Base station site selection method and apparatus, device, and storage medium

Abstract

A base station site selection method, device, equipment and storage medium are disclosed, which belong to the field of communication technology. The method comprises: modeling a target area to obtain a three-dimensional model of the target area, the target area comprising a plurality of sampling points and a plurality of positions for placing base stations; using a genetic algorithm to determine the positions of k base stations under the condition that the coverage rate of the k base stations is a first coverage rate, each chromosome in the genetic algorithm being used to indicate that the k base stations are placed at k different positions, the k different positions being any k positions in the plurality of positions, the coverage rate being used to indicate the proportion of sampling points with received power greater than or equal to a first power threshold in the plurality of sampling points, the first coverage rate being the maximum coverage rate of the k base stations in the target area, and the three-dimensional model being used to determine the received power of each sampling point. The method can improve the accuracy of the determined received power and simplify the base station site selection process.

Description

Technical Field

[0001] This disclosure relates to the field of communication technology, and in particular to a base station site selection method, apparatus, device and storage medium. Background Technology

[0002] A cell is an area covered by a base station or a portion of a base station (fan antenna) in a cellular mobile communication system, within which mobile stations can reliably communicate with the base station via a wireless channel. With technological advancements, the radius of cells has continuously decreased, leading to the emergence of microcells and picocells. For microcell or picocell systems, base station site selection is crucial.

[0003] In related technologies, base station site selection methods include: determining multiple base station candidate points and sampling points; calculating the received power of each sampling point when the base station is located at each base station candidate point to determine the coverage rate when the base station is located at each base station candidate point, and selecting the base station candidate point with the highest coverage rate as the base station site.

[0004] However, when using the above-mentioned base station location method, the received power of the calculated sampling points is not accurate enough, which leads to inaccurate coverage. Furthermore, the coverage needs to be calculated once for each base station candidate point. When there are many base stations, base station candidate points, and sampling points, the amount of calculation is large, which makes the base station location process more complicated. Summary of the Invention

[0005] This disclosure provides a base station site selection method, apparatus, device, and storage medium, which can improve the accuracy of the determined received power and simplify the base station site selection process. The technical solution includes at least the following:

[0006] In a first aspect, a base station site selection method is provided, comprising: modeling a target area to obtain a three-dimensional model of the target area, the target area including multiple sampling points and multiple locations for placing base stations; employing a genetic algorithm to determine the locations of k base stations when the coverage rate of k base stations is a first coverage rate, wherein each chromosome in the genetic algorithm is used to indicate that the k base stations are placed in k different locations, the k different locations being any k locations among the multiple locations, the coverage rate being used to indicate the proportion of sampling points with received power greater than or equal to a first power threshold among the multiple sampling points, the first coverage rate being the maximum coverage rate of the k base stations in the target area, and the three-dimensional model being used to determine the received power of each sampling point; wherein k is a positive integer, and k is greater than 0.

[0007] Optionally, when using a genetic algorithm to determine the coverage rate of k base stations as a first coverage rate, the placement of the k base stations includes: determining the first received power of each sampling point when the k base stations are placed at the location indicated by the first chromosome based on the three-dimensional model, the line-of-sight probability model, and the ray tracing algorithm, where the first chromosome is any chromosome in any iteration; determining the coverage rate of the first chromosome based on the first received power of each sampling point and the power threshold, where the coverage rate of the first chromosome is the fitness value of the first chromosome.

[0008] Optionally, determining the first received power of each sampling point when k base stations are placed at the positions indicated by the first chromosome, based on the three-dimensional model, the line-of-sight probability model, and the ray tracing algorithm, includes: determining the type of line-of-sight path from each base station to each sampling point based on the three-dimensional model and the line-of-sight probability model, wherein the type of line-of-sight path includes direct line-of-sight path, reflected line-of-sight path, and diffracted line-of-sight path; determining the line-of-sight path from each base station to each sampling point based on the ray tracing algorithm, the type of line-of-sight path, and the three-dimensional model; determining the path loss of each line-of-sight path; and determining the first received power of each sampling point based on the path loss of each line-of-sight path.

[0009] Optionally, when using a genetic algorithm to determine the coverage rate of k base stations as a first coverage rate, the placement of the k base stations further includes: generating a fourth chromosome based on the second and third chromosomes for any crossover in any iteration, wherein the second and third chromosomes are two different chromosomes generated in the previous iteration of any iteration, and the placement of the k base stations indicated by the fourth chromosome is the average of the placement of the k base stations indicated by the second and third chromosomes.

[0010] Optionally, each base station has the same transmission power, and the sum of the transmission powers of the k base stations is a second power threshold.

[0011] Optionally, the method further includes: when the first coverage rate is less than a coverage rate threshold, based on a genetic algorithm, determining the coverage rate of k+1 base stations as a second coverage rate, the location of the k+1 base stations, the second coverage rate being the maximum coverage rate of the k+1 base stations in the target area, and the sum of the transmission powers of the k+1 base stations being the second power threshold.

[0012] Optionally, the method further includes: when the first coverage rate is greater than a coverage rate threshold, reducing the transmission power of multiple base stations so that the sum of the transmission power of the multiple base stations is less than a second power threshold; based on the reduced transmission power of the multiple base stations, using a genetic algorithm to determine the location of the multiple base stations when the coverage rate of the multiple base stations is a third coverage rate, where the third coverage rate is the maximum coverage rate after the transmission power of the multiple base stations is reduced; if the third coverage rate is equal to the coverage rate threshold, using the location of the multiple base stations when the coverage rate of the multiple base stations is the third coverage rate as the site selection of the multiple base stations.

[0013] Optionally, modeling the target region to obtain a three-dimensional model of the target region includes: acquiring a first set of aerial images, the first set of aerial images including multiple aerial images of the target region; determining first point cloud data of the target region based on the multiple aerial images; performing semantic segmentation on the first point cloud data; performing incremental reconstruction on the semantically segmented first point cloud data to obtain second point cloud data; and performing surface reconstruction based on the second point cloud data to obtain the three-dimensional model.

[0014] Secondly, a base station location device is also provided, comprising: a modeling module for modeling a target area to obtain a three-dimensional model of the target area, the target area including multiple sampling points and multiple locations for placing base stations; a genetic algorithm module for using a genetic algorithm to determine the locations of the k base stations when the coverage rate of the k base stations is a first coverage rate, wherein each chromosome in the genetic algorithm is used to indicate that the k base stations are placed in k different locations, the k different locations being any k locations among the multiple locations, the coverage rate indicating the proportion of sampling points with received power greater than or equal to a first power threshold among the multiple sampling points, the first coverage rate being the maximum coverage rate of the k base stations in the target area, and the three-dimensional model for determining the received power of each sampling point; wherein k is a positive integer, and k is greater than 0.

[0015] Optionally, the genetic algorithm module is further configured to determine, based on the three-dimensional model, the line-of-sight probability model, and the ray tracing algorithm, the first received power of each sampling point when k base stations are placed at the positions indicated by the first chromosome, wherein the first chromosome is any chromosome in any iteration; and to determine the coverage of the first chromosome based on the magnitude of the first received power of each sampling point and the power threshold, wherein the coverage of the first chromosome is the fitness value of the first chromosome.

[0016] Optionally, the genetic algorithm module is further configured to determine the type of line-of-sight path from each base station to each sampling point based on the three-dimensional model and the line-of-sight probability model, wherein the type of line-of-sight path includes direct line-of-sight path, reflected line-of-sight path, and diffracted line-of-sight path; determine the line-of-sight path from each base station to each sampling point based on the ray tracing algorithm, the type of line-of-sight path, and the three-dimensional model; determine the path loss of each line-of-sight path; and determine the first received power of each sampling point based on the path loss of each line-of-sight path.

[0017] Optionally, the genetic algorithm module is further used to generate a fourth chromosome based on the second and third chromosomes for crossover in any iteration, wherein the second and third chromosomes are two different chromosomes generated in the previous iteration, and the positions of the k base stations indicated by the fourth chromosome are the average of the positions of the k base stations indicated by the second and third chromosomes.

[0018] Optionally, the genetic algorithm module is further configured to, when the first coverage rate is less than the coverage rate threshold, determine the location of the k+1 base stations based on the genetic algorithm, where the coverage rate of the k+1 base stations is the second coverage rate, the second coverage rate is the maximum coverage rate of the k+1 base stations in the target area, and the sum of the transmission power of the k+1 base stations is the second power threshold.

[0019] Optionally, the genetic algorithm module is further configured to, when the first coverage rate is greater than a coverage rate threshold, reduce the transmission power of multiple base stations so that the sum of the transmission power of the multiple base stations is less than the second power threshold; based on the reduced transmission power of the multiple base stations, use a genetic algorithm to determine the location of the multiple base stations when the coverage rate of the multiple base stations is a third coverage rate, where the third coverage rate is the maximum coverage rate after the transmission power of the multiple base stations is reduced; if the third coverage rate is equal to the coverage rate threshold, the location of the multiple base stations when the coverage rate of the multiple base stations is the third coverage rate is used as the site selection of the multiple base stations.

[0020] Optionally, the modeling module is further configured to acquire a first set of aerial images, the first set of aerial images including multiple aerial images of the target area; determine first point cloud data of the target area based on the multiple aerial images; perform semantic segmentation on the first point cloud data; perform incremental reconstruction on the semantically segmented first point cloud data to obtain second point cloud data; and perform surface reconstruction based on the second point cloud data to obtain the three-dimensional model.

[0021] Thirdly, a computer device is also provided, comprising: a memory and a processor, wherein the memory stores at least one computer program, the at least one computer program being loaded and executed by the processor to perform the base station location method described in the above embodiments.

[0022] Fourthly, a computer-readable storage medium is also provided, wherein at least one computer program is stored in the computer-readable storage medium, the at least one computer program being loaded and executed by a processor to perform the base station location method described in the above embodiments.

[0023] Fifthly, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the method described in the first aspect.

[0024] The beneficial effects of the technical solutions provided in this disclosure include at least the following:

[0025] In this embodiment, the received power of each sampling point is calculated using a three-dimensional model of the target area. The impact of building obstruction on the intensity of the received power at each sampling point is considered, thereby improving the accuracy of the calculated received power and coverage rate. Furthermore, since a genetic algorithm is used for base station site selection, a typical random algorithm, it eliminates the need to calculate coverage for each location, greatly simplifying the base station site selection process. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 A flowchart illustrating a base station location method provided in an exemplary embodiment of this disclosure is shown.

[0028] Figure 2 A flowchart of a base station location method provided in another exemplary embodiment of this disclosure is shown;

[0029] Figure 3 A flowchart for modeling the target region;

[0030] Figure 4 This is a schematic diagram of the line-of-sight probability model;

[0031] Figure 5 This is a schematic diagram illustrating the types of line-of-sight paths;

[0032] Figure 6 This illustration shows a schematic diagram of the structure of a base station addressing device provided in an exemplary embodiment of the present disclosure;

[0033] Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this disclosure. Detailed Implementation

[0034] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this disclosure pertains. The terms “first,” “second,” “third,” and similar terms used in this patent application specification and claims do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms “an” or “a” and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms “comprising” or “including” and similar terms mean that the elements or objects preceding “comprising” or “including” encompass the elements or objects listed following “comprising” or “including” and their equivalents, but do not exclude other elements or objects.

[0035] To make the objectives, technical solutions, and advantages of this disclosure clearer, the embodiments of this disclosure will be described in further detail below with reference to the accompanying drawings.

[0036] To facilitate understanding of the embodiments of this disclosure, microcells and picocells will be briefly described below.

[0037] Currently, due to business needs, microcell technology has been proposed for areas not covered by traditional cellular networks. Compared to microcells, regular cellular networks are called macrocells. Microcells have lower transmission power and smaller coverage areas, and are generally used in areas not covered by macrocells, such as basements and tunnels, as a supplement to wireless coverage. Based on the same principle, there is also picocell technology, which has even lower power and is typically used to cover communication "hotspots" such as conference rooms.

[0038] Macrocells do not require consideration of base station site selection and planning because macrocell base stations are typically located at high elevations, allowing their emitted wireless signals to travel from high to low without obstruction. In contrast, microcells and picocells are usually located at lower elevations, such as in the middle of a city, meaning their base station antennas may be lower than surrounding buildings. Building obstruction significantly hinders wireless signal transmission, resulting in a substantial reduction in received power at sampling points blocked by buildings.

[0039] The base station site selection methods in related technologies do not consider the impact of building obstruction on the intensity of the received power at the sampling points, resulting in inaccurate calculations of the received power at the sampling points and consequently, inaccurate calculations of the coverage rate.

[0040] Figure 1 A flowchart illustrating a base station location method provided in an exemplary embodiment of this disclosure is shown. See also: Figure 1 The method includes:

[0041] In step 101, the target region is modeled to obtain a three-dimensional model of the target region.

[0042] The target area includes multiple sampling points and multiple locations for placing base stations.

[0043] The target area refers to the region where microcell or picocell base station site selection is required. The location of the target area can be set according to requirements, and this embodiment does not limit this.

[0044] In the multiple locations used to place base stations, each location can be represented by three-dimensional coordinates (x, y, z), where x is the x-axis coordinate of a location in the world coordinate system, y is the y-axis coordinate of a location in the world coordinate system, and z is the z-axis coordinate of a location in the world coordinate system.

[0045] Optionally, multiple sampling points and multiple locations for placing base stations can be set according to actual needs, and this disclosure does not limit this.

[0046] For example, the multiple locations used to place the base station can be all locations in the target area except for locations where the base station cannot be placed.

[0047] In step 102, a genetic algorithm is used to determine the locations of the k base stations when the coverage rate of the k base stations is the first coverage rate.

[0048] In the genetic algorithm, each chromosome indicates that k base stations are placed in k distinct locations. These k distinct locations are any k locations from a set of multiple locations. Coverage rate indicates the proportion of sampling points with received power greater than or equal to a first power threshold among the multiple sampling points. The first coverage rate is the maximum coverage rate of the k base stations in the target area. The 3D model is used to determine the received power of each sampling point. k is a positive integer, greater than 0.

[0049] Optionally, in the initial case, such as when planning the location of a base station for the first time in the target area, k equals 1.

[0050] Optionally, each base station has the same transmission power, and the sum of the transmission powers of the k base stations is the second power threshold.

[0051] In this embodiment of the disclosure, when k equals 1, there is only one base station, and the transmission power of that base station is the second power threshold.

[0052] When k is greater than or equal to 2, there exist k base stations, each with the same transmission power, and the sum of the transmission powers of the k base stations equals the second power threshold, which is the transmission power of each base station. Here, R max This is the second power threshold.

[0053] In the genetic algorithm, each chromosome indicates the placement of k base stations at k different locations. That is, each chromosome contains the three-dimensional coordinates of k locations, and each of the k locations on each chromosome is used to place one base station. Since the transmission power of each base station is the same when k is greater than or equal to 2, they can be considered identical. Therefore, when encoding or decoding chromosomes, the order of the k locations does not need to be considered. This embodiment does not limit the specific chromosome encoding rules, as long as each chromosome can indicate k different locations.

[0054] If the received power of a sampling point is greater than or equal to the first power threshold, it indicates that the received power of that sampling point is relatively strong, and therefore that the sampling point can receive the wireless signal emitted by the base station well. If the received power of a sampling point is less than the first power threshold, it indicates that the received power of that sampling point is relatively weak, that is, a bad point. The coverage rate is the proportion of sampling points with received power greater than or equal to the first power threshold among all sampling points, which can be expressed by formula (1).

[0055]

[0056] In formula (1), f t Where N is the coverage rate, N is the total number of sampling points, and n is the number of bad pixels.

[0057] In this embodiment of the disclosure, the objective of the genetic algorithm is to maximize the coverage rate. Therefore, the first coverage rate is the coverage rate when the k base stations are located in the k locations indicated by the optimal solution of the genetic algorithm.

[0058] In this embodiment, the received power of each sampling point is calculated using a three-dimensional model of the target area. The impact of building obstruction on the intensity of the received power at each sampling point is considered, thereby improving the accuracy of the calculated received power and coverage rate. Furthermore, since a genetic algorithm is used for base station site selection, a typical random algorithm, it eliminates the need to calculate coverage for each location, greatly simplifying the base station site selection process.

[0059] Figure 2 A flowchart illustrating a base station location method provided in an exemplary embodiment of this disclosure is shown. This method can be executed by a computer device. See also... Figure 2 The method includes:

[0060] In step 201, the target region is modeled to obtain a three-dimensional model of the target region.

[0061] For details regarding the target area, please refer to step 101 above, which will be omitted here.

[0062] Figure 3 A flowchart for modeling the target region, such as Figure 3 As shown, optionally, step 201 includes the following step ae:

[0063] Step a: Obtain the first set of aerial images.

[0064] The first aerial image set includes multiple aerial images of the target area. For example, the first aerial image set can be obtained by taking aerial photographs of the target area.

[0065] Step b: Based on multiple aerial images, determine the first point cloud data of the target area.

[0066] Optionally, the first point cloud data of the target area can be determined based on multiple aerial images using Sfm (Structure from Motion).

[0067] Sfm is a traditional 3D reconstruction method that performs 3D reconstruction based on images from multiple perspectives. It infers 3D information from multiple 2D images over a time series. The first point cloud data obtained using Sfm is sparse point cloud data. Sparse point cloud data has a small number of points and large spacing between points, resulting in insufficient accuracy. Therefore, 3D modeling cannot be directly performed using sparse point cloud data; further processing is required.

[0068] There are many ways to implement Sfm, so we will omit the details here.

[0069] Step c: Perform semantic segmentation on the first point cloud data.

[0070] After obtaining the first point cloud data, it's impossible to determine which point cloud data belong to the same building or object. Therefore, semantic segmentation is required. This step is equivalent to classifying the point cloud data and assigning labels based on the classification. Point cloud data of the same category share the same label. Point cloud data of the same category are those used to describe the same building or object.

[0071] There are many technologies available for implementing semantic segmentation of the first point cloud data, which will not be detailed here.

[0072] Step d involves incrementally reconstructing the first point cloud data after semantic segmentation to obtain the second point cloud data.

[0073] Incremental reconstruction of the first point cloud data is essentially an incremental process of processing sparse point cloud data. The resulting second point cloud data is dense point cloud data, which can then be used for 3D modeling.

[0074] There are many methods for implementing incremental reconstruction, which will not be detailed here.

[0075] Step e: Surface reconstruction is performed based on the second point cloud data to obtain a three-dimensional model.

[0076] Optionally, step e includes: using a planar extraction method to reconstruct the surface based on the second point cloud data, thereby obtaining a three-dimensional model, which is a triangular mesh model.

[0077] Optionally, step e further includes: simplifying the triangular mesh model, for example, by using a quadratic edge collapse extraction algorithm (also known as an edge collapse algorithm based on quadratic metric error). Since the 3D model generated from dense point cloud data is quite complex and contains many triangular meshes, potentially leading to a large computational load, the triangular mesh model can be simplified, and the simplified triangular mesh model can be used as the 3D model of the target region.

[0078] There are many implementation methods for plane extraction algorithms and quadratic edge collapse extraction algorithms, which will not be detailed here.

[0079] In this embodiment of the disclosure, the three-dimensional model of the target area can be obtained relatively simply and accurately through the above-described step ae.

[0080] In step 202, a genetic algorithm is used to determine the locations of the k base stations when the coverage rate of the k base stations is the first coverage rate.

[0081] When using a genetic algorithm to determine the coverage rate of k base stations as the first coverage rate, the placement of the k base stations requires multiple iterations.

[0082] Before the first iteration of a multi-iteration process, chromosome coding needs to be set based on the cellular coverage problem, and population initialization needs to be completed. For details regarding chromosome coding, please refer to step 102 above; these details are omitted here. After determining the rules for chromosome coding, some initial chromosomes can be randomly generated, thus completing population initialization.

[0083] For any iteration in the multiple iterations of the genetic algorithm, the process includes: determining the fitness of each chromosome among the multiple chromosomes generated in the previous iteration, and performing replication, crossover, and mutation on the chromosomes in the previous generation population based on the fitness of each chromosome in the previous generation population.

[0084] In the first iteration, the fitness of each chromosome among the multiple chromosomes generated in the previous iteration is determined, which is to determine the fitness of each chromosome in the first generation population.

[0085] In this embodiment of the disclosure, the fitness of a chromosome is the coverage rate when k base stations are placed at the k locations indicated by the chromosome. Optionally, step fg is used to determine the fitness of each chromosome.

[0086] Step f: Based on the 3D model, line-of-sight probability model, and ray tracing algorithm, determine the first received power of each sampling point when the k base stations are placed at the positions indicated by the first chromosome.

[0087] The first chromosome is any chromosome in any iteration.

[0088] Optionally, step f includes the following four steps:

[0089] The first step is to determine the type of line-of-sight path from each base station to each sampling point based on the 3D model and the line-of-sight probability model.

[0090] The types of line-of-sight paths include direct line-of-sight paths, reflected line-of-sight paths, and diffracted line-of-sight paths.

[0091] Figure 4 For a schematic diagram of the line-of-sight probability model, see [link / reference]. Figure 4 In the line-of-sight probability model, the center point represents base station 401. Multiple dense line-of-sight paths 402 are emitted outward from base station 401. The starting point of each line-of-sight path is base station 401, and the ending point of each line-of-sight path is a sampling point. The line-of-sight probability model can predict the type of line-of-sight path from each base station to each sampling point based on the starting and ending points of the line-of-sight paths and the three-dimensional model. The line-of-sight path can represent the propagation path of the wireless signal. The three-dimensional model includes multiple models, such as model 403.

[0092] Figure 5 This is a schematic diagram of the types of line-of-sight paths. Figure 5 Part (a) is a schematic diagram of the direct line of sight path, as shown below. Figure 5 As shown in part (a), a direct line of sight starts at RS and ends at R0, and the line of sight is not blocked by a 3D model between its start and end.

[0093] Figure 5 Part (b) is a schematic diagram of the reflected line of sight path, such as Figure 5 As shown in part (b), there are two reflecting surfaces: a first reflecting surface P1 and a second reflecting surface P2. The line of sight originates from the source point S (the starting point), and after multiple reflections through the first reflecting surface P1 and the second reflecting surface P2, reaches the receiving point R (the focal point). Image point M1 is symmetrical to the source point S via the first reflecting surface P1, and image point M2 is symmetrical to the source point S via the second reflecting surface P2; secondary image point m1 is symmetrical to image point M1 via the second reflecting surface P2, and secondary image point m2 is symmetrical to image point M2 via the first reflecting surface P1. Once the source point S, receiving point R, first reflecting surface P1, second reflecting surface P2, image point M1, image point M2, secondary image point m1, and secondary image point m2 are determined, the reflected line of sight path originating from the source point S and reaching the receiving point S after reflections through the first reflecting surface P1 and the second reflecting surface P2 can be determined.

[0094] Figure 5 Part (c) is a schematic diagram of the diffraction line of sight path, as shown below. Figure 5 As shown in part (c), the incident plane of a certain diffracted line of sight path is the plane containing TxD1B1, and the starting point Tx of this diffracted line of sight path is located in the incident plane. After the diffracted line of sight originates from the starting point, it passes through I diffracting planes and finally reaches the ending point Rx. Among them, D1B1D2B2 is a diffracting plane.

[0095] There are many implementation methods for line-of-sight probability models, which will not be detailed here.

[0096] The second step is to determine the line-of-sight path from each base station to each sampling point based on the ray tracing algorithm, the type of line-of-sight path, and the 3D model.

[0097] Once the type of line-of-sight path is determined, a ray tracing algorithm can be used to determine the line-of-sight path from each base station to each sampling point based on the 3D model.

[0098] For example, consider k base stations and n sampling points, where k and n are both positive integers. For each of the k base stations, a line-of-sight path needs to be determined between that base station and the n sampling points. This results in a total of k × n line-of-sight paths being determined.

[0099] There are many implementation methods for ray tracing algorithms in related technologies, so they will not be detailed here.

[0100] If the line-of-sight path from each base station to each sampling point is determined directly based on the 3D model using a ray tracing algorithm without considering the influence of the line-of-sight path type, errors will occur in the calculated line-of-sight path. In this embodiment, a line-of-sight probability model is used to determine the line-of-sight path type. After determining the line-of-sight path type, the ray tracing algorithm is then used to determine the line-of-sight path from each base station to each sampling point according to the line-of-sight path type, which can effectively improve the accuracy of the determined line-of-sight path.

[0101] In this embodiment, since ray tracing is required to calculate the fitness of each chromosome, it is necessary to improve the execution efficiency of the ray tracing algorithm as much as possible, thereby improving the execution efficiency of the genetic algorithm. For example, the execution efficiency of the ray tracing algorithm can be improved by adding parallel processing tasks.

[0102] The third step is to determine the path loss for each line of sight.

[0103] Once the line-of-sight path is determined, the path loss generated during the process of the wireless signal propagating from the base station to the sampling point along the line-of-sight path can be calculated.

[0104] Optionally, for a direct line-of-sight path, the path loss is calculated using a free-space attenuation model; for a reflected line-of-sight path, the path loss is calculated using geometric optics theory; and for a diffracted line-of-sight path, the path loss is calculated using the principle of uniform diffraction.

[0105] For calculating path loss using geometric optics theory, taking a reflected line-of-sight path undergoing one reflection as an example, the path loss is the sum of the path loss from the starting point to the reflection point, the signal strength at the reflection point, and the path loss from the reflection point to the destination. The path loss from the starting point to the reflection point and the path loss from the reflection point to the destination can be calculated using a free-space attenuation model.

[0106] When the frequency of a wireless signal is sufficiently high, the wavelength of the wireless signal is much smaller than the edge length of urban buildings. Therefore, the reflection path of the wireless signal is similar to that of light wave reflection. The formula for the intensity of the wireless signal at the reflection point is similar to the formula for the intensity of a light wave reflection signal, i.e., the signal intensity at the reflection point = incident signal intensity × reflection coefficient. The reflection coefficient is determined according to the actual situation, and this disclosure does not limit it.

[0107] There are many implementation methods for the free space decay model and the principle of uniform diffraction, which will not be detailed here.

[0108] The fourth step is to determine the first received power for each sampling point based on the path loss of each line of sight path.

[0109] The first received power at each sampling point is calculated using formula (2):

[0110] P1 = P0 - P2 (2)

[0111] Where P1 is the first received power, P0 is the transmit power of each base station, and P2 is the path loss of the line-of-sight path.

[0112] Step g: Based on the magnitude of the first received power and the power threshold at each sampling point, determine the coverage of the first chromosome. The coverage of the first chromosome is the fitness value of the first chromosome.

[0113] The coverage of the first chromosome can be calculated using formula (1). Here, the first received power of each sampling point has been calculated in step f, and the number of sampling points with a first received power less than the power threshold is the number of bad receiving points. The total number of sampling points is known. Therefore, the coverage of the first chromosome can be achieved using formula (1). The determined coverage of the first chromosome is the fitness value of the first chromosome.

[0114] Once the fitness of each chromosome is determined, operations such as replication, crossover, and mutation can be performed on multiple chromosomes in this iteration based on their fitness.

[0115] Optionally, for any crossover in an iteration, a fourth chromosome is generated based on the second and third chromosomes.

[0116] The second and third chromosomes are two different chromosomes generated in the previous iteration of any iteration, and the positions of the k base stations indicated by the fourth chromosome are the average of the positions of the k base stations indicated by the second and third chromosomes.

[0117] For example, if the second chromosome is used to indicate placing a base station at (x1, y1, z1), the third chromosome is used to indicate placing the base station at (x2, y2, z2), then the fourth chromosome is used to indicate placing the base station at... Place.

[0118] In this embodiment, a fourth chromosome is generated based on the second and third chromosomes during crossover, resulting in a reduced number of chromosomes compared to conventional crossover methods. To ensure that the number of chromosomes in the population remains unchanged, the fourth chromosome is replicated in this embodiment, effectively resulting in two fourth chromosomes after crossover, thus ensuring that the population size does not change.

[0119] Optionally, common replication and mutation methods can be used for replication and mutation in any iteration. For example, when replicating, chromosomes with high fitness in the previous generation of any iteration can be selected for replication and retention; while when mutating, single-point mutation or multi-point mutation can be used. The embodiments of this disclosure do not limit the implementation method of replication and mutation.

[0120] After multiple iterations using a genetic algorithm, an optimal solution can be obtained, and the fitness value of this optimal solution is the first coverage rate. At this point, it is necessary to determine the relationship between the first coverage rate and the coverage threshold. If the first coverage rate is greater than the coverage threshold, proceed to step 204; if the first coverage rate is less than the coverage threshold, proceed to step 203; if the first coverage rate is equal to the coverage threshold, proceed to step 206.

[0121] In step 203, based on the genetic algorithm, the location of k+1 base stations is determined when the coverage rate of k+1 base stations is the second coverage rate.

[0122] The second coverage rate is the maximum coverage rate of k+1 base stations in the target area, and the sum of the transmission power of k+1 base stations is the second power threshold.

[0123] Step 203 essentially means that after k equals k+1, step 202 is executed, which will be omitted in detail here. After step 203 is completed, an optimal solution can also be obtained. At this time, it is still necessary to determine the relationship between the fitness value of the optimal solution and the coverage threshold. If the fitness value of the optimal solution is less than the coverage threshold, step 203 is repeated until the fitness value of the optimal solution obtained in a certain iteration is greater than or equal to the coverage threshold.

[0124] If the fitness value of the optimal solution obtained in a certain iteration is greater than the coverage threshold, step 204 is executed; if the fitness value of the optimal solution obtained in a certain iteration is equal to the coverage threshold, step 206 is executed.

[0125] In step 204, the transmit power of multiple base stations is reduced so that the sum of the transmit power of the multiple base stations is less than the second power threshold.

[0126] When performing step 204, the optimal solution could be k base stations, k+1 base stations, k+2 base stations, and so on. The following explanation uses the example of multiple optimal solutions with k base stations.

[0127] For example, the transmit power of each base station is Reducing the transmit power of multiple base stations can be achieved by setting the transmit power of each base station to [a specific value]. Among them, P n This refers to the amount by which the transmit power of each base station is reduced. P nThe specific value can be set based on experience, and this embodiment does not limit it.

[0128] In step 205, based on the reduced transmit power of the multiple base stations, a genetic algorithm is used to determine the placement of the multiple base stations when the coverage rate is the third coverage rate.

[0129] The third coverage rate is the maximum coverage rate after multiple base stations reduce their transmission power.

[0130] Step 205 essentially involves the transmit power of each base station. Step 202 is then executed based on the above steps, which will be omitted here.

[0131] After performing step 205, an optimal solution can be obtained, and the fitness value of the optimal solution is the third coverage.

[0132] If the third coverage rate is greater than the coverage threshold, repeat steps 204-205 until the third coverage rate equals the coverage threshold.

[0133] By using steps 204-205, when the fitness value of the iterated optimal solution is greater than the coverage threshold, the transmit power of the base station can be effectively reduced, thereby enabling the most economical base station site selection.

[0134] In step 206, the locations of the multiple base stations indicated by the optimal solution are used as the locations of the multiple base stations.

[0135] When the third coverage rate equals the coverage rate threshold, step 206 is to use the location of the multiple base stations as the site selection of the multiple base stations when the coverage rate of the multiple base stations is the third coverage rate.

[0136] Optionally, the coverage range of each base station is 30cm.

[0137] Those skilled in the art will understand that the coverage rate equal to the coverage rate threshold in the embodiments of this disclosure is essentially a reference to the solution with the coverage rate closest to the coverage rate threshold, and is not limited to making the coverage rate exactly equal to the coverage rate threshold.

[0138] The following are device embodiments of this application. For details not described in detail in the device embodiments, please refer to the above method embodiments.

[0139] Figure 6 A schematic diagram of a base station addressing device provided in an exemplary embodiment of this disclosure is shown below. See also: Figure 6 The base station location selection device 600 includes a modeling module 601 and a genetic algorithm module 602.

[0140] The modeling module 601 is used to model the target area to obtain a three-dimensional model of the target area, which includes multiple sampling points and multiple locations for placing base stations.

[0141] The genetic algorithm module 602 is used to determine the placement of k base stations when the coverage rate of k base stations is a first coverage rate. Each chromosome in the genetic algorithm is used to indicate that the k base stations are placed in k different locations. The k different locations are any k locations from multiple locations. The coverage rate is used to indicate the proportion of sampling points with received power greater than or equal to a first power threshold among multiple sampling points. The first coverage rate is the maximum coverage rate of the k base stations in the target area. The three-dimensional model is used to determine the received power of each sampling point. Here, k is a positive integer and k is greater than 0.

[0142] Optionally, the genetic algorithm module 602 is also used to determine the first received power of each sampling point when k base stations are placed at the positions indicated by the first chromosome, based on the three-dimensional model, the line-of-sight probability model, and the ray tracing algorithm, where the first chromosome is any chromosome in any iteration; and to determine the coverage of the first chromosome based on the first received power of each sampling point and the power threshold, where the coverage of the first chromosome is the fitness value of the first chromosome.

[0143] Optionally, the genetic algorithm module 602 is also used to determine the type of line-of-sight path from each base station to each sampling point based on the three-dimensional model and the line-of-sight probability model. The types of line-of-sight paths include direct line-of-sight paths, reflected line-of-sight paths, and diffracted line-of-sight paths. Based on the ray tracing algorithm, the type of line-of-sight path, and the three-dimensional model, the line-of-sight path from each base station to each sampling point is determined. The path loss of each line-of-sight path is determined. Based on the path loss of each line-of-sight path, the first received power of each sampling point is determined.

[0144] Optionally, the genetic algorithm module 602 is also used to generate a fourth chromosome based on the second and third chromosomes for crossover in any iteration. The second and third chromosomes are two different chromosomes generated in the previous iteration of any iteration, and the positions of the k base stations indicated by the fourth chromosome are the average of the positions of the k base stations indicated by the second and third chromosomes.

[0145] Optionally, the genetic algorithm module 602 is further configured to determine the location of the k+1 base stations based on the genetic algorithm when the first coverage rate is less than the coverage rate threshold, and when the coverage rate of the k+1 base stations is the second coverage rate, wherein the second coverage rate is the maximum coverage rate of the k+1 base stations in the target area, and the sum of the transmission power of the k+1 base stations is the second power threshold.

[0146] Optionally, the genetic algorithm module 602 is further configured to, when the first coverage rate is greater than the coverage rate threshold, reduce the transmission power of multiple base stations so that the sum of the transmission power of the multiple base stations is less than the second power threshold; based on the reduced transmission power of the multiple base stations, use the genetic algorithm to determine the location of the multiple base stations when the coverage rate of the multiple base stations is the third coverage rate, where the third coverage rate is the maximum coverage rate after the transmission power of the multiple base stations is reduced; if the third coverage rate is equal to the coverage rate threshold, the location of the multiple base stations when the coverage rate of the multiple base stations is the third coverage rate is used as the site selection of the multiple base stations.

[0147] Optionally, the modeling module 601 is further configured to acquire a first set of aerial images, which includes multiple aerial images of the target area; determine the first point cloud data of the target area based on the multiple aerial images; perform semantic segmentation on the first point cloud data; perform incremental reconstruction on the semantically segmented first point cloud data to obtain second point cloud data; and perform surface reconstruction based on the second point cloud data to obtain a three-dimensional model.

[0148] It should be noted that the base station location selection device provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the base station location selection device and the base station location selection method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0149] The module division in this embodiment is illustrative and represents only one logical functional division. In actual implementation, other division methods are possible. Furthermore, the functional modules in the various embodiments of this disclosure can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0150] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a terminal device (which may be a personal computer, mobile phone, or communication device, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this disclosure. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0151] Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this disclosure. For example... Figure 7 As shown, the computer device 700 includes a processor 701 and a memory 702.

[0152] Processor 701 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 701 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 701 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 701 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 701 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0153] The memory 702 may include one or more computer-readable storage media, which may be non-transitory. The memory 702 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 702 is used to store at least one instruction, which is executed by the processor 701 to implement the base station addressing method provided in the embodiments of this disclosure.

[0154] Those skilled in the art will understand that Figure 7 The structure shown does not constitute a limitation on the computer device 700, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0155] This disclosure also provides a non-transitory computer-readable storage medium, wherein when the instructions in the storage medium are executed by the processor of a computer device, the computer device is able to execute the base station location method provided in this disclosure.

[0156] This disclosure also provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the base station location method provided in this disclosure.

[0157] The above description is merely an optional embodiment of this disclosure and is not intended to limit this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the protection scope of this disclosure.

Claims

1. A base station siting method, characterized by, The method comprises: modeling a target area to obtain a three-dimensional model of the target area, the target area comprising a plurality of sampling points and a plurality of positions for placing base stations; determining, using a genetic algorithm, positions of k base stations under a condition that a coverage rate of the k base stations is a first coverage rate, each chromosome in the genetic algorithm being used to indicate that the k base stations are placed at k different positions, the k different positions being any k positions in the plurality of positions, the coverage rate being used to indicate a proportion of the sampling points with a received power greater than or equal to a first power threshold in the plurality of sampling points, the first coverage rate being a maximum coverage rate of the k base stations in the target area, and the three-dimensional model being used to determine the received power of each sampling point; wherein k is a positive integer greater than 0; the determining, using the genetic algorithm, the positions of the k base stations under the condition that the coverage rate of the k base stations is the first coverage rate comprises: determining, based on the three-dimensional model, a line-of-sight probability model and a ray tracing algorithm, a first received power of each sampling point when the k base stations are placed at positions of the k base stations indicated by a first chromosome, the first chromosome being any chromosome in a current iteration; determining, based on a size of the first received power of each sampling point and the power threshold, a coverage rate of the first chromosome, the coverage rate of the first chromosome being a fitness value of the first chromosome; the determining, based on the three-dimensional model, the line-of-sight probability model and the ray tracing algorithm, the first received power of each sampling point when the k base stations are placed at the positions of the k base stations indicated by the first chromosome comprises: determining, based on the three-dimensional model and the line-of-sight probability model, a type of a line-of-sight path from each base station to each sampling point, the type of the line-of-sight path comprising a direct line-of-sight path, a reflected line-of-sight path and a diffracted line-of-sight path; determining, based on the ray tracing algorithm, the type of the line-of-sight path and the three-dimensional model, the line-of-sight path from each base station to each sampling point; determining a path loss of each line-of-sight path, the path loss of the direct line-of-sight path being calculated by a free space attenuation model, the path loss of the reflected line-of-sight path being calculated by geometric optics theory, and the path loss of the diffracted line-of-sight path being calculated by a uniform theory of diffraction; determining the first received power of each sampling point based on the path loss of each line-of-sight path.

2. The method of claim 1, wherein, the determining, using the genetic algorithm, the positions of the k base stations under the condition that the coverage rate of the k base stations is the first coverage rate comprises: generating, based on a second chromosome and a third chromosome, a fourth chromosome for a crossover in a current iteration, the second chromosome and the third chromosome being two different chromosomes generated in a last iteration of the current iteration, and positions of the k base stations indicated by the fourth chromosome being an average of positions of the k base stations indicated by the second chromosome and the third chromosome.

3. The method of claim 1, wherein, transmission powers of the k base stations are equal, and a sum of the transmission powers of the k base stations is a second power threshold.

4. The method of claim 3, wherein, The method further comprises: In a case where the first coverage is less than the coverage threshold, positions of k+1 base stations are determined based on a genetic algorithm in a case where coverages of the k+1 base stations are second coverages, the second coverages are maximum coverages of the k+1 base stations in the target area, and a sum of transmission powers of the k+1 base stations is the second power threshold.

5. The method of claim 3, wherein, The method further includes: In a case where the first coverage is greater than the coverage threshold, transmission powers of the plurality of base stations are reduced so that the sum of the transmission powers of the plurality of base stations is less than the second power threshold; positions of the plurality of base stations are determined based on a genetic algorithm in a case where coverages of the plurality of base stations are third coverages after the transmission powers of the plurality of base stations are reduced, the third coverages are maximum coverages of the plurality of base stations after the transmission powers of the plurality of base stations are reduced; In a case where the third coverage is equal to the coverage threshold, the positions of the plurality of base stations in the case where the coverages of the plurality of base stations are the third coverages are taken as the site selection of the plurality of base stations.

6. The method of claim 1, wherein, The modeling of the target area includes: a first set of aerial images is acquired, the first set of aerial images including a plurality of aerial images of the target area; first point cloud data of the target area is determined based on the plurality of aerial images; the first point cloud data is subjected to semantic segmentation; the first point cloud data after the semantic segmentation is subjected to incremental reconstruction to obtain second point cloud data; the three-dimensional model is obtained based on surface reconstruction of the second point cloud data.

7. A base station site selection device, characterized in that, The apparatus includes: a modeling module configured to model a target area to obtain a three-dimensional model of the target area, the target area including a plurality of sampling points and a plurality of positions for placing base stations; a genetic algorithm module configured to determine positions of k base stations in a case where coverages of the k base stations are a first coverage by using a genetic algorithm, each chromosome in the genetic algorithm being used to indicate that the k base stations are placed at k different positions, the k different positions being any k positions of the plurality of positions, the coverage being used to indicate a proportion of the sampling points with a received power greater than or equal to a first power threshold in the plurality of sampling points, the first coverage being a maximum coverage of the k base stations in the target area, and the three-dimensional model being used to determine the received power of each sampling point, k being a positive integer greater than 0; the genetic algorithm module is further configured to determine a first received power of each sampling point in a case where the k base stations are placed at positions of the k base stations indicated by a first chromosome by using a line-of-sight probability model and a ray tracing algorithm, the first chromosome being any chromosome in any iteration; a coverage of the first chromosome is determined based on a size of the first received power of each sampling point and the power threshold, the coverage of the first chromosome being a fitness value of the first chromosome. ​ The genetic algorithm module is further configured to determine a type of line-of-sight path from each base station to each sampling point based on the three-dimensional model and a line-of-sight probability model, the type of line-of-sight path including a direct line-of-sight path, a reflected line-of-sight path and a diffracted line-of-sight path; determine the line-of-sight path from each base station to each sampling point based on the ray tracing algorithm, the type of line-of-sight path and the three-dimensional model; determine a path loss of each line-of-sight path, for the direct line-of-sight path, the path loss is calculated by a free space attenuation model, for the reflected line-of-sight path, the path loss is calculated by geometric optics theory, and for the diffracted line-of-sight path, the path loss is calculated by a uniform theory of diffraction; determine the first received power of each sampling point based on the path loss of each line-of-sight path.

Images