costs. Candidate heterogeneous base stations are activated sequentially based on the marginal benefit per unit cost, and the corresponding communication quality, interference status, service association, and communication resources are redefined after each activation of a candidate heterogeneous base station. Redundant candidate heterogeneous base stations that do not reduce deployment feasibility or degrade deployment evaluation results after being adjusted to an inactive state are removed to form a deployment state.

[0016] In some embodiments, step S1 further includes obtaining the construction cost of candidate heterogeneous base stations, the construction budget for deploying candidate heterogeneous base stations, and the mutual exclusion restrictions between candidate heterogeneous base stations; Step S4 further includes: adjusting the activation status of candidate heterogeneous base stations through destruction and repair strategies, wherein: The destruction strategy determines the activated candidate heterogeneous base stations to be adjusted from the deployment status based on at least one of the resource utility, interference status, construction cost, and spatial proximity relationship between activated candidate heterogeneous base stations, and adjusts the activated candidate heterogeneous base stations to be adjusted to the inactive state to form the deployment status after destruction. The repair strategy is based on at least one of the following: coverage gain of unactivated candidate heterogeneous base stations to ground users and low-altitude users, coverage capability to uncovered areas, impact on interference status after activation, and activation ratio of different base station types among the candidate heterogeneous base stations. It identifies candidate heterogeneous base stations to be activated from the unactivated candidate heterogeneous base stations, and adjusts the candidate heterogeneous base stations to be activated to the activated state under the conditions of meeting the construction budget and mutual exclusion constraints, so as to repair the deployment state after destruction and form a new deployment state. During the iterative execution of steps S4 and S5 in step S6, when step S4 evaluates the new deployment status based on the new resource configuration status, a new deployment evaluation result corresponding to the new deployment status is obtained, and the new deployment evaluation result is compared with the deployment evaluation result corresponding to the deployment status before the formation of the new deployment status according to the preset evaluation rules. If the new deployment evaluation result is not worse than the deployment evaluation result corresponding to the deployment state before the formation of the new deployment state, the new deployment state shall be accepted; if the new deployment evaluation result is worse than the deployment evaluation result corresponding to the deployment state before the formation of the new deployment state, the decision on whether to accept the new deployment state shall be made according to the preset acceptance rules. When accepting a new deployment state, the new deployment state and its corresponding new resource configuration state are respectively used as the current deployment state and the current resource configuration state; when not accepting a new deployment state, the adjustments made to the activation state of the candidate heterogeneous base stations through the destruction and repair strategies when the new deployment state is formed are revoked, and the deployment state before the formation of the new deployment state and its corresponding resource configuration state are restored to the current deployment state and the current resource configuration state, respectively. Based on the new deployment evaluation results corresponding to the new deployment status, rewards are given for the destruction and repair strategies adopted in this round. The selection weight of the corresponding destruction and repair strategies in subsequent iterations is adjusted according to the cumulative rewards and usage times of each destruction and repair strategy.

[0017] In some embodiments, step S1 further includes: obtaining the priorities of ground users and low-altitude users and the construction costs of candidate heterogeneous base stations; Step S4 includes: determining the coverage status of ground users and low-altitude users based on service association and resource configuration status, and determining the satisfaction of communication requirements based on resource configuration status; The priority-weighted coverage index is determined based on the coverage status and priority of ground users and low-altitude users; the communication demand satisfaction index is determined based on communication needs and their satisfaction; the interference evaluation index is determined based on the interference status; and the construction cost index is determined based on the construction cost of the candidate heterogeneous base stations that have been activated in the deployment status. The deployment evaluation result is determined based on the priority-weighted coverage index, the communication demand satisfaction index, the interference evaluation index, and the construction cost index. A Pareto non-dominated deployment state set is established based on the deployment evaluation results of each deployment state obtained through iteration; When the number of deployment states in the Pareto non-dominated deployment state set exceeds the preset limit, the most crowded deployment state is deleted according to the target space congestion distance of each deployment state until the number of deployment states in the Pareto non-dominated deployment state set does not exceed the preset limit. After the iteration stopping condition is met, the final deployment state is determined from the Pareto non-dominated deployment state set based on the deployment evaluation results.

[0018] In some embodiments, minimum coverage requirements are set for ground users and low-altitude users respectively; When establishing service associations and configuring communication resources, the coverage status of ground users and low-altitude users are determined separately. Based on the satisfaction of ground users and low-altitude users with their respective minimum coverage requirements, ground users or low-altitude users are alternately selected to establish service associations and configure communication resources.

[0019] In some embodiments, a heterogeneous base station collaborative deployment optimization system for air-ground converged communication networks is also provided, including an information acquisition module, a communication status determination module, a resource configuration module, a deployment optimization module, and an iterative control module; The information acquisition module is used to acquire spatial information of candidate heterogeneous base stations, ground users and low-altitude users, as well as the communication needs of ground users and low-altitude users, and to form a deployment status that represents the activation status of the base station based on the candidate heterogeneous base stations. The communication status determination module is used to determine the communication quality between the candidate heterogeneous base stations and ground users and low-altitude users, as well as the interference status formed by multiple activated candidate heterogeneous base stations, based on spatial information and the deployed status of the candidate heterogeneous base stations. The resource configuration module is used to establish service associations between enabled candidate heterogeneous base stations and ground users and low-altitude users based on communication quality, interference status and communication requirements, and configure communication resources according to the service associations to form a resource configuration status corresponding to the deployment status. The deployment optimization module is used to evaluate the deployment status based on the resource configuration status, obtain the deployment evaluation result, and determine whether the preset iteration stop condition is met based on the deployment evaluation result and the number of iterations. When the iteration stop condition is met, the final deployment status is determined from the deployment status obtained by iteration based on the deployment evaluation result corresponding to each deployment status. When the iteration stop condition is not met, the activation status of the candidate heterogeneous base stations is adjusted according to the deployment evaluation result to form a new deployment status. The iterative control module is used to control the communication state determination module to redetermine the corresponding communication quality and interference state for the new deployment state after a new deployment state is formed, and to control the resource configuration module to re-establish service associations and configure communication resources based on the redetermined communication quality and interference state, so as to form a new resource configuration state corresponding to the new deployment state; it is also used to control the deployment optimization module, communication state determination module and resource configuration module to iterate based on the new deployment state and the new resource configuration state until the iteration stop condition is met, and to determine the set of base stations to be built based on the final deployment state.

[0020] In some embodiments, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the heterogeneous base station collaborative deployment optimization method of the present invention.

[0021] The present invention has the following beneficial effects: This invention obtains spatial information of candidate heterogeneous base stations, ground users, and low-altitude users, as well as the communication needs of ground users and low-altitude users. Based on the candidate heterogeneous base stations, a deployment status characterizing the base station's activation status is formed. This enables the activation status of candidate heterogeneous base stations to establish a correspondence with the spatial distribution and communication needs of ground users and low-altitude users, providing a unified status basis for subsequent communication condition analysis, resource allocation, and deployment evaluation for different deployment statuses. By determining the communication quality between the candidate heterogeneous base stations and ground users and low-altitude users, as well as the interference status formed by multiple activated candidate heterogeneous base stations, based on spatial information and the activated candidate heterogeneous base stations in the deployment status, the communication conditions and interference situation corresponding to different deployment statuses can be reflected with the actual activation status of the candidate heterogeneous base stations. This reduces the deviation caused by evaluating the deployment status solely based on spatial location or static coverage relationship, and helps to improve the matching degree between the deployment status and the corresponding communication environment. By establishing service associations between activated candidate heterogeneous base stations and ground users and low-altitude users based on communication quality, interference status, and communication requirements, and configuring communication resources according to service associations, a resource configuration status corresponding to the deployment status is formed. This allows the corresponding communication conditions to be further implemented in service associations and communication resource configuration, so that the evaluation of the deployment status is based on its corresponding user service relationship and resource usage conditions. This is conducive to improving the matching degree between the deployment plan and communication requirements and resource constraints, and improving the engineering feasibility of the deployment results. By evaluating the deployment status based on the resource configuration status, the deployment evaluation results are obtained. Based on the deployment evaluation results and the number of iterations, it is determined whether the preset iteration stop condition is met. This allows the activation status of candidate heterogeneous base stations to be adjusted according to the deployment effect reflected by the corresponding resource configuration status. At the same time, when the iteration stop condition is met, the final deployment status is determined from the deployment status obtained by iteration based on the deployment evaluation results corresponding to each deployment status. This avoids using only the deployment status formed in a single iteration or the last deployment status as the final result, which is conducive to improving the overall performance of the final deployment status. By adjusting the activation status of candidate heterogeneous base stations based on deployment evaluation results when the iteration stopping condition is not met, a new deployment status is formed. The corresponding communication quality and interference status are redefined, service associations are re-established, and communication resources are reconfigured for the new deployment status, so as to form a new resource configuration status corresponding to the new deployment status. This allows the corresponding communication conditions, service associations, and communication resources to be updated as the activation status of candidate heterogeneous base stations changes. This avoids using the communication conditions and resource configurations of the original deployment status to evaluate the new deployment status, which is conducive to maintaining the matching relationship between the deployment status and the corresponding communication environment and resource constraints. By iteratively executing steps S4 and S5 based on the new deployment status and the new resource configuration status until the iteration stop condition is met, the deployment status adjustment, communication quality and interference status redetermination, service association establishment, communication resource configuration and deployment evaluation can continuously form a closed loop during the iteration process, and finally determine the set of base stations to be built based on the final deployment status.

[0022] In summary, this invention assigns different deployment states to corresponding communication quality, interference status, service association, communication resources, and resource configuration status, and re-establishes these correspondences after changes in the activation status of candidate heterogeneous base stations. This allows each deployment state to be evaluated under matching communication environment and resource constraints. Based on this, the deployment evaluation results drive iterative adjustments to the activation status of candidate heterogeneous base stations, and the final deployment state is determined from the iteratively obtained deployment states. This ensures that the impact of communication environment, communication requirements, and resource constraints on base station deployment results is coordinated during continuous optimization, thereby improving the matching degree between heterogeneous base station deployment schemes and communication environment and resource constraints in complex air-ground converged communication networks, and obtaining base station deployment results with high comprehensive performance and engineering feasibility.

[0023] Other beneficial effects of the present invention will be further described below. Attached Figure Description

[0024] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a two-dimensional distribution and 2 km grid density map of candidate base stations generated based on OpenCellID records in one embodiment of the present invention; Figure 2 This is a diagram showing the final base station site selection and ground and low-altitude user coverage results in one embodiment of the present invention; Figure 3 This is a bandwidth and capacity utilization graph of the 30 sites with the highest resource utilization in one embodiment of the present invention; Figure 4 This is a Pareto non-dominated solution set and a recommended compromise solution graph in one embodiment of the present invention; Figure 5 This is a convergence curve of the current value and the optimal value for adaptive large neighborhood search in one embodiment of the present invention. Detailed Implementation

[0025] The embodiments of the present invention will be described in detail below. It should be emphasized that the following description is merely exemplary and not intended to limit the scope and application of the present invention.

[0026] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of embodiments of the present invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0027] This invention provides an optimization method for the collaborative deployment of heterogeneous base stations in air-ground converged communication networks, in order to solve the problems that existing methods struggle to address in a coordinated manner, such as three-dimensional coverage, dynamic interference, heterogeneous base station types, user association, bandwidth resources, capacity, concurrent user limits, and multi-objective trade-offs.

[0028] This invention provides an optimization method for the collaborative deployment of heterogeneous base stations in air-ground converged communication networks, which improves the matching degree between the heterogeneous base station deployment scheme and the communication environment and resource constraints in complex air-ground converged communication networks, and obtains base station deployment results with high comprehensive performance and engineering feasibility.

[0029] It should be noted that, for the convenience of specific calculations and algorithm descriptions, some technical objects in the following embodiments are represented by specific names, abbreviations, variables or algorithmic terms; these representations are only used to illustrate specific implementations and should not be construed as limitations on the corresponding technical objects.

[0030] Candidate heterogeneous base stations may include candidate macro base stations and candidate micro base stations, and can also be represented by candidate locations, candidate sites, or candidate stations during specific site selection; ground users and low-altitude users can be collectively referred to as users when there is no need to distinguish their types; spatial information can be represented by planar coordinates and altitude information to represent three-dimensional location information, and communication requirements can be represented by service requirements. In this embodiment, the transmit power can be determined by the maximum transmit power combined with a power scaling factor, the antenna gain can be expressed as the base station antenna gain, the available bandwidth can be represented by the system bandwidth, and the priority can be represented by priority weights; the minimum received power and the minimum signal-to-interference-plus-noise ratio can be represented by corresponding thresholds in the calculation.

[0031] Deployment status is used to characterize the activation status of candidate heterogeneous base stations, and can be represented by a deployment vector composed of base station deployment variables or a set of site locations. During iteration, the deployment status serves as the basis for adjustments, the newly formed status, and the final determined status; in the algorithm description, it can be referred to as the current solution, the candidate solution, and the recommended solution, respectively. Communication quality can be represented by received power, dynamic signal-to-interference-plus-noise ratio (SINR), or effective link status. Three-dimensional propagation loss can be determined by line-of-sight loss, non-line-of-sight loss, expected path loss, or free space loss, depending on the propagation scenario. Interference status can be represented by dynamic co-channel interference or corresponding interference indicators. Service association can be represented by user association, and service rate can be represented by theoretically achievable rate.

[0032] Communication resources can be represented by bandwidth resources, and resource configuration can be constrained by the maximum service capacity and the maximum number of concurrent users; resource configuration constraints can be called hard resource constraints in the algorithm description. Bandwidth resources can also be calculated by bandwidth ratio, and the minimum bandwidth ratio obtained by the inverse solution of equation (12) is an intermediate quantity for determining the minimum bandwidth resources; the remaining bandwidth resources, remaining service capacity and remaining number of concurrent users can be simply referred to as remaining bandwidth, remaining capacity and remaining number of concurrent users, respectively. The service association order can be reflected by the corresponding user sorting, and the bandwidth load, capacity load, concurrent user load and interference load can be represented by the corresponding occupancy degree or normalized index. The resource configuration status can be represented by at least one of user association, bandwidth allocation, capacity occupancy and concurrent user occupancy; the coverage status can be represented by the effective service status of users or the coverage result, and the minimum coverage requirement can be represented by the minimum coverage rate or coverage threshold.

[0033] Coverage gain can be further represented by priority-weighted coverage gain; communication demand satisfaction gain can be simply referred to as demand satisfaction gain; marginal benefit per unit cost can be represented by marginal score per unit cost; deployment feasibility can be reflected by indicators such as feasibility gap; redundant candidate heterogeneous base stations can be simply referred to as redundant sites; construction budget can be simply referred to as budget or total budget; mutual exclusion constraints can be expressed as mutual exclusion conditions or mutual exclusion constraints. Destruction and repair strategies can be implemented using destruction and repair operators, respectively. Resource utility, spatial proximity, uncovered areas, and the activation ratio of different base station types can be characterized, for example, by inefficiency, spatial proximity or spatial clustering, coverage blind spots, and the ratio of macro to micro base stations. Preset acceptance rules can be implemented, for example, using a simulated annealing acceptance mechanism.

[0034] The deployment evaluation results can include a multi-objective evaluation vector and a comprehensive evaluation value formed by priority-weighted coverage indicators, communication demand satisfaction indicators, interference evaluation indicators, and construction cost indicators. These indicators can be represented, for example, by priority-weighted coverage rate, demand satisfaction rate, normalized interference target, and normalized construction cost. The multi-objective evaluation vector can be used for Pareto non-dominated judgment, and the comprehensive evaluation value can be used for state acceptance and compromise scheme selection. The Pareto non-dominated deployment state set can be maintained using Pareto archives or Pareto solution sets, and the congestion level of the target space can be represented by congestion distance. The cumulative reward and selection weight can be referred to as operator reward and operator weight, respectively. The iteration stopping condition can be a maximum number of iterations or a consecutive preset number of times without improvement. Interference coordination parameters can be represented by residual interference coordination coefficients, power adjustment parameters can be represented by base station power scaling coefficients, and preset deployment feasibility requirements can be determined through constraints such as budget, coverage, bandwidth, capacity, and concurrent users. The set of base stations to be built can be called the recommended deployment set or the recommended site set; when the deployment information and network operation evaluation information are specifically output, they can be represented by specific information such as base station identifier or site number, base station type or site type, spatial location, resource utilization rate, user coverage rate, communication demand satisfaction rate, signal-to-interference-plus-noise ratio and interference evaluation results.

[0035] For system implementation, the information acquisition module, communication status determination module, resource configuration module, deployment optimization module, and iterative control module can be implemented through data acquisition and status construction, propagation and interference calculation, user association and resource allocation, deployment status evaluation and adjustment, and iterative update process, respectively.

[0036] The method steps of the present invention are described below by way of example.

[0037] Step S1: Constructing an air-ground integrated heterogeneous communication scenario

[0038] Obtain open cellular base station records or operator-planned site records for the target area. Each original site record includes at least the site or cell identifier, radio access standard, longitude, latitude, and optional estimated coverage radius. Perform coordinate validity checks, duplicate record processing, and study area cropping on the records, and convert the latitude and longitude into local planar coordinates in meters with the data geometric center as the origin. This yields a set of candidate locations. The raw, open data is only used to determine buildable locations and spatial distribution.

[0039] Based on the planning scenario, configure engineering station types for candidate locations and establish a candidate base station set. ,in For the candidate macro base station set, Establish a candidate micro base station set; establish a user set. , in For ground users, For low-altitude user groups. Candidate base stations. The position is recorded as User The position is recorded as .

[0040] Configure construction costs for each candidate base station Bandwidth Maximum transmission power Base station antenna gain Maximum business capacity and maximum concurrent users Configure business requirements for each user. Priority weight Minimum received power and lowest signal-to-interference-plus-noise ratio Before solving, the east-west and north-south spans, envelope density, nearest neighbor distances, and number of stations within the grid of candidate locations are statistically analyzed to identify hotspot clusters, sparse peripheral areas, and potential coverage blind spots, and to provide spatial basis for subsequent spatial cluster destruction and blind spot repair operators.

[0041] Step S2: Define joint decision variables

[0042] Define base station deployment variables User-related variables User effective service variables and bandwidth ratio variable . in Indicates the candidate position Building base stations Indicates user Associated base stations , Indicates user The communication needs are met. Indicates base station Assigned to user The proportion of bandwidth of this site.

[0043] Step S3: Establish a three-dimensional propagation and dynamic interference model

[0044] Candidate base stations With users The three-dimensional distance between them is

[0045] Determine the line-of-sight probability based on user type, elevation angle, and environmental parameters. For ground users, macro base stations use an urban macrocell propagation model, while micro base stations use an urban microcell street canyon propagation model. In one implementation, the carrier frequency... In GHz, the line-of-sight probability is

[0046] Two-dimensional distance for

[0047] And calculate the expected path loss.

[0048] A series of path loss formulas based on empirical measurements are defined in 3GPP TR 38.901. Here, the ground line-of-sight loss for macro base stations is defined as...

[0049] Its non-line-of-sight loss is

[0050] The ground line-of-sight loss of micro base stations is

[0051] Its non-line-of-sight loss is

[0052] For low-altitude users, the free space loss considering elevation angle and line-of-sight probability is:

[0053] The received power is expressed as

[0054] in For antenna gain. After converting the received power to the watt domain, given the deployment vector... The dynamic signal-to-interference-plus-noise ratio at that time is

[0055] in For noise power, This is the base station power scaling factor. This represents the remaining interference coordination coefficient. In response to the conservative scenario of full frequency synchronization and full activity, This is used to characterize the comprehensive suppression effect brought about by time-frequency peak shifting, beam isolation, base station silence, and interference coordination. When determining the interference state corresponding to the current deployment state, the co-channel interference power generated by other enabled candidate heterogeneous base stations to the corresponding users is summarized, and the summarized co-channel interference power is corrected using the remaining interference coordination coefficient. The corrected co-channel interference power is used to determine the dynamic signal-to-interference-plus-noise ratio.

[0056] Step S4: Establish user association and bandwidth allocation model

[0057] link The approximate model of the theoretically achievable rate during the planning phase is expressed as follows:

[0058] User association and resource allocation satisfy

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065] Equation (15) prevents bandwidth from being allocated to users who have not established an association. Because... are binary variables and The original model is in the form of The bilinear term can be precisely replaced by After fixing the deployment vector in the outer layer, Since is a constant, the bandwidth allocation subproblem can be solved by linear programming or by the resource feasible sorting algorithm that satisfies equations (13)–(19).

[0066] In the process of resource feasibility ranking, bandwidth load, capacity load, concurrent user load and interference load can be determined, for example, based on bandwidth utilization, service capacity utilization, concurrent user utilization and normalized interference level, respectively. The overall load can be determined by weighting the above loads after normalization.

[0067] Step S5: Establish deployment, communication quality, and cluster coverage constraints

[0068] Deployment budget meets

[0069] For candidate stations that cannot be built simultaneously , satisfy

[0070] Ground users and low-altitude users respectively meet the minimum coverage requirements

[0071]

[0072] Construct a set of effective links based on the applicable domain of the propagation model and the minimum received power.

[0073] in, This is the minimum receive power threshold; This is the linear dimensionless threshold for the lowest signal-to-interference-plus-noise ratio.

[0074] right Direct order The signal-to-interference-plus-noise ratio (SIR) condition for association-triggered activation of effective links should meet the following requirements:

[0075] Step S6: Establish a multi-objective evaluation model

[0076] First, we define priority-weighted coverage and demand fulfillment rate as follows:

[0077]

[0078] Obviously .

[0079] for Users, define Its only associated service base station, namely ,definition

[0080] Then the normalized interference target is

[0081] The empty solution takes a value of 1 as a penalty for algorithm evaluation, used to avoid division by zero, and does not indicate that the empty network has the actual maximum physical interference.

[0082] Normalized construction cost is

[0083] Multi-objective vector is defined as The scalar evaluation function used for searching and selecting acceptable and compromise solutions is:

[0084] in , Normalized violation amounts for constraints such as budget, mutual exclusion, and cluster coverage. This is the penalty coefficient.

[0085] Step S7: Solve using a greedy initial solution and an adaptive large neighborhood search.

[0086] First, construct a set of empty sites. For each candidate site that is budget-feasible and does not violate the mutual exclusion condition, estimate its priority-weighted coverage gain and demand satisfaction gain for unserved users, and normalize them according to construction cost. Select the candidate site with the largest marginal score and add it to the current solution. Then, recalculate the dynamic interference using Equation (11) and re-execute user association and bandwidth allocation. If the candidate site reduces the feasibility gap or increases the comprehensive target value, then accept the site. Repeat the above operation until the coverage constraint, budget limit, or no positive marginal gain is reached, and delete redundant sites that do not affect feasibility and target value to obtain the greedy initial solution.

[0087] An adaptive large neighborhood search is performed based on the initial greedy solution. Destruction operators include random destruction, inefficient site destruction, high-interference site destruction, spatial cluster destruction, and high-cost site destruction; repair operators include coverage gain repair, blind zone repair, interference-aware repair, macro-micro site balancing repair, and random repair. Each repair is subject to budget and mutual exclusion constraints, and a complete dynamic communication state evaluation is performed after repair.

[0088] Among them, the destruction of inefficient sites can be determined based on resource utility, which can be evaluated based on at least one of the following: the number of associated users, bandwidth utilization, capacity utilization, and contribution to meeting communication needs; spatial cluster destruction is used to reflect spatial proximity, blind spot repair is used to reflect the coverage capability of uncovered areas, and macro-micro station balance repair is used to reflect the activation ratio of different base station types.

[0089] The objective difference between the candidate solution and the current solution is denoted as . when Accept candidate solutions when;

[0090]

[0091] The probability of accepting candidate solutions is given, the temperature decreases, and the temperature is increased again after several consecutive failures to improve the solution.

[0092] Destroy or repair operator The weights are updated based on segmented feedback:

[0093] in The reaction coefficient, For the cumulative rewards in this section, The number of uses determines the reward, from highest to lowest, for each action: generating a global improvement, being added to the Pareto archive, improving the current solution, or being accepted.

[0094] Algorithm 1: A collaborative deployment algorithm combining greedy initialization and adaptive large neighborhood search.

[0095] Input: Candidate station set User set Link matrix, budget Coverage threshold

[0096] Output: Recommended deployment set User association Bandwidth allocation Pareto Archives

[0097] 1: Calculate the evaluation results of the empty solution.

[0098] 2: while there exists a budget and mutually exclusive candidate station do with positive marginal revenue.

[0099] 3: Calculate the unit cost coverage – marginal demand score for each candidate station.

[0100] 4: Add the highest-scoring site Recalculate dynamic SINR, correlation, and bandwidth.

[0101] 5. If the feasibility gap or overall objectives are not improved, the candidate site shall be withdrawn.

[0102] 6: end while

[0103] 7: Remove redundant sites that do not degrade the solution quality, and obtain...

[0104] 8: , Initialize operator weights and temperature and archives

[0105] 9: for up to the maximum number of iterations do

[0106] 10: Select the destruction operator based on the weighted roulette wheel. and repair operator

[0107] 11: and implement budget and mutual exclusion feasibility repair.

[0108] 12: Yes Recalculate dynamic SINR, user association, bandwidth, capacity, and concurrency usage.

[0109] 13: Decide whether to use formula (32) replace

[0110] 14: If If not dominated, update the Pareto archive.

[0111] 15: Update the current optimal solution, operator reward, operator weight, and temperature.

[0112] 16: end for

[0113] 17: From Select the recommended compromise solution according to equation (31) and output it.

[0114] In the above iteration, when a candidate solution is accepted, the candidate solution is taken as the current deployment state, and its corresponding user association, bandwidth allocation, capacity usage, and concurrent user usage are taken as the current resource configuration state; when a candidate solution is not accepted, the original current deployment state and its corresponding current resource configuration state are retained.

[0115] For a fixed site set, the user association and bandwidth repair process is as follows: First, links with insufficient received power or signal-to-interference-plus-noise ratio are filtered out; then, users are sorted according to the number of available base stations from few to many, and the combination of demand and priority, and the ground and low-altitude user coverage targets are promoted alternately; for each user, the base station with the smallest overall load is selected from the base stations that meet the requirements in terms of remaining bandwidth, remaining capacity and remaining concurrency; the minimum required to meet the demand is solved by inverse solution of equation (12). This ensures that each set of sites obtained from the outer neighborhood search corresponds to an inner solution with executable resources.

[0116] Among them, the minimum obtained by the inverse solution of equation (12) The minimum bandwidth ratio is the actual minimum bandwidth resource determined by the minimum bandwidth ratio and the available bandwidth of the corresponding candidate heterogeneous base stations.

[0117] Step S8: Maintain Pareto solution sets and output deployment results

[0118] Each feasible candidate solution is represented as a four-dimensional target vector. If a candidate solution is not dominated by any solution in the archive, the archive solution dominated by that candidate solution is deleted and added to the candidate solution; when the archive exceeds the upper limit, the most crowded solution is deleted according to the target space congestion distance to maintain scheme diversity. Finally, the system outputs at least one of the following for the candidate heterogeneous base stations in the set of base stations to be built: base station identifier (e.g., site number), base station type (e.g., station type), spatial location, station height, construction cost, transmit power, available bandwidth, maximum service capacity, and maximum number of concurrent users. It also outputs at least one of the following network operation evaluation information: number of associated users, bandwidth utilization, capacity utilization, user coverage, communication demand satisfaction rate, signal-to-interference-plus-noise ratio, and interference index. Pareto alternative schemes can also be output. It should be noted that the target space congestion distance is used to characterize the interval between the corresponding deployment state and its adjacent deployment states in the target space; the smaller the target space congestion distance, the closer the corresponding deployment state is to its adjacent deployment state. Therefore, the deployment state with the smallest target space congestion distance can be regarded as the most crowded deployment state.

[0119] The following description, in conjunction with the accompanying drawings, provides an exemplary account of specific embodiments of the present invention.

[0120] Example 1: Data Construction and Parameter Setting

[0121] Retrieve 500 base station records for a specific area from the OpenCellID open cellular database. The original fields include radio access standard, mobile country code, mobile network code, location area or tracking area identifier, cell identifier, longitude, latitude, and estimated coverage radius. First, check the latitude and longitude range and retain the source row number. Then, using the geometric center of the sample's latitude and longitude as the projection center, convert the 500 records into a local plane. coordinate.

[0122] Figure 1 The left side shows the two-dimensional projection of the candidate locations, and the right side shows the 2 km grid density. Sites form continuous hotspot clusters in the central-southern region, while exhibiting a strip-like or discrete distribution in the western, northern, and peripheral regions. This indicates that the input is not a uniform random set of points, but rather possesses a combination of local high density, peripheral low density, and a multi-center structure. This distribution characteristic necessitates subsequent site selection that addresses both bandwidth competition and co-channel interference in hotspot areas, as well as coverage blind spots in sparse areas.

[0123] In the engineering parameter scenario of this embodiment, the 500 candidate locations are configured as 100 candidate macro base station locations and 400 candidate micro base station locations. The 500 m grid is only used for data quality checks and mutual exclusion candidate point identification. The optimization algorithm still uses continuous two-dimensional coordinates and station height, and does not force the candidate locations to move to the regular grid.

[0124] User network data consists of user demand samples generated within the spatial envelope of candidate stations. During generation, a combination of hotspot and background distribution is used, concentrating some users in densely populated candidate station areas while retaining users in sparse areas to test blind spot repair capabilities. This embodiment generates 1250 users, including 1000 ground users and 250 low-altitude users; the total service demand is approximately... Gbps.

[0125] The main parameters of this embodiment are shown in Table 1.

[0126] Table 1 Main parameters of the embodiment

[0127] To reflect the interference coordination scenario, the remaining interference coordination coefficient in the main scenario is set to... This is equivalent to a 30dB combined effect of time-frequency activity, beam isolation, and coordinated suppression. This parameter needs to be calibrated during actual deployment based on frequency reuse, antenna pattern, traffic activity factor, or field measurement results. Therefore, this embodiment simultaneously performs different... Sensitivity analysis of the selected values.

[0128] Example 2: Greedy Initial Solution and ALNS Optimization Results

[0129] The resource balancing greedy selection is 112 stations (macro 32, micro 80), cost 240; ground coverage 89.90%, low-altitude coverage 96.40%, weighted coverage 91.60%, demand satisfaction 89.03%, and overall target 0.7748.

[0130] After ALNS optimization, the recommended solution is 113 stations (30 macro stations, 83 micro stations), with a cost of 233; serving 1144 users, with ground coverage of 90.30%, low-altitude coverage of 96.40%, weighted coverage of 92.00%, demand satisfaction of 90.38%, and an overall target of 0.7909.

[0131] Table 2 shows a comparison between the greedy initial solution and the recommended solution.

[0132] Table 2 Comparison between the initial greedy solution and the ALNS recommended solution

[0133] Figure 2The results show the recommended sites and user coverage. Unserved users are mainly distributed at the spatial edge or in areas with sparse candidate sites, indicating that while the spatial blind spot repair operator achieves high coverage in hotspot areas, there are still edge service gaps limited by budget and candidate locations.

[0134] Figure 3 The algorithm identifies the 30 sites with the highest bandwidth utilization. No site had a bandwidth sum exceeding 1, a capacity utilization exceeding 1, or a concurrent user count exceeding [a certain threshold]. The situation.

[0135] Example 3: Comparative Analysis with Traditional Optimization Algorithms

[0136] To evaluate the performance of adaptive large neighborhood search compared to traditional metaheuristic algorithms, genetic algorithm (GA) and tabu search (TS) were selected as benchmarks. GA employs site set encoding, tournament selection, set crossover, random mutation, elite preservation, and budget repair; the population size is 10. TS employs fixed neighborhood site replacement operation and optimal solution pardon criterion; 10 neighborhood solutions are generated per round, for a total of 15 rounds. All three algorithms use the same candidate site data, user data, propagation model, resource allocation model, and interference coordination coefficient. 1. Initial solution using a common greedy algorithm. Data loading and the initial solution using a common greedy algorithm are not included in the algorithm time. Each algorithm was run independently 5 times using a random seed, and the mean was reported. ALNS allows early termination when it meets the convergence condition; in this experiment, an average of 355 evaluations were performed. ALNS obtained feasible solutions in all five runs, and its overall objective value was better than GA and similar to TS. ALNS achieved competitive coverage and requirement satisfaction results with a shorter average running time than TS. Statistical results are shown in Table 3.

[0137] Table 3. Comparison of five independent runs of the traditional algorithm and ALNS (data in the table are averages).

[0138] Example 4: Pareto Scheme and Convergence Analysis

[0139] After performing ALNS optimization based on the initial greedy solution, 18 Pareto feasible non-dominated solutions were retained, satisfying the constraints of budget, coverage, capacity, concurrent users, and bandwidth. The horizontal axis represents priority-weighted coverage, and the vertical axis represents demand satisfaction rate. Each point also has different construction costs and interference targets, so no single point can be strictly improved by another point in all four objectives. The red points represent the recommended compromise solutions. Planners can adjust... to The system selects deployment options from the available resources that prioritize coverage, business requirements, low disruption, or low cost, without needing to redefine the constraint model. Location P12, with the highest overall goal and requirement fulfillment rate, was therefore chosen as the final recommended compromise.

[0140] Figure 5 The figure illustrates the convergence process of the ALNS algorithm in this embodiment. The vertical axis represents the historical best comprehensive objective value accumulated during the search process, and the horizontal axis represents the number of ALNS iterations (due to the use of historical best values, the curve exhibits a monotonically non-decreasing step-like characteristic). This embodiment sets the maximum number of ALNS iterations to 500 and employs an early termination condition of 50 consecutive iterations without global improvement. The algorithm's optimal comprehensive objective value rapidly increases in the early stages of the search, and after several plateau periods in the middle stages, it continues to produce small improvements. The 296th iteration yields the final optimal value of 0.790869, and thereafter, it remains unchanged for 54 consecutive iterations, stabilizing until the 350th iteration. This demonstrates that ALNS can quickly obtain high-quality solutions in the early stages and, through destruction, repair, and restart mechanisms, escape local optima, further finding better deployment schemes in the later stages of the search.

[0141] Example 5: Interference Coordination Sensitivity Analysis

[0142] Keep the recommended site set unchanged, and set them separately. Sensitivity analysis was performed using values ​​of 0.01, 0.003, and 0.001, and user association and bandwidth allocation were recalculated in each scenario.

[0143] Table 4 Residual Interference Coordination Coefficients

[0144] The results show that coverage and demand are sensitive to residual interference. Before deployment, calibration should be performed based on frequency reuse, antenna patterns, service activity records, or on-site measurements. Simulation parameters should not be interpreted as the performance of the existing network that can be obtained without configuration.

[0145] Example 6: Recommended Site Output

[0146] In this embodiment, the high proportion of micro base stations in the final output scheme is a result of the combined effects of the current macro-micro base station cost ratio, capacity parameters, candidate locations, and budget. In other engineering scenarios, the algorithm output can be adapted to the specific scenario by configuring macro-micro base station costs, minimum or maximum number of base station types, backhaul capacity, coverage radius, or maintenance costs.

[0147] The above embodiments illustrate that the present invention can generate a set of heterogeneous sites, user associations and bandwidth schemes that meet the hard constraints of resources, starting from candidate site data and user demand data, and provide auditable results such as convergence, Pareto trade-offs, resource utilization and interference sensitivity.

[0148] Compared with the prior art, the present invention has at least the following beneficial effects: 1. Simultaneously describe the three-dimensional location, service requirements, and communication quality thresholds of both ground users and low-altitude users to improve the applicability of air-ground converged network planning; 2. Recalculate dynamic co-channel interference after each site change to avoid deployment evaluation bias caused by using a fixed coverage matrix; 3. Employ an outer-layer addressing, inner-layer associativity, and minimum bandwidth allocation structure that meets service requirements, through... Ensure consistency in bandwidth and enforce hard constraints on bandwidth, maximum service capacity, and maximum concurrent users; 4. A greedy initial solution is adopted to narrow the search range, and multiple destruction and repair operators and simulated annealing acceptance mechanism are used to escape local optima; 5. Maintain Pareto non-dominated solution sets, providing multiple options for engineering trade-offs between coverage, capacity, interference, and cost; 6. Output auditable resource results for each station and each user, which facilitates planners to verify the plan and implement subsequent parameter calibration.

[0149] Those skilled in the art will understand that embodiments of the present invention may take the form of a method, a system, or a computer-readable storage medium. The computer-readable storage medium may store a computer program that, when executed by a processor, implements the above-described heterogeneous base station collaborative deployment optimization method; the computer-readable storage medium may, for example, include a disk storage device, a CD-ROM, an optical storage medium, etc.

[0150] This invention can be described with reference to the method flow and system functional relationships, and the relevant functions can be implemented by computer program instructions. These computer program instructions can be executed by a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to implement the corresponding method steps or system functions.

[0151] The computer program instructions may be stored in a computer-readable storage medium, causing a processor to read and execute the computer program instructions.

[0152] The computer program instructions can also be loaded onto a computer or other programmable data processing equipment to achieve the above functions by executing corresponding operation steps.

[0153] The background section of this invention may include background information about the problems or environment in which the invention is being developed, and is not necessarily a description of prior art. Therefore, the content included in the background section does not constitute an admission of prior art by the applicant.

[0154] The above description provides a further detailed explanation of the present invention in conjunction with specific or preferred embodiments, and it should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various substitutions or modifications can be made to these described embodiments without departing from the concept of the present invention, and all such substitutions or modifications should be considered within the scope of protection of the present invention. In the description of this specification, the reference to terms such as "an embodiment," "some embodiments," "preferred embodiment," "example," "specific example," or "some examples," etc., indicates that the specific features, structures, materials, or characteristics described in connection with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. Without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification and the features of different embodiments or examples. Although the embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions, and modifications can be made herein without departing from the scope of protection of the patent application.

Claims

1. A method for optimizing the collaborative deployment of heterogeneous base stations in an air-to-ground converged communication network, characterized in that, include: S1. Obtain spatial information of candidate heterogeneous base stations, ground users and low-altitude users, as well as the communication needs of the ground users and the low-altitude users, and form a deployment status characterizing the base station activation status based on the candidate heterogeneous base stations. S2. Based on the spatial information and the candidate heterogeneous base stations that have been enabled in the deployment status, determine the communication quality between the candidate heterogeneous base stations and the ground user and the low-altitude user, as well as the interference status formed by multiple enabled candidate heterogeneous base stations; wherein, when determining the interference status, the residual interference coordination coefficient is used to correct the co-channel interference, and the residual interference coordination coefficient is used to characterize the comprehensive interference suppression effect produced by the interference coordination measures. S3. Based on the communication quality, the interference status, and the communication requirements, establish a service association between the enabled candidate heterogeneous base station and the ground user and the low-altitude user, and configure communication resources according to the service association to form a resource configuration status corresponding to the deployment status. S4. Evaluate the deployment status based on the resource configuration status to obtain the deployment evaluation result, and determine whether the preset iteration stop condition is met based on the deployment evaluation result and the number of iterations; if the iteration stop condition is met, determine the final deployment status from the deployment status obtained by iteration based on the deployment evaluation result corresponding to each deployment status; if the iteration stop condition is not met, adjust the activation status of the candidate heterogeneous base stations based on the deployment evaluation result to form a new deployment status. S5. For the new deployment state, redetermine the corresponding communication quality and interference state, and re-establish service associations and configure communication resources to form a new resource configuration state corresponding to the new deployment state. S6. If the iteration stop condition is not met, based on the new deployment state formed in step S4 and the new resource configuration state formed in step S5, iteratively execute steps S4 and S5 until the iteration stop condition is met, and determine the set of base stations to be built according to the final deployment state.

2. The method for optimizing the collaborative deployment of heterogeneous base stations according to claim 1, characterized in that, The candidate heterogeneous base stations include candidate macro base stations and candidate micro base stations; the spatial information includes the three-dimensional location information of the candidate macro base stations, the candidate micro base stations, the ground users, and the low-altitude users; Step S1 further includes: obtaining the transmit power and antenna gain of the candidate heterogeneous base station, as well as the minimum received power and minimum signal-to-interference-plus-noise ratio of the ground user and the low-altitude user; Step S2 further includes: determining the three-dimensional propagation loss between the candidate heterogeneous base station and the corresponding user based on the three-dimensional position information of the candidate heterogeneous base station, the ground user, and the low-altitude user; The received power between the candidate heterogeneous base station and the corresponding user is determined based on the three-dimensional propagation loss, the transmit power, and the antenna gain. The dynamic signal-to-interference-plus-noise ratio is determined based on the received power of the activated candidate heterogeneous base stations in the deployment state and the co-channel interference generated by other activated candidate heterogeneous base stations to the corresponding users. The communication quality is determined based on the comparison between the received power and the minimum received power, and the comparison between the dynamic signal-to-interference-plus-noise ratio and the minimum signal-to-interference-plus-noise ratio, and the interference state is determined based on the co-channel interference.

3. The method for optimizing the collaborative deployment of heterogeneous base stations according to claim 1, characterized in that, Step S1 also includes obtaining the available bandwidth, maximum service capacity, and maximum concurrent users of the candidate heterogeneous base stations; In step S3, the configuration of the service association and the communication resources satisfies resource configuration constraints, which include: Each ground user or low-altitude user may establish the service association with at most one enabled candidate heterogeneous base station; The communication resources are configured only for ground users or low-altitude users who have established the service association. The communication resources include bandwidth resources. The total amount of bandwidth resources allocated by each enabled candidate heterogeneous base station to ground users and low-altitude users that have established service associations with it does not exceed its available bandwidth. The sum of the service rates provided by the enabled candidate heterogeneous base station through each service association does not exceed its maximum service capacity, and the number of users that have established service associations with the enabled candidate heterogeneous base station does not exceed its maximum number of concurrent users.

4. The method for optimizing the collaborative deployment of heterogeneous base stations according to claim 3, characterized in that, Step S1 also includes obtaining the priorities of the ground users and the low-altitude users; In step S3, under the condition of satisfying the resource configuration constraints, for the enabled candidate heterogeneous base stations that can establish service associations with ground users and low-altitude users in various places, the minimum bandwidth resources required to satisfy the communication requirements through the corresponding enabled candidate heterogeneous base stations are determined according to the communication requirements and the communication quality between the corresponding enabled candidate heterogeneous base stations and the ground users or the low-altitude users. The order of service association is determined based on the number of activated candidate heterogeneous base stations for which service associations can be established for ground users and low-altitude users in each region, as well as their communication needs and priorities. According to the service association order, for ground users or low-altitude users to be associated with services, from the enabled candidate heterogeneous base stations with remaining bandwidth resources not less than the corresponding minimum bandwidth resources, remaining service capacity to meet the communication requirements, and remaining number of concurrent users to meet the association requirements, the comprehensive load is determined based on bandwidth load, capacity load, concurrent user load, and interference load. The enabled candidate heterogeneous base station with the smallest comprehensive load is selected to establish the service association, and the corresponding minimum bandwidth resources are configured for the ground user or the low-altitude user.

5. The method for optimizing the collaborative deployment of heterogeneous base stations according to claim 1, characterized in that, In step S1, the deployment status is formed in the following way: For the candidate heterogeneous base stations that can be activated, the marginal revenue per unit cost of the candidate heterogeneous base stations is determined based on the coverage gain and communication demand satisfaction gain they generate for ground users and low-altitude users who have not yet established service associations, as well as the corresponding construction costs. Candidate heterogeneous base stations are activated sequentially based on the marginal benefit per unit cost, and the corresponding communication quality, interference status, service association, and communication resources are re-determined after each activation of a candidate heterogeneous base station. Redundant candidate heterogeneous base stations that do not reduce deployment feasibility or degrade the deployment evaluation results after being adjusted to an inactive state are deleted to form the deployment state.

6. The method for optimizing the collaborative deployment of heterogeneous base stations according to claim 1, characterized in that, Step S1 also includes obtaining the construction cost of the candidate heterogeneous base station, the construction budget for deploying the candidate heterogeneous base station, and the mutual exclusion restrictions between the candidate heterogeneous base stations; Step S4 further includes: adjusting the activation status of the candidate heterogeneous base stations through a destruction strategy and a repair strategy, wherein: The destruction strategy determines the activated candidate heterogeneous base stations to be adjusted from the deployment state based on at least one of the resource utility, interference status, construction cost, and spatial proximity relationship between activated candidate heterogeneous base stations, and adjusts the activated candidate heterogeneous base stations to be adjusted to the inactive state to form the deployment state after destruction. The repair strategy determines candidate heterogeneous base stations to be activated from the unactivated candidate heterogeneous base stations based on at least one of the following: coverage gain of the unactivated candidate heterogeneous base stations to the ground users and the low-altitude users, coverage capability to the uncovered areas, impact on the interference state after activation, and activation ratio of different base station types among the candidate heterogeneous base stations. Under the condition of meeting the construction budget and the mutual exclusion limit, the candidate heterogeneous base stations to be activated are adjusted to the activated state to repair the deployment state after destruction and form the new deployment state. During the iterative execution of steps S4 and S5 in step S6, when step S4 evaluates the new deployment state based on the new resource configuration state, a new deployment evaluation result corresponding to the new deployment state is obtained, and the new deployment evaluation result is compared with the deployment evaluation result corresponding to the deployment state before the formation of the new deployment state according to the preset evaluation rules. If the new deployment evaluation result is not worse than the deployment evaluation result corresponding to the deployment state before the formation of the new deployment state, the new deployment state is accepted; if the new deployment evaluation result is worse than the deployment evaluation result corresponding to the deployment state before the formation of the new deployment state, it is determined whether to accept the new deployment state according to the preset acceptance rules. When the new deployment state is accepted, the new deployment state and its corresponding new resource configuration state are respectively used as the current deployment state and the current resource configuration state; when the new deployment state is not accepted, the adjustment made to the activation state of the candidate heterogeneous base station through the destruction strategy and the repair strategy when the new deployment state was formed is revoked, and the deployment state before the formation of the new deployment state and its corresponding resource configuration state are restored to the current deployment state and the current resource configuration state, respectively. Based on the new deployment evaluation results corresponding to the new deployment status, rewards are given to the destruction and repair strategies adopted in this round. Based on the cumulative rewards and usage times of each destruction and repair strategy, the selection weight of the corresponding destruction and repair strategies in subsequent iterations is adjusted.

7. The method for optimizing the collaborative deployment of heterogeneous base stations according to claim 1, characterized in that, Step S1 further includes: obtaining the priorities of the ground users and the low-altitude users, as well as the construction cost of the candidate heterogeneous base stations; Step S4 includes: determining the coverage status of the ground user and the low-altitude user based on the service association and the resource configuration status, and determining the satisfaction status of the communication requirements based on the resource configuration status; A priority-weighted coverage index is determined based on the coverage status and priority of the ground users and the low-altitude users; a communication demand satisfaction index is determined based on the communication demand and its satisfaction status; an interference evaluation index is determined based on the interference status; and a construction cost index is determined based on the construction cost of the candidate heterogeneous base stations that have been enabled in the deployment status; the deployment evaluation result is determined based on the priority-weighted coverage index, the communication demand satisfaction index, the interference evaluation index, and the construction cost index. A Pareto non-dominated deployment state set is established based on the deployment evaluation results of each deployment state obtained through iteration; When the number of deployment states in the Pareto non-dominated deployment state set exceeds a preset upper limit, the most crowded deployment state is deleted according to the target space congestion distance of each deployment state until the number of deployment states in the Pareto non-dominated deployment state set does not exceed the preset upper limit. After the iteration stopping condition is met, the final deployment state is determined from the Pareto non-dominated deployment state set based on the deployment evaluation result.

8. The method for optimizing the collaborative deployment of heterogeneous base stations according to claim 1, characterized in that, Set minimum coverage requirements for the ground users and the low-altitude users respectively; When establishing the service association and configuring the communication resources, the coverage status of the ground user and the low-altitude user are determined respectively, and the service association is established and the communication resources are configured alternately for the ground user or the low-altitude user based on whether the ground user and the low-altitude user meet their respective minimum coverage requirements.

9. A heterogeneous base station collaborative deployment optimization system for air-ground converged communication networks, characterized in that, It includes an information acquisition module, a communication status determination module, a resource configuration module, a deployment optimization module, and an iteration control module; The information acquisition module is used to acquire spatial information of candidate heterogeneous base stations, ground users and low-altitude users, as well as the communication needs of the ground users and the low-altitude users, and to form a deployment status characterizing the base station activation status based on the candidate heterogeneous base stations. The communication status determination module is used to determine the communication quality between the candidate heterogeneous base stations and the ground user and the low-altitude user, as well as the interference status formed by multiple activated candidate heterogeneous base stations, based on the spatial information and the activated candidate heterogeneous base stations in the deployment status. The resource configuration module is used to establish service associations between the enabled candidate heterogeneous base stations and the ground users and the low-altitude users according to the communication quality, the interference status and the communication requirements, and configure communication resources according to the service associations to form a resource configuration status corresponding to the deployment status. The deployment optimization module is used to evaluate the deployment status based on the resource configuration status, obtain the deployment evaluation result, and determine whether a preset iteration stop condition is met based on the deployment evaluation result and the number of iterations; when the iteration stop condition is met, the final deployment status is determined from the deployment status obtained by iteration based on the deployment evaluation result corresponding to each deployment status; If the iteration stop condition is not met, the activation status of the candidate heterogeneous base stations is adjusted according to the deployment evaluation results to form a new deployment status; The iteration control module is configured to, after the formation of the new deployment state, control the communication state determination module to re-determine the corresponding communication quality and interference state for the new deployment state, and control the resource configuration module to re-establish service associations and configure communication resources based on the re-determined communication quality and interference state, so as to form a new resource configuration state corresponding to the new deployment state; it is also configured to control the deployment optimization module, the communication state determination module, and the resource configuration module to iterate based on the new deployment state and the new resource configuration state until the iteration stop condition is met, and determine the set of base stations to be built based on the final deployment state.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the heterogeneous base station collaborative deployment optimization method according to any one of claims 1 to 8.

Images