A method for edge base station deployment with border user prioritized coverage

Abstract

The application discloses a method for deploying an edge base station based on boundary user priority coverage, which comprises the following steps: S1, calculating the coverage of each candidate position on a user according to the spatial coordinates of the user and the candidate position; S2, initializing a candidate base station position set as all candidate base station positions and a user set to be covered as a set composed of users; S3, identifying boundary users from the user set to be covered and adding the boundary users into a boundary user set; S4, calculating the number of boundary users covered by each candidate position according to the boundary user set, and screening a candidate base station position set covering the most boundary users; S5, calculating the total number of all users covered by each position according to the candidate base station position set, and selecting a position covering the most users from the candidate base station position set; S6, obtaining all users covered by the selected position; and S7, updating a problem state variable. The application can realize comprehensive and efficient coverage of a complex area by preferentially identifying and covering boundary users with fewer base stations.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method for deploying edge base stations with priority coverage for border users. Background Technology

[0002] In recent years, deep learning technology has been widely applied in the mobile and wireless network fields, making it increasingly complex for various mobile applications to extract high-value information from user data generated by mobile devices. To effectively address the limitations of mobile devices' battery and computing power, as well as the high network latency and difficulty in guaranteeing privacy and security in centralized computing models (such as cloud computing and cluster computing), edge computing has emerged and has received widespread attention from academia and industry in recent years. Edge computing aims to extend the capabilities of mobile devices and reduce transmission latency between data sources and data processing resources by deploying computing resources close to distributed data sources. Generally, an edge computing system contains multiple geographically dispersed edge computing centers (referred to as edge nodes). Each edge node deploys a network access point, allowing mobile devices to access the node's resources or forward requests to other edge nodes or cloud computing centers to achieve edge-cloud collaboration. Since the types of network access points deployed on edge nodes are diverse, such as micro base stations (BS), WiFi, ZigBee, and Bluetooth Low Energy (BLE), they are all used to provide wireless network connections between devices and edge nodes. Given the limitations of resource costs and space, each edge node has limited processing power, so edge resources must be carefully managed to avoid low resource utilization and performance degradation.

[0003] To provide services via edge computing, the primary task is to determine the deployment strategy for edge nodes. Edge deployment mainly involves two steps: base station deployment and resource allocation (edge ​​server placement). First, service providers must select locations from numerous candidate sites to deploy base stations, aiming to maximize network coverage while minimizing the number of base stations, thereby reducing capital expenditure and ensuring all users have access to edge computing services. Second, given the base station deployment strategy, the amount of resources allocated to each base station to handle received requests is determined, optimizing service quality and resource costs.

[0004] Current research on edge deployment problems mainly falls into three categories: heuristic methods, clustering-based methods, and global optimization search methods. Heuristic methods typically iteratively select locations to deploy base stations that best cover the most users, aiming to maximize user coverage. Clustering-based methods first cluster users based on their locations and then deploy base stations at the locations closest to the cluster centers. The performance of these two methods is significantly affected by users located in sparse areas or at the edges, potentially leading to incomplete coverage or requiring excessive base station deployment to achieve full coverage, thus increasing capital expenditure and operating costs. Global optimization search methods employ stochastic global search strategies inspired by natural and social rules. While theoretically offering better solutions, for large-scale optimization problems (e.g., tens of thousands of candidate locations), the search space grows exponentially with the problem size, making it difficult to find the globally optimal solution. Furthermore, more candidate locations mean more cost-effective base station deployment options, further expanding the search space and increasing the difficulty of finding the optimal solution using global optimization search methods.

[0005] In real-world scenarios, when designing base station deployment strategies, edge users are crucial for optimizing the number of base stations while achieving maximum coverage. However, existing base station deployment methods typically consider edge users last when searching for solutions, leading to either low coverage or an excessive number of base stations deployed. Summary of the Invention

[0006] To address the aforementioned problems, the present invention aims to provide a method for deploying edge base stations with priority coverage for border users.

[0007] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: A method for deploying edge base stations with priority coverage for border users includes the following steps: S1. Based on the spatial coordinates of the user and the candidate locations, calculate the coverage of the user by each candidate location. The calculation formula is as follows: (1) (2) (3) in, To calculate the first The user and the An intermediate variable representing the distance between candidate locations; , They represent the first Latitude and longitude of each user; , They represent the first The latitude and longitude of each candidate location; Indicates the first The user and the The distance between candidate positions; Indicates the first Is the first user...? The indicator variables covered by each candidate location; Indicates the maximum coverage area of ​​the base station; Represents the arctangent mathematical function; S2. Initialize the candidate base station location set For all candidate base station locations ,in, Indicates the first One candidate position Indicates the number of candidate base station locations; initializes the set of users to be covered. A collection of users ,in, Indicates the first One user, Indicates the number of users; S3, from the set of users to be covered Identify border users and add them to the border user set. The process includes the following sub-steps: S3.1 Initialize the boundary user set An empty, non-boundary set of users Empty; for the set of users to be covered Each user in Convert its latitude and longitude to ordinates on a plane using the following formula: (4) (5) in, , They represent the first The x-coordinate and y-coordinate values ​​of each user on a plane coordinate system; S3.2, From the set of users to be covered Find the user with the smallest x-coordinate value. If the following formula is satisfied: (6) Add it to the border user set as the first border user. ,Right now ; S3.3 Add eligible edge users to the boundary user set. This continues until the boundary user set and the non-boundary user set contain all users to be covered, i.e.: The steps are as follows: S3.31 Calculate the set of users to be covered Non-boundary user set Boundary User Set Each node to the user The distances are sorted from closest to furthest, as follows: (7) (8) in, Indicates user With users The distance between them; For users The planar coordinates of the location; Indicates user The planar coordinates of the location; Indicates user With users The distance between them; Represents the set of users to be covered Non-boundary user set Boundary User Set Mid-range users No. Nearest users; , They represent the sets of users to be covered. Non-boundary user set The number of users included. Represents the set of users to be covered Non-boundary user set Boundary User Set Number of users included; initialization =1; S3.32, Calculate User With users The equation of the straight line whose position is determined is expressed as: As shown in the following formula: (9) in, Indicates user The planar coordinates of the location; Indicates user The planar coordinates of the location; S3.33, Calculation and Lines Vertical and through and The two straight lines representing the positions are respectively: and As shown in the following formula: (10) (11) S3.34, Find the set of users to be covered. Non-boundary user set Boundary User Set The middle is located on the straight line and Users between them are shown in the following formula: (12) in, Indicates that it lies on a straight line and The set of users between; Represents a straight line and Located in the user Both sides of the position; S3.4 Determine if it lies on a straight line and The set of users between Are all user locations in the middle a straight line? On the same side, that is, whether the proposition expressed by formula (12) is true; expressed as: (13) If the proposition is true, the user At the boundary, the user Join the boundary user set middle: , will users Replace with user : And execute step S3.31; otherwise, the user Join the non-boundary user set middle, Increment by 1 and proceed to step S3.32; S4. Based on the boundary user set obtained in step S3 For each candidate location, the number of boundary users it can cover is calculated, and a set of candidate base station locations that can cover the most boundary users is obtained. S5. Based on the candidate base station location set obtained in step S4, calculate the total number of users that each location can cover, and select the location with the most covered users. S6. Obtain all users covered by the location selected in step S5; if this location can cover at least one user, determine to deploy a base station at this location. S7. Update the problem status variables: Remove the selected location from the candidate location set in step S6; add the user to be covered to the base station covered user set; remove the user from the uncovered user set.

[0008] Furthermore, in step S4 above, if the candidate base station location set contains only one candidate location, then this candidate location is directly selected.

[0009] Furthermore, in step S5 above, if there are multiple candidate base station locations with the same number of users covered, one of them is randomly selected.

[0010] Due to the adoption of the technical solution described above, the present invention has the following advantages: The edge base station deployment method for prioritizing coverage of boundary users in this invention employs a boundary user identification method that performs objective boundary identification based on the geometric relationship of user geographical locations, making it mathematically more rigorous. Furthermore, this method is a deterministic algorithm, where a given input corresponds to a unique output, eliminating randomness and offering higher reliability and debuggability in practical engineering deployments, surpassing algorithms such as alpha-shape that rely on parameter settings. The boundary user identification method can identify a more comprehensive set of boundary users than traditional convex hull algorithms, especially including boundary users distributed in locally concave or non-convex regions. This provides more complete and detailed boundary information for subsequent base station deployment optimization, significantly improving the overall algorithm's convergence efficiency and coverage quality.

[0011] The edge base station deployment method for prioritizing coverage of boundary users in this invention can achieve comprehensive and efficient coverage of complex areas with fewer base stations by prioritizing the identification and coverage of boundary users. It cleverly addresses the problem of uneven user distribution (dense in the center and sparse at the edge) in real-world scenarios and avoids coverage gaps or base station redundancy caused by traditional greedy methods or clustering methods that process sparse peripheral users last.

[0012] This invention discloses an edge base station deployment method that prioritizes coverage for border users. This method can more accurately identify "border users" in geographically complex areas and assess the actual coverage capability of base stations, thereby avoiding misjudgments caused by terrain factors during deployment decisions. Ultimately, it achieves 100% coverage for all users with fewer base stations, effectively solving the problems of low base station deployment efficiency and incomplete coverage in geographically challenging areas. It is suitable for providing edge computing services and 5G network coverage in remote or geographically complex areas (such as mountains, islands, vast farmlands, or desert edges). It adopts mature Geographic Information System (GIS) technology and wireless propagation models, and has high rationality and practicality. Attached Figure Description

[0013] Figure 1This is a flowchart of the edge base station deployment method for prioritizing coverage of boundary users according to the present invention; Figure 2 This is a schematic diagram of an embodiment of a base station deployment scenario; where A, B, and C represent three dispersed candidate locations for base station deployment; D represents 10 users who need to be covered by the deployed base station; and E represents the coverage area of ​​the central base station. Figure 3 The edge base station deployment method based on the present invention prioritizes coverage for edge users. Figure 2 The diagram illustrates the deployment of base stations in the scenario shown; where F represents a boundary user, and there are a total of 5 users. Figure 4 This is a diagram illustrating the status of mobile phones accessing the Internet via base stations in Telecom's data collection. Figure 5 This is a user location distribution map in the EUA dataset; Figure 6a This is a comparison chart of the base station deployment volume of the present invention and various algorithms in the prior art on the EUA dataset; Figure 6b This is a comparison chart of base station deployment volume for eight existing global optimization algorithms on the EUA dataset; Figure 7a This is a comparison chart of the base station deployment volume of this invention and various algorithms in the prior art on the Telcom dataset; Figure 7b This is a comparison chart of base station deployment volume using various global optimization algorithms in existing technologies on the Telcom dataset. Detailed Implementation

[0014] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can fully understand and implement the present invention.

[0015] like Figure 1 As shown, a method for deploying edge base stations with priority coverage for edge users includes the following steps: S1. Based on the spatial coordinates of the user and the candidate locations, calculate the coverage of the user by each candidate location. The calculation formula is as follows: (1) (2) (3) in, To calculate the first The user and the An intermediate variable representing the distance between candidate locations; , They represent the first Latitude and longitude of each user; , They represent the first The latitude and longitude of each candidate location; Indicates the first The user and the The distance between candidate positions; Indicates the first Is the first user...? The indicator variables covered by each candidate location; Indicates the maximum coverage area of ​​the base station; Represents the arctangent mathematical function; S2. Initialize the candidate base station location set For all candidate base station locations ,in, Indicates the first One candidate position Indicates the number of candidate base station locations; initializes the set of users to be covered. A collection of users ,in, Indicates the first One user, Indicates the number of users; S3, from the set of users to be covered Identify border users and add them to the border user set. The process includes the following sub-steps: S3.1 Initialize the boundary user set An empty, non-boundary set of users Empty; for the set of users to be covered Each user in Convert its latitude and longitude to ordinates on a plane using the following formula: (4) (5) in, , They represent the first The x-coordinate and y-coordinate values ​​of each user on a plane coordinate system; , They represent the first Latitude and longitude of each user; S3.2, From the set of users to be covered Find the user with the smallest x-coordinate value. If the following formula is satisfied: (6) Add it to the border user set as the first border user. ,Right now ; S3.3 Add eligible edge users to the boundary user set. This continues until the boundary user set and the non-boundary user set contain all users to be covered, i.e.: The steps are as follows: S3.31 Calculate the set of users to be covered Non-boundary user set Boundary User Set Each node to the user The distances are sorted from closest to furthest, as follows: (7) (8) in, Indicates user With users The distance between them; For users The planar coordinates of the location; Indicates user The planar coordinates of the location; Indicates user With users The distance between them; Represents the set of users to be covered Non-boundary user set Boundary User Set Mid-range users No. Nearest users; , They represent the sets of users to be covered. Non-boundary user set The number of users included, therefore Represents the set of users to be covered Non-boundary user set Boundary User Set Number of users included; initialization =1; S3.32, Calculate User With users The equation of the straight line whose position is determined is expressed as: As shown in the following formula: (9) in, Indicates user The planar coordinates of the location; Indicates user The planar coordinates of the location; S3.33, Calculation and Lines Vertical and through and The two straight lines representing the positions are respectively: and As shown in the following formula: (10) (11) S3.34, Find the set of users to be covered. Non-boundary user set Boundary User Set The middle is located on the straight line and Users between them are shown in the following formula: (12) in, Indicates that it lies on a straight line and The set of users between; Represents a straight line and Located in the user Both sides of the position; S3.4 Determine if it lies on a straight line and The set of users between Are all user locations in the middle a straight line? On the same side, that is, whether the proposition expressed by formula (12) is true; expressed as: (13) If the proposition is true, the user At the boundary, the user Join the boundary user set middle: , will users Replace with user : And execute step S3.31; otherwise, the user Join the non-boundary user set middle, Increment by 1 and proceed to step S3.32; S4. Based on the set of boundary users obtained in step S3, for each candidate location, calculate the number of boundary users it can cover, and filter to obtain a set of candidate base station locations that can cover the most boundary users; if the set of candidate base station locations contains only one candidate location, directly select this candidate location and proceed to step S6. S5. Based on the obtained set of candidate base station locations, calculate the total number of users that each location can cover, and select the location with the most covered users; if there are multiple candidate base station locations with the same number of covered users, randomly select one of them. S6. Obtain all users covered by the location selected in step S4 or step S5; if this location can cover at least one user, determine to deploy a base station at this location. S7. Update the problem status variables: Remove the selected location from the candidate location set in step S6; add the user to be covered to the base station covered user set; remove the user from the uncovered user set.

[0016] against Figure 2 Base station deployment scenarios, such as Figure 3 As shown, the method of the present invention first identifies five boundary users, and then selects candidate base station deployment location C as the first base station deployment point. This is because candidate base station deployment location C covers three boundary users, while candidate base station deployment locations A and B each cover only two, and users already covered by candidate base station deployment location C are removed. At this point, the base station deployment optimization problem is simplified to a smaller-scale problem involving only two candidate locations and five users. In this simplified problem, there are four boundary users, of which all users can be covered by candidate base station deployment location A, and three users can be covered by candidate base station deployment location B. Therefore, candidate base station deployment location A is selected as the second base station deployment point, thereby obtaining the optimal solution.

[0017] The effectiveness of the edge base station deployment method for prioritizing coverage of boundary users in this invention will be verified below.

[0018] like Figure 4 , 5 As shown, a comparative experiment was conducted based on two real datasets: the EUA dataset and the Telecom dataset. The EUA dataset contains geolocation information of base stations and users in Australia, including 4748 user locations and 95562 base station locations. The Telecom dataset, provided by Shanghai Telecom, contains over 7.2 million request records from 9481 mobile phones through 3233 base stations, with each record including the location of the base station receiving the request. This experiment used data from the first half of June in this dataset, which contains 563914 records (records with empty base station locations have been excluded); from... Figure 4The results show the density of mobile phone access to base stations. Base stations in city centers have high loads, frequent user connections, and densely distributed requests; while base station coverage in suburban areas is sparse, user activity is less, and requests are sparsely distributed. Since the Telecom dataset does not contain the specific locations of user requests, this experiment treats each record as a single user record and randomly generates the user's location within a 50-meter radius of the base station in that record. The coverage area of ​​each base station is set as a circular area with a radius of 100 meters.

[0019] The edge base station deployment method prioritizing coverage of boundary users in this invention is compared with three types of algorithms in the prior art. The performance indicators used in the experiment include: user coverage rate (the higher the better, 100% is optimal) and the number of base stations deployed (the fewer the better).

[0020] (1) Two heuristic algorithms: Greedy algorithm repeatedly selects candidate locations that can cover the most users until there are no candidate locations or no uncovered users; The applicant's prior patent number 202410090170.7, invention title "A heuristic edge base station deployment method with minimum coverage users first", adopts the technical solution of iteratively selecting base stations that can cover users covered by the fewest candidate sites (Fewest Covered User First, FCUF).

[0021] (2) Eight global optimization algorithms: including five representative and widely used metaheuristic algorithms: GA (genetic algorithm), PSO (particle swarm optimization algorithm), DE (differential evolution algorithm), MVO (multiverse optimization algorithm) and GWO (grey wolf optimization algorithm); in the experiment, all of the above single global optimization algorithms were implemented through the MEALPY public library and used their default parameters.

[0022] The other three hybrid metaheuristic algorithms include: PSOM (Particle Swarm Optimization with Mutation), GAPSO, and PSOGA. The PSOM algorithm performs a mutation operation on each individual at the end of each iteration. The GAPSO algorithm uses GA and PSO sequentially to evolve the population. The PSOGA algorithm uses PSO and GA sequentially to evolve the population.

[0023] (3) Two clustering-based algorithms: The KMeansBS algorithm first uses K-means++ to cluster all users, where the number of clusters equals the number of candidate locations; then, the FF algorithm is used to select base station deployment points from the candidate locations closest to each cluster center. The KMeansBound algorithm iteratively increases the number of clusters starting from k=2, using K-means++ to cluster users until the distance between each user and its cluster center does not exceed the coverage radius of the base station; then, the base station deployment point is selected from the candidate locations closest to the cluster center.

[0024] Regarding user coverage, for the EUA dataset, the KMeansBound algorithm achieved 99.98%; for the Telecom dataset, the GA, GAPSO, ABC, PSO, and GWO algorithms achieved 99.992%–99.997%, with all other algorithms reaching 100% coverage in the remaining cases. The results for the number of base stations deployed are as follows: Figure 6a , 6b and Figure 7a , 7b As shown, the edge base station deployment method for prioritizing coverage of boundary users according to the present invention requires the fewest number of base stations to be deployed.

[0025] Experiments show that the edge base station deployment method for prioritizing coverage of edge users in this invention can achieve full user coverage and significantly reduce the number of base stations required to be deployed on publicly available real datasets through a more efficient search process, thereby effectively reducing network construction and operation costs while ensuring service quality.

[0026] The scope of protection of this invention is not limited to the specific embodiments described above. Any changes made based on the core principles of this invention, including but not limited to equivalent substitutions of technical solutions and structural improvements, should be considered to fall within the scope of protection of the claims of this invention. Various modifications and adjustments made to the implementation schemes by those skilled in the art without departing from the design concept of this invention also fall within the scope of protection of this invention.

Claims

1. A method for deploying edge base stations with priority coverage for edge users, characterized by: It includes the following steps: S1. Based on the spatial coordinates of the user and the candidate locations, calculate the coverage of the user by each candidate location. The calculation formula is as follows: (1) (2) (3) in, To calculate the first The user and the first An intermediate variable representing the distance between candidate locations; , They represent the first Latitude and longitude of each user; , They represent the first The latitude and longitude of each candidate location; Indicates the first The user and the first The distance between candidate positions; Indicates the first Is the first user...? The indicator variables covered by each candidate location; Indicates the maximum coverage area of ​​the base station; Represents the arctangent mathematical function; S2. Initialize the candidate base station location set For all candidate base station locations ,in, Indicates the first Candidate positions, Indicates the number of candidate base station locations; initializes the set of users to be covered. A collection of users ,in, Indicates the first One user, Indicates the number of users; S3, from the set of users to be covered Identify border users and add them to the border user set. The process includes the following sub-steps: S3.1 Initialize the boundary user set An empty, non-boundary set of users Empty; for the set of users to be covered Each user in Convert its latitude and longitude to ordinates on a plane using the following formula: (4) (5) in, , They represent the first The x-coordinate and y-coordinate values ​​of each user on a plane coordinate system; S3.2, From the set of users to be covered Find the user with the smallest x-coordinate value. If the following formula is satisfied: (6) Add it to the border user set as the first border user. ,Right now ; S3.3 Add eligible edge users to the boundary user set. This continues until the boundary user set and the non-boundary user set contain all users to be covered, i.e.: The steps are as follows: S3.31 Calculate the set of users to be covered Non-boundary user set Boundary User Set Each node to the user The distances are sorted from closest to furthest, as follows: (7) (8) in, Indicates user With users The distance between them; For users The planar coordinates of the location; Indicates user The planar coordinates of the location; Indicates user With users The positional distance between them; Represents the set of users to be covered Non-boundary user set Boundary User Set Mid-range users No. Nearest users; , These represent the sets of users to be covered. Non-boundary user set The number of users included. Represents the set of users to be covered Non-boundary user set Boundary User Set Number of users included; initialization =1; S3.32, Calculate User With users The equation of the straight line whose position is determined is expressed as: As shown in the following formula: (9) in, Indicates user The planar coordinates of the location; Indicates user The planar coordinates of the location; S3.33, Calculation and Lines Vertical and through and The two straight lines representing the positions are respectively: and As shown in the following formula: (10) (11) S3.34, Find the set of users to be covered. Non-boundary user set Boundary User Set The middle is located on the straight line and Users between them are shown in the following formula: （12） in, Indicates that it lies on a straight line and The set of users between; Represents a straight line and Located in the user Both sides of the position; S3.4 Determine if it lies on a straight line and The set of users between Are all user locations in the middle a straight line? On the same side, that is, whether the proposition expressed by formula (12) is true; expressed as: （13） If the proposition is true, the user At the boundary, the user Join the boundary user set middle: , will users Replace with user : And execute step S3.31; otherwise, the user Join the non-boundary user set middle, Increment by 1 and proceed to step S3.32; S4. Based on the boundary user set obtained in step S3 For each candidate location, the number of boundary users it covers is calculated, and the candidate base station locations that cover the most boundary users are selected. S5. Based on the candidate base station location set obtained in step S4, calculate the total number of all users covered by each location, and select the location with the most covered users. S6. Obtain all users covered by the location selected in step S5; if this location covers at least one user, determine to deploy a base station at this location; S7. Update the problem status variables: Remove the selected location from the candidate location set in step S6; add the user to be covered to the base station covered user set; remove the user from the uncovered user set.

2. The edge base station deployment method for priority coverage of edge users according to claim 1, characterized in that: In step S4, if the candidate base station location set contains only one candidate location, then this candidate location is selected directly.

3. The edge base station deployment method for priority coverage of edge users according to claim 1, characterized in that: In step S5, if there are multiple candidate base station locations with the same number of users covered, one of them is randomly selected.

Images