Base station selection method, system, device and storage medium based on clustering algorithm

Abstract

The present invention provides a base station selection method, system, device and storage medium based on a clustering algorithm, wherein the method includes: in the initial base station set of the target to be located, the number of base stations in each group is preset h, the initial base stations are arranged and combined, and the corresponding positioning coordinates are obtained for each, and a positioning coordinate set is established; the positioning coordinate set is clustered in turn based on the number of clusters k, and then the classification accuracy is evaluated to obtain the optimal number of clusters k; according to the clustering of the optimal number of clusters k, the coordinate mean of the cluster with the largest number of positioning coordinates in the cluster matches the nearest positioning coordinate, and the positioning coordinates of the target to be located and the corresponding positioning base station combination are obtained according to the positioning coordinates. The present invention can select an implementation scheme with a suitable number and position from the candidate base stations, while improving the resource utilization rate of the network element of the positioning management function, and improving the precision and accuracy of 5G positioning coordinates.

Description

Background Art

[0002] Since 5G communication is characterized by high speed, low latency, and large connections, its key technologies include large-scale antenna arrays, ultra-dense networking, new multiple access, full spectrum access, and new network architecture. In the continuous evolution of standards, 5G itself has also added high-precision positioning functions, such as industrial AGV, asset tracking, etc., especially precise indoor positioning. However, satellite positioning cannot be used indoors, and LTE and WiFi positioning technologies are not accurate (in fact, Bluetooth AOA, UWB and other technologies are already available for high-precision indoor positioning). For this reason, 5G has added a positioning function in the R16 version. It uses the MIMO multi-beam characteristics to define indoor positioning technologies such as the round-trip time (RTT) of the signal based on the cellular cell, the time difference of arrival (TDOA), the angle of arrival measurement method (AoA), and the angle of departure measurement method (AoD).

[0003] 5G positioning is to measure the reference signal through the terminal under test, and the location management function (LMF) calculates the measurement results. LMF calculates the location information of the terminal under test based on the stored known base station location information. According to the positioning principle, two-dimensional positioning requires at least three positioning base stations to complete the position calculation, but in actual applications, there are generally many base stations measured, and there is a large amount of redundant base station information. For example, calculating a large amount of base station data will result in a large amount of calculation and waste of LMF resources; if only three base stations are selected for calculation, the accuracy will be insufficient and the positioning will produce a large error.

[0004] Therefore, how to select the appropriate number and location of base stations from these candidate base stations is an important issue to be solved, which determines the precision and accuracy of subsequent positioning coordinates and is of great value for local or limited area positioning and segmentation of positioning areas.

[0005] In view of this, the present invention proposes a base station selection method, system, device and storage medium based on a clustering algorithm.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0007] In response to the problems in the prior art, the purpose of the present invention is to provide a base station selection method, system, device and storage medium based on a clustering algorithm, which overcomes the difficulties of the prior art and can select an implementation solution with a suitable number and position from the candidate base stations, thereby improving the resource utilization rate of the network element of the positioning management function and improving the precision and accuracy of 5G positioning coordinates.

[0008] An embodiment of the present invention provides a base station selection method based on a clustering algorithm, comprising the following steps:

[0009] In the initial base station set of the target to be located, the number of base stations in each group is preset to be h, where h≥3, the initial base stations are arranged and combined, and the corresponding positioning coordinates are obtained for each of them to establish a positioning coordinate set;

[0010] Clustering the positioning coordinate set based on the number of clusters k in turn, and then performing classification accuracy evaluation to obtain the optimal number of clusters k;

[0011] According to the clustering with the optimal number of clusters k, the coordinate mean of the cluster with the largest number of positioning coordinates matches the nearest positioning coordinate, and the positioning coordinates of the target to be positioned and the corresponding positioning base station combination are obtained according to the positioning coordinate.

[0012] Preferably, in the initial base station set of the target to be located, the number of base stations in each group is preset h, h≥3, the initial base stations are arranged and combined, and the corresponding positioning coordinates are obtained for each, and before the positioning coordinate set is established, the following steps are also included:

[0013] Based on the position of the target to be located, an initial base station set is obtained. The isolation forest algorithm is used to filter isolated base stations in the initial base station set according to the strength of the reference signal of the initial base station. If there are no redundant base stations, the positioning coordinates and the corresponding positioning base station combination are obtained; if there are still redundant base stations, the subsequent steps are executed.

[0014] Preferably, in the initial base station set of the target to be located, the number of base stations in each group is preset to be h, where h≥3, the initial base stations are arranged and combined, and the corresponding positioning coordinates are obtained for each, so as to establish a positioning coordinate set, including:

[0015] Based on the preset number of base stations h in each group, the value range of the number of base stations is [3, M], and the corresponding initial base stations are obtained according to different values ​​of the number of base stations h for permutation and combination;

[0016] Obtain a sum W of permutation and combination types obtained by different numbers of base stations h;

[0017] Each of the permutations and combinations obtains positioning coordinates based on the target to be positioned, and establishes a mapping relationship between the positioning coordinates and the permutations and combinations.

[0018] Preferably, the number of base stations h of each group is preset, the value range of the number of base stations is [3, M], and the corresponding initial base stations are obtained according to different values ​​of the number of base stations h for permutation and combination, including:

[0019]

[0020] Wherein, M represents the total number of initial base stations; P represents the total number of isolated base stations filtered.

[0021] Preferably, clustering the positioning coordinate sets based on the number of clusters k in sequence, and then performing classification accuracy evaluation to obtain the optimal number of clusters k, includes:

[0022] Selecting a first initial cluster center in the positioning coordinate set according to the density peak value;

[0023] Calculate the distance between each remaining positioning coordinate in the positioning coordinate set and the initial cluster center and the sum of the distances;

[0024] The positioning coordinate farthest from the initial cluster center is used as the next initial cluster center, the value range of the number of clusters k is [2, T], and each positioning coordinate is assigned to the nearest cluster; T is the smallest natural number greater than the root result of W;

[0025] The classification accuracy of the clustering results is evaluated, and the clustering result with the best evaluation result is selected as the final clustering result through iteration, and the value of the number of clusters k corresponding to the final clustering result is regarded as the optimal number of clusters.

[0026] Preferably, selecting a first initial cluster center in the positioning coordinate set according to the density peak value includes:

[0027] Get local density formula positioning ρ i It means that the circle with point i as the center is contained in the preset radius d. c The number of points in the circle, d ij is the distance between point j and point i in the positioning coordinate set, χ(x) indicates that the distance to data object i is less than the preset radius d c The number of data objects, when x<0, then χ(x)=1, when x≥0, then χ(x)=0;

[0028] The location coordinates with the largest local density are taken as the first initial cluster center.

[0029] Preferably, the clustering results are evaluated for classification accuracy, the clustering result with the best evaluation result is selected as the final clustering result through iteration, and the value of the number of clusters k corresponding to the final clustering result is regarded as the optimal number of clusters, including:

[0030] The clustering algorithm evaluation index DBI is used for evaluation.

[0031]

[0032] Among them, avg(C i ) is Ci The average Euclidean distance from a class sample to its class center, avg(C j ) is C j The average Euclidean distance from a class sample to its class center, d cen (u i ,u j ) is the C i and C j The Euclidean distance of the class center of the class;

[0033] When the DBI value is smaller, the dispersion is lower and the clustering result is better;

[0034] By iterating each round k=k+1, the evaluation results of the number of clusters k under different values ​​are obtained;

[0035] When k=T, the clustering result with the best evaluation result is selected as the final clustering result, and the value of the number of clusters k corresponding to the final clustering result is regarded as the optimal number of clusters.

[0036] Preferably, in the clustering according to the optimal number of clusters k, the coordinate mean of the cluster with the largest number of positioning coordinates in the cluster matches the nearest positioning coordinate, and the positioning coordinates of the target to be positioned and the corresponding positioning base station combination obtained according to the positioning coordinates include:

[0037] Perform clustering according to the optimal number of clusters, and select the cluster with the largest number of positioning coordinates as the optimal cluster;

[0038] Obtaining the average value of all positioning coordinates in the optimal cluster;

[0039] The positioning coordinates closest to the average value are used as the positioning coordinates of the target to be positioned.

[0040] Preferably, in the clustering according to the optimal number of clusters k, the coordinate mean of the cluster with the largest number of positioning coordinates in the cluster matches the nearest positioning coordinate, and the positioning coordinates of the target to be positioned and the corresponding positioning base station combination obtained according to the positioning coordinates also include:

[0041] The arrangement and combination of the initial base stations corresponding to the positioning coordinates are obtained according to the mapping relationship.

[0042] An embodiment of the present invention further provides a base station selection system based on a clustering algorithm, which is used to implement the above-mentioned base station selection method based on a clustering algorithm. The base station selection system based on a clustering algorithm includes:

[0043] The permutation and combination module presets the number of base stations in each group h, where h≥3, in the initial base station set of the target to be located, performs permutation and combination on the initial base stations, obtains corresponding positioning coordinates for each of the base stations, and establishes a positioning coordinate set;

[0044] A coordinate clustering module clusters the positioning coordinate set based on the number of clusters k in turn, and then performs classification accuracy evaluation to obtain the optimal number of clusters k;

[0045] The positioning coordinate module matches the nearest positioning coordinate with the coordinate mean of the cluster with the largest number of positioning coordinates in the cluster according to the clustering of the optimal number of clusters k, and obtains the positioning coordinate of the target to be positioned and the corresponding positioning base station combination according to the positioning coordinate.

[0046] An embodiment of the present invention further provides a base station selection device based on a clustering algorithm, comprising:

[0047] processor;

[0048] a memory storing executable instructions of the processor;

[0049] The processor is configured to execute the steps of the above-mentioned base station selection method based on the clustering algorithm by executing the executable instructions.

[0050] An embodiment of the present invention further provides a computer-readable storage medium for storing a program, wherein the program, when executed, implements the steps of the above-mentioned base station selection method based on the clustering algorithm.

[0051] The purpose of the present invention is to provide a base station selection method, system, device and storage medium based on a clustering algorithm, which can select an implementation scheme with a suitable number and position from candidate base stations, thereby improving the resource utilization rate of network elements of the positioning management function and improving the precision and accuracy of 5G positioning coordinates. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Other features, objectives and advantages of the present invention will become more apparent from a reading of the detailed description of non-limiting embodiments made with reference to the following accompanying drawings.

[0053] Figure 1 It is a flow chart of the base station selection method based on clustering algorithm of the present invention.

[0054] Figure 2 It is a flowchart of step S120 in an embodiment of the base station selection method based on clustering algorithm of the present invention.

[0055] Figure 3 It is a flowchart of step S130 in an embodiment of the base station selection method based on clustering algorithm of the present invention.

[0056] Figure 4 It is a flowchart diagram of step S140 in an embodiment of the base station selection method based on clustering algorithm of the present invention.

[0057] Figure 5It is a flowchart of specific steps for implementing the base station selection method based on clustering algorithm of the present invention.

[0058] Figure 6 , 7 , 8, and 9 are schematic diagrams of the step scenarios for implementing the base station selection method based on clustering algorithm of the present invention.

[0059] Fig.10 It is a module schematic diagram of a system for implementing the base station selection method based on clustering algorithm of the present invention.

[0060] Fig.11 It is a module schematic diagram of the permutation and combination module in the embodiment of the base station selection system based on clustering algorithm of the present invention.

[0061] Fig.12 It is a module schematic diagram of a coordinate clustering module in an embodiment of a base station selection system based on a clustering algorithm of the present invention.

[0062] Fig.13 It is a module schematic diagram of a positioning coordinate module in an embodiment of a base station selection system based on a clustering algorithm of the present invention.

[0063] Fig.14 It is a schematic diagram of a base station selection device based on a clustering algorithm of the present invention. DETAILED DESCRIPTION

[0064] The following describes the implementation methods of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in the present application. The present application can also be implemented or applied through other different specific implementation methods, and the details in the present application can also be modified or changed in various ways according to different viewpoints and application systems without departing from the spirit of the present application. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0065] The following is a detailed description of the embodiments of the present application with reference to the accompanying drawings so that those skilled in the art can easily implement the present application. The present application can be embodied in a variety of different forms and is not limited to the embodiments described herein.

[0066] In the representations of the present application, the representations with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. mean that the specific features, structures, materials or characteristics represented in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the represented specific features, structures, materials or characteristics may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples represented in the present application and the features of the different embodiments or examples, unless they contradict each other.

[0067] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the representation of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0068] In order to clearly describe the present application, components irrelevant to the description are omitted, and the same reference symbols are given to the same or similar components throughout the specification.

[0069] Throughout the specification, when a device is said to be "connected" to another device, this includes not only the case of "direct connection" but also the case of "indirect connection" by placing other elements therebetween. In addition, when a device is said to "include" a certain component, unless otherwise stated, it does not exclude other components, but means that other components may be included.

[0070] When a device is said to be "on" another device, it may be directly on the other device, but there may also be other devices between it. In contrast, when a device is said to be "directly" on another device, there are no other devices between it.

[0071] Although the terms first, second, etc. are used to represent various elements in some examples, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, the first interface and the second interface, etc. are represented. Moreover, as used in this article, the singular forms "one", "one" and "the" are intended to also include plural forms, unless there is an opposite indication in the context. It should be further understood that the terms "comprising" and "including" indicate the existence of features, steps, operations, elements, components, projects, kinds, and / or groups, but do not exclude the existence, occurrence or addition of one or more other features, steps, operations, elements, components, projects, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Therefore, "A, B or C" or "A, B and / or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B and C". Only when the combination of elements, functions, steps or operations is inherently mutually exclusive in some way, will there be an exception to this definition.

[0072] The technical terms used herein are only used to refer to specific embodiments and are not intended to limit the present application. The singular form used herein also includes the plural form unless the sentence clearly indicates the contrary meaning. The meaning of "including" used in the specification is to specify specific characteristics, regions, integers, steps, operations, elements and / or components, and does not exclude the existence or addition of other characteristics, regions, integers, steps, operations, elements and / or components.

[0073] Although not defined differently, all terms, including technical and scientific terms used herein, have the same meaning as those generally understood by those skilled in the art to which this application belongs. Terms defined in commonly used dictionaries are additionally interpreted as having meanings consistent with the relevant technical literature and the contents of the current disclosure, and unless defined, shall not be overly interpreted as ideal or very formal meanings.

[0074] Figure 1 FIG. 1 is a flow chart of a base station selection method based on a clustering algorithm of the present invention. Figure 1 As shown, the base station selection method based on the clustering algorithm of the present invention, the process of the present invention includes:

[0075] S120 , in the initial base station set of the target to be located, preset the number of base stations in each group h, where h≥3, arrange and combine the initial base stations, and obtain corresponding positioning coordinates for each of them to establish a positioning coordinate set.

[0076] S130 , clustering the positioning coordinate sets based on the number of clusters k in sequence, and then performing classification accuracy evaluation to obtain the optimal number of clusters k.

[0077] S140 , according to the clustering with the optimal number of clusters k, the coordinate mean of the cluster with the largest number of positioning coordinates matches the nearest positioning coordinate, and the positioning coordinates of the target to be positioned and the corresponding positioning base station combination are obtained according to the positioning coordinates.

[0078] The present invention provides a base station selection optimization method for 5G positioning. The method of the present invention first eliminates edge base stations with low measurement signals through an isolation forest algorithm; subsequently, a certain number of base stations are combined and the positioning results (positioning points) of each combination are calculated, and these scattered positioning points are clustered using an improved K-Means++ algorithm, and finally the base station combination is reversely determined based on the positioning point, thereby achieving the purpose of selecting the best base station combination. The method of the present invention can reversely select a base station combination based on the best positioning result from the perspective of reducing positioning errors, thereby improving indoor positioning accuracy, and is of great value for local or limited area positioning and segmentation of positioning areas.

[0079] Figure 2 It is a flowchart of step S120 in an embodiment of the base station selection method based on clustering algorithm of the present invention. Figure 3 It is a flowchart of step S130 in an embodiment of the base station selection method based on clustering algorithm of the present invention. Figure 4 It is a flowchart diagram of step S140 in an embodiment of the base station selection method based on clustering algorithm of the present invention. Figures 2 to 4 As shown, in Figure 1 In the embodiment, on the basis of steps S120, S130, and S140, step S110 is included before step S120, step S120 is replaced by S121, S122, and S123, step S130 is replaced by S131, S132, S133, and S134, and step S140 is replaced by S141, S142, S143, and S144. Each step is described below:

[0080] S110, based on the position of the target to be located, an initial base station set is obtained, and an isolation forest algorithm is used to filter isolated base stations in the initial base station set according to the strength of the reference signal of the initial base station. If there is no redundant base station, the positioning coordinates and the corresponding positioning base station combination are obtained; if there is still a redundant base station, the subsequent steps are executed. In this embodiment, the number of base stations in the filtered initial base station set is greater than three as a judgment condition for the existence of redundant base stations, and if it is less than or equal to three, it is considered that there is no redundant base station, so the positioning coordinates are directly output. Among them, the isolation forest algorithm is an anomaly detection method based on Ensemble, so it has a linear time complexity. And it has high accuracy and fast speed when processing big data, so it is currently widely used in the industry. Common scenarios include: attack detection in network security, financial transaction fraud detection, disease detection, noise data filtering (data cleaning), etc. Different algorithms are used for detection for different types of anomalies, and the isolation forest algorithm is mainly aimed at abnormal points in continuous structured data. The premise of using the isolation forest is to define abnormal points as those "outliers that are easily isolated" - which can be understood as points that are sparsely distributed and far away from high-density groups. From a statistical point of view, in data space, if there are only sparsely distributed points in an area, it means that the probability of data points falling in this area is very low. Therefore, the points in these areas can be considered abnormal, but this is not limited to this.

[0081] S121, based on the preset number of base stations in each group h, the value range of the number of base stations is [3, M], the corresponding initial base stations are obtained according to different values ​​of the number of base stations h for permutation and combination, wherein,

[0082]

[0083] Wherein, M represents the total number of initial base stations; P represents the total number of isolated base stations filtered.

[0084] S122. Obtain a sum W of permutation and combination types obtained with different numbers of base stations h.

[0085] S123. Each permutation and combination obtains positioning coordinates based on the target to be located, and establishes a mapping relationship between the positioning coordinates and the permutation and combination.

[0086] S131, select the first initial cluster center in the positioning coordinate set according to the density peak. Step S131, including: obtaining the local density formula positioning ρ i It means that the circle with point i as the center is contained in the preset radius d. c The number of points in the circle, d ij is the distance between point j and point i in the positioning coordinate set, χ(x) indicates that the distance to data object i is less than the preset radius d cThe number of data objects, when x<0, then χ(x)=1, when x≥0, then χ(x)=0; the positioning coordinate with the largest local density is taken as the first initial clustering center, but it is not limited to this.

[0087] S132: Calculate the distance from each of the remaining positioning coordinates in the positioning coordinate set to the initial cluster center and the sum of the distances.

[0088] S133, the positioning coordinate farthest from the initial cluster center is used as the next initial cluster center, the value range of the number of clusters k is [2, T], and each positioning coordinate is assigned to the nearest cluster; T is the smallest natural number greater than the root result of W.

[0089] S134, evaluating the classification accuracy of the clustering results, selecting the clustering result with the best evaluation result as the final clustering result through iteration, and considering the value of the number of clusters k corresponding to the final clustering result as the optimal number of clusters. Step S134 includes:

[0090] The clustering algorithm evaluation index DBI is used for evaluation.

[0091]

[0092] Among them, avg(C i ) is C i The average Euclidean distance from a class sample to its class center, avg(C j ) is C j The average Euclidean distance from a class sample to its class center, d cen (u i ,u j ) is the C i and C j The Euclidean distance of the class center of the class;

[0093] When the DBI value is smaller, the dispersion is lower and the clustering result is better;

[0094] By iterating each round k=k+1, the evaluation results of the number of clusters k under different values ​​are obtained;

[0095] Until k = T, the clustering result with the best evaluation result is selected as the final clustering result, and the value of the number of clusters k corresponding to the final clustering result is regarded as the optimal number of clusters. The Davidson-Bouldin Index (DBI), also known as the classification accuracy index, is an indicator proposed by David L. Davis and Donald Bouldin to evaluate the quality of clustering algorithms, but is not limited to this.

[0096] S141 , clustering is performed according to the optimal number of clusters, and the cluster with the largest number of positioning coordinates is selected as the optimal cluster.

[0097] S142. Obtain the average value of all positioning coordinates in the optimal cluster.

[0098] S143: Use the positioning coordinates closest to the average value as the positioning coordinates of the target to be positioned.

[0099] S144. Obtain the arrangement and combination of initial base stations corresponding to the positioning coordinates according to the mapping relationship.

[0100] The present invention can remove the influence of isolated base stations on clustering results through the isolation forest algorithm. Based on the initial measurement results, preliminary screening is performed. The distance between individual base stations and the target to be measured is far, and the reference signal measurement result is low, which will affect the accuracy of positioning. The isolation forest algorithm can quickly separate P sparsely distributed base stations through super cutting.

[0101] The present invention obtains clustering results and positioning coordinates based on the improved K-Means++ algorithm. The basic idea of ​​the k-means++ algorithm to select the initial seed point is that the distance between the initial cluster centers should be as far as possible. 1. Randomly select a point from the input data point set as the first cluster center ("seed point"). 2. For each point x in the data set, calculate its distance D(x) from the nearest cluster center (referring to the selected cluster center). 3. Select a new data point as the new cluster center. The principle of selection is: the point with a larger D(x) has a greater probability of being selected as the cluster center. 4. Repeat 2 and 3 until k cluster centers are selected. 5. Use these k initial cluster centers to run the standard k-means algorithm.

[0102] In the present invention, first, candidate base stations are grouped into h (h is variable) to form multiple base station combinations (a total of Then, the initial point selection: select the first initial cluster center according to the density peak method to ensure that the first cluster center is generated at the core of the whole. Then, the number of clusters is determined: the number of clusters ranges from The first round is preset to 2. As the number of clustering rounds increases, the number of clusters increases until it reaches the upper limit. Use the maximum distance sum criterion proposed by K-Means++ to select the remaining initial cluster centers. Use an iterative algorithm to find a more suitable cluster center. Finally, evaluate the clustering results using the DBI evaluation index to obtain an evaluation result. Repeat steps ②-⑤, where the number of clusters set in step ③ increases, and use step ⑤ to evaluate the clustering results corresponding to the number of clusters. Use the DBI evaluation method to compare the clustering results of different preset numbers of clusters, and select the clustering result with the best evaluation.

[0103] In addition, the present invention infers the base station based on the best positioning coordinates. According to the characteristic that the positioning result is generated around the real position, the cluster with the most coordinates in the clustering result is selected, the mean value of the coordinates within the cluster is calculated, and the coordinates closest to the mean value of the coordinates are selected as the final positioning coordinates. According to the coordinates, the combination of positioning base stations can be inferred, thereby determining the selected base station.

[0104] The base station selection method based on clustering algorithm of the present invention can select an implementation scheme with a suitable number and position from the candidate base stations, thereby improving the resource utilization rate of the positioning management function network element and enhancing the precision and accuracy of 5G positioning coordinates.

[0105] Figure 5 FIG. 1 is a flowchart of the specific steps of implementing the base station selection method based on the clustering algorithm of the present invention. Figure 5 The template shown implements the specific steps of the base station selection method based on the clustering algorithm of the present invention, including:

[0106] Determine the terminal to be located.

[0107] The initial measurement result of the terminal to be located is obtained, and the positioning assistance data is obtained at the same time.

[0108] Generate initial positioning coordinates, use the isolation forest algorithm to eliminate isolated base stations, and determine whether the number of remaining base stations is greater than 3.

[0109] If the number is less than or equal to 3, the initial positioning coordinates are directly output as the positioning coordinates of the terminal to be positioned.

[0110] If there are more than 3, cluster the positioning results after removing isolated points based on density peak and K-Means++ algorithm. Use internal validity index DBI to evaluate clustering quality and output clustering results. Record the number of internal coordinates of each cluster and select the one with the largest number as class C. Calculate the average coordinate value of class C. The coordinates closest to it are selected as the final positioning coordinates. The positioning coordinates of the terminal to be positioned are output, and the base station combination is inferred based on the positioning coordinates.

[0111] Figure 6 , 7 , 8, and 9 are schematic diagrams of the steps of implementing the base station selection method based on the clustering algorithm of the present invention. Figure 5 , 6As shown in Figures 7, 8, and 9, the present invention uses 5G positioning to measure the reference signal through the terminal under test, and the location management function network element (Location Management Function, LMF) calculates the measurement results. LMF calculates the location information of the terminal under test 12 based on the stored known base station location information. However, there are many positioning base stations 11 measured, and there is a large amount of redundant base station information. For example, calculating a large amount of base station data will result in a large amount of calculation and waste of LMF resources; and if only 3 base stations are selected for calculation, the accuracy will be insufficient and the positioning will produce a large error. After adding the 5G positioning base station selection function:

[0112] Step 1: LMF performs screening based on the initial reference signal measurement results. P isolated base stations 10 that are relatively far from the target to be measured are eliminated through the Isolation Forest Algorithm (IForest). IForest mainly includes two parts, the training phase and the testing phase.

[0113] Training phase: First, randomly select n sample points from the training data set and put them into the root node; second, randomly select a dimension and cut it within the data range; finally, repeat the above steps until each node has only one data or reaches the preset height.

[0114] Testing phase: First, let the training data x traverse each iTree; second, calculate the layer of the tree where x falls; third, calculate the average height of x in each tree; finally, when the anomaly score s is greater than the set threshold and the signal strength from the base station to the target to be tested is less than the average of all base stations, it is determined to be an isolated point.

[0115] s(x,n)=2^(-E(h(x)) / c(n))

[0116] For example, the target to be measured can communicate with 10 nearby base stations (M1, M2, M3, M4, M5, M6, M7, M8, M9, M10), and the signal strength is [-44-42-52-78-30-36-54-36-72-47], in dBm. The isolation forest algorithm can identify "-78 and -72" in the above one-dimensional array as abnormal values, and then eliminate the two isolated base stations (M4 and M9) to avoid affecting the accuracy of positioning.

[0117] Step 2: If there are still redundant base stations after removing isolated base stations, continue to perform subsequent steps. First, build multiple base station combinations, each of which contains h (3≤h≤M) base stations. Traversal can be obtained Base station combinations and corresponding Where M represents the total number of initial base stations; P represents the number of removed base stations; h represents a group of h base stations, which can be used to calculate a corresponding positioning point) (the multiple combined positioning results 13 obtained in this step are the coordinate points scattered in the positioning area, which are the source of samples for clustering in subsequent steps).

[0118] For example, after eliminating two base stations through the isolation forest, there are still eight candidate base stations (M1, M2, M3, M5, M6, M7, M8, M10). There are three base station combinations, each of which can obtain corresponding positioning results.

[0119] Step 3: Use the improved K-Means++ algorithm for clustering. The specific steps of the improved algorithm are as follows:

[0120] (1) Select the first initial cluster center based on the density peak. Local density formula positioning χ(x) indicates that the distance to data object i is less than the cutoff distance dist c The number of data objects.

[0121] (2) Calculate the sum of the distances from each coordinate point to the existing cluster center. Select the coordinate with the largest value as the next initial cluster center. (The number of clusters is initially set to 2)

[0122] (3) Assign each positioning result to the cluster with the highest similarity and find a reasonable cluster center through iteration.

[0123] (4) Use the improved DBI method to evaluate the clustering results. Use the following formula Evaluate the clustering results within this range.

[0124] After the first round of steps (2)-(4) is completed, the set number of clusters is incremented (the range of cluster numbers is ), execute steps (2)-(4) again, and output the clustering evaluation result DBI(i) of each round. Compare the DBI(i) of each round, select the one with the best evaluation result as the final clustering result, and regard the number of clusters corresponding to this round as the optimal number of clusters.

[0125] For example, based on the total number of positioning results of 219, the number of clusters is set in the range of [2, 15]. In the first round, the number of clusters is set to 2, the improved K-Means++ algorithm is executed, and the clustering quality is evaluated by DBI. Within the range of the number of clusters, the above steps are executed again until the number of clusters reaches 15. Assume that the minimum value of DBI in the evaluation result is 0.5, and the number of clusters is 6. That is, when the number of clusters is 6, the clustering result is the best.

[0126] Step 4: According to the clustering results, select the cluster with the largest number of intra-cluster coordinates as the optimal cluster. Calculate the mean of the intra-cluster coordinates The positioning coordinates closest to the average coordinates 15 are the final positioning coordinates 16, and the base station combination [base station 141, base station 142, base station 143, base station 144] is inferred based on the positioning results.

[0127] For example: among the 6 clusters, select the cluster with the largest number of coordinates as the optimal cluster. Assume that the mean value of the coordinates in this cluster is (15.74, 20.23). And assume that the coordinate closest to the coordinate (15.74, 20.23) in the 219 positioning results is (13, 17). The final positioning coordinate is (13, 17). Assume that the positioning coordinate (13, 17) is generated by four positioning base stations M1, M2, M5, and M6, then the 5G positioning base stations finally selected are M1, M2, M5, and M6.

[0128] Compared with the existing technology, the main advantages of this patent are:

[0129] 1. This solution designs an algorithm for selecting base stations in 5G positioning, which can select the optimal base station combination based on the initial measurement results to improve positioning accuracy.

[0130] 2. Using the Isolation Forest Algorithm (IForest), based on the size of the reference signal measured for the first time, P isolated base stations that are relatively far away from the target to be measured are eliminated to avoid affecting the accuracy of positioning.

[0131] 3. Use the improved K-Means++ algorithm to obtain clustering results and positioning coordinates:

[0132] (1) Improve the selection of initial points. Select the point with the largest density peak as the initial clustering point. The largest density means that the probability of the measured terminal being at this location is the highest, so it is more reasonable than random point selection.

[0133] (2) Improve the selection of k value. In this scheme, the selection of k value is not fixed. By introducing the DBI evaluation method, the clustering results corresponding to each selected k value are evaluated, and the final clustering result is determined based on the evaluation result.

[0134] (3) Generation of clustered sample points. This scheme can set a variety of base station combinations, and each combination can calculate a positioning result, that is, a coordinate point. Therefore, there is no need to pre-set the number of base stations in the combination, which is more reasonable.

[0135] 4. Infer the base station combination based on the clustering results and positioning coordinates. According to the rules, a positioning coordinate point is selected, and this point corresponds to a base station combination, so the number and location of the optimal base station combination can be determined.

[0136] This method is aimed at 5G positioning. In the case of a large number of redundant base stations, selecting appropriate positioning base stations is a practical problem that LMF needs to solve in positioning calculations. In addition, by eliminating isolated base stations and using clustering algorithms to select base stations, the number and location of positioning base stations are determined by algorithms, avoiding positioning errors caused by subjective determination of the number.

[0137] The present invention can be applied to local or limited area positioning in 5G positioning, and to the segmentation of positioning base stations in the positioning area. It solves the problem of base station selection in the process of LMF position calculation, avoids the positioning error caused by subjective selection of base station combination, and improves positioning precision and accuracy. It can also determine the positioning base station combination corresponding to the local positioning area, segment the positioning area, and determine the base station combination corresponding to a certain positioning area.

[0138] Fig.10 Schematic diagram of the modules of the system for implementing the base station selection method based on clustering algorithm of the present invention. Fig.10 As shown, the base station selection system based on clustering algorithm of the present invention includes but is not limited to:

[0139] The permutation and combination module 52 presets the number of base stations in each group h, h≥3, in the initial base station set of the target to be located, permutes and combines the initial base stations, obtains corresponding positioning coordinates for each, and establishes a positioning coordinate set.

[0140] The coordinate clustering module 53 clusters the positioning coordinate sets based on the number of clusters k in sequence, and then performs classification accuracy evaluation to obtain the optimal number of clusters k.

[0141] The positioning coordinate module 54 matches the coordinate mean of the cluster with the largest number of positioning coordinates in the cluster according to the clustering of the optimal number of clusters k with the nearest positioning coordinate, and obtains the positioning coordinates of the target to be positioned and the corresponding positioning base station combination according to the positioning coordinates.

[0142] The implementation principle of the above modules can be found in the relevant introduction of the base station selection method based on clustering algorithm, which will not be repeated here.

[0143] The base station selection system based on clustering algorithm of the present invention can select the implementation scheme of appropriate number and location from the candidate base stations, while improving the resource utilization rate of the network element of the positioning management function, and improving the precision and accuracy of 5G positioning coordinates.

[0144] Fig.11 It is a module schematic diagram of the permutation and combination module in the embodiment of the base station selection system based on clustering algorithm of the present invention. Fig.12 It is a module schematic diagram of a coordinate clustering module in an embodiment of a base station selection system based on a clustering algorithm of the present invention. Fig.13It is a module schematic diagram of a positioning coordinate module in an embodiment of a base station selection system based on a clustering algorithm of the present invention. Figures 11 to 13 As shown, in Fig.10 On the basis of the device embodiment, the base station selection system based on the clustering algorithm of the present invention also includes a base station filtering module 51, and the permutation and combination module 52 is replaced by the base station grouping module 521, the category sum module 522, and the mapping relationship module 523. The coordinate clustering module 53 is replaced by the initial clustering module 531, the distance calculation module 532, the coordinate allocation module 533, and the iterative clustering module 534. The positioning coordinate module 54 is replaced by the optimal clustering module 541, the coordinate averaging module 542, the coordinate positioning module 543, and the base station combination module 544. The following is a description of each module:

[0145] The base station filtering module 51 is configured to obtain an initial base station set based on the position of the target to be located, and use an isolation forest algorithm to filter isolated base stations in the initial base station set according to the strength of the reference signal of the initial base station. If there are no redundant base stations, the positioning coordinates and the corresponding positioning base station combination are obtained; if there are still redundant base stations, the subsequent modules are executed.

[0146] The base station grouping module 521 is configured to obtain corresponding initial base stations for permutation and combination based on the preset number of base stations in each group h, where the value range of the number of base stations is [3, M], and the number of base stations h is different.

[0147]

[0148] Wherein, M represents the total number of initial base stations; P represents the total number of isolated base stations filtered.

[0149] A category summing module 522 is configured to obtain a sum W of permutation and combination categories obtained with different numbers of base stations h;

[0150] The mapping relationship module 523 is configured to obtain positioning coordinates for each permutation and combination based on the target to be located, and establish a mapping relationship between the positioning coordinates and the permutation and combination.

[0151] The initial clustering module 531 is configured to select the first initial clustering center in the positioning coordinate set according to the density peak value; including: obtaining the local density formula positioning ρ i It means that the circle with point i as the center is contained in the preset radius d. c The number of points in the circle, d ij is the distance between point j and point i in the positioning coordinate set, χ(x) indicates that the distance to data object i is less than the preset radius d cThe number of data objects, when x<0, then χ(x)=1, when x≥0, then χ(x)=0; the location coordinate with the largest local density is taken as the first initial clustering center.

[0152] The distance calculation module 532 is configured to calculate the distance between each of the remaining positioning coordinates in the positioning coordinate set and the initial cluster center and the sum of the distances;

[0153] The coordinate allocation module 533 is configured to use the positioning coordinate farthest from the initial cluster center as the next initial cluster center, the value range of the number of clusters k is [2, T], and each positioning coordinate is allocated to the nearest cluster; T is the smallest natural number greater than the root result of W;

[0154] The iterative clustering module 534 is configured to evaluate the classification accuracy of the clustering results, select the clustering result with the best evaluation result through iteration as the final clustering result, and regard the value of the number of clusters k corresponding to the final clustering result as the optimal number of clusters. It includes:

[0155] The clustering algorithm evaluation index DBI is used for evaluation.

[0156]

[0157] Among them, avg(C i ) is C i The average Euclidean distance from a class sample to its class center, avg(C j ) is C j The average Euclidean distance from a class sample to its class center, d cen (u i ,u j ) is the C i and C j The Euclidean distance of the class center of the class;

[0158] When the DBI value is smaller, the dispersion is lower and the clustering result is better;

[0159] By iterating each round k=k+1, the evaluation results of the number of clusters k under different values ​​are obtained;

[0160] When k=T, the clustering result with the best evaluation result (the clustering result with the smallest DBI value) is selected as the final clustering result, and the value of the number of clusters k corresponding to the final clustering result is regarded as the optimal number of clusters.

[0161] The optimal clustering module 541 is configured to perform clustering according to the optimal number of clusters, and select the cluster with the largest number of positioning coordinates in the clusters as the optimal cluster.

[0162] The coordinate averaging module 542 is configured to obtain an average value of all positioning coordinates in the optimal cluster.

[0163] The coordinate positioning module 543 is configured to use the positioning coordinates closest to the average value as the positioning coordinates of the target to be positioned.

[0164] The base station combination module 544 is configured to obtain the arrangement and combination of initial base stations corresponding to the positioning coordinates according to the mapping relationship.

[0165] The implementation principles of the above steps can be found in the relevant introduction of the base station selection method based on clustering algorithm, which will not be repeated here.

[0166] The embodiment of the present invention further provides a base station selection device based on a clustering algorithm, comprising a processor and a memory, wherein executable instructions of the processor are stored, wherein the processor is configured to execute the steps of a base station selection method based on a clustering algorithm by executing the executable instructions.

[0167] As shown above, the base station selection system based on the clustering algorithm of this embodiment of the present invention can select an implementation scheme with a suitable number and position from the candidate base stations, thereby improving the resource utilization rate of the positioning management function network element and enhancing the precision and accuracy of the 5G positioning coordinates.

[0168] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as systems, methods or program products. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, which may be collectively referred to herein as "circuits", "modules" or "platforms".

[0169] Fig.14 Schematic diagram of the base station selection device based on clustering algorithm of the present invention. Fig.14 An electronic device 600 according to this embodiment of the present invention is described. Fig.14 The electronic device 600 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0170] like Fig.14 As shown, the electronic device 600 is in the form of a general computing device. The components of the electronic device 600 may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0171] The storage unit stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 executes the steps of various exemplary embodiments of the present invention described in the electronic prescription circulation processing method section of this specification. For example, the processing unit 610 can execute the following steps: Figure 1 Follow the steps shown in .

[0172] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0173] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: a processing system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0174] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0175] The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 650. Furthermore, the electronic device 600 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 via a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0176] The embodiment of the present invention also provides a computer-readable storage medium for storing a program, and the steps of the base station selection method based on the clustering algorithm are implemented when the program is executed. In some possible implementations, various aspects of the present invention can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps of various exemplary embodiments of the present invention described in the above-mentioned electronic prescription circulation processing method section of this specification.

[0177] As shown above, the base station selection system based on the clustering algorithm of this embodiment of the present invention can select an implementation scheme with a suitable number and position from the candidate base stations, thereby improving the resource utilization rate of the positioning management function network element and enhancing the precision and accuracy of the 5G positioning coordinates.

[0178] The program product 800 for implementing the above method according to an embodiment of the present invention may adopt a portable compact disk read-only memory (CD-ROM) and include program code, and may be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto, and in this document, a readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device.

[0179] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0180] Computer readable storage media may include data signals propagated in baseband or as part of a carrier wave, wherein readable program codes are carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program codes contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0181] Program code for performing the processes of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, and the like, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0182] In summary, the purpose of the present invention is to provide a base station selection method, system, device and storage medium based on a clustering algorithm, which can select an appropriate number and location from the candidate base stations, thereby improving the resource utilization rate of the positioning management function network element and improving the precision and accuracy of 5G positioning coordinates.

[0183] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A base station selection method based on clustering algorithm, characterized in that: The following steps are involved: An initial base station set is obtained based on the position of the target to be located, and an isolation forest algorithm is used to filter isolated base stations in the initial base station set according to the strength of the reference signal of the initial base station. If there is no redundant base station, the positioning coordinates and the corresponding positioning base station combination are obtained; If there are still redundant base stations, execute the subsequent steps; Based on the preset number of base stations h in each group, the value range of the number of base stations is [3, M], and the corresponding initial base stations are obtained according to different values ​​of the number of base stations h for permutation and combination; Obtaining a sum W of permutation and combination types obtained by different numbers of base stations h; obtaining positioning coordinates for each permutation and combination based on the target to be located, and establishing a mapping relationship between the positioning coordinates and the permutation and combination; Select the first initial cluster center in the positioning coordinate set according to the density peak value; calculate the distance from each of the remaining positioning coordinates in the positioning coordinate set to the initial cluster center and the sum of the distances; take the positioning coordinate farthest from the initial cluster center as the next initial cluster center, the value range of the number of clusters k is [2, T], and assign each positioning coordinate to the nearest cluster; T is the smallest natural number greater than the root result of W; The clustering results are evaluated for classification accuracy, and the clustering result with the best evaluation result is selected as the final clustering result through iteration, and the value of the number of clusters k corresponding to the final clustering result is regarded as the optimal number of clusters; According to the clustering with the optimal number of clusters k, the coordinate mean of the cluster with the largest number of positioning coordinates matches the nearest positioning coordinate, and the positioning coordinate of the target to be positioned and the corresponding positioning base station combination are obtained according to the positioning coordinate.

2. The base station selection method based on clustering algorithm according to claim 1, characterized in that: The number of base stations h of each group is preset, the value range of the number of base stations is [3, M], and the corresponding initial base stations are obtained according to different values ​​of the number of base stations h for permutation and combination, including: Wherein, M represents the total number of initial base stations; P represents the total number of isolated base stations filtered.

3. The base station selection method based on clustering algorithm as claimed in claim 1, characterized in that: The selecting a first initial cluster center in the positioning coordinate set according to the density peak value includes: Get local density formula positioning ρ i It means that the circle with point i as the center is contained in the preset radius d. c The number of points in the circle, d ij is the distance between point j and point i in the positioning coordinate set, χ(x) indicates that the distance to data object i is less than the preset radius d c The number of data objects, when x<0, then χ(x)=1, when x≥0, then χ(x)=0; The location coordinates with the largest local density are taken as the first initial cluster center.

4. The base station selection method based on clustering algorithm as claimed in claim 1, characterized in that: The clustering results are evaluated for classification accuracy, the clustering result with the best evaluation result is selected as the final clustering result through iteration, and the value of the number of clusters k corresponding to the final clustering result is regarded as the optimal number of clusters, including: The clustering algorithm evaluation index DBI is used for evaluation. Among them, avg(C i ) is C i The average Euclidean distance from a class sample to its class center, avg(C j ) is C j The average Euclidean distance from a class sample to its class center, d cen (u i ,u j ) is the C i and C j The Euclidean distance of the class center of the class; When the DBI value is smaller, the dispersion is lower and the clustering result is better; By iterating each round k=k+1, the evaluation results of the number of clusters k under different values ​​are obtained; When k=T, the clustering result with the best evaluation result is selected as the final clustering result, and the value of the number of clusters k corresponding to the final clustering result is regarded as the optimal number of clusters.

5. The base station selection method based on clustering algorithm according to claim 1, characterized in that: In the clustering according to the optimal number of clusters k, the coordinate mean of the cluster with the largest number of positioning coordinates in the cluster matches the nearest positioning coordinate, and the positioning coordinates of the target to be positioned and the corresponding positioning base station combination are obtained according to the positioning coordinates, including: Perform clustering according to the optimal number of clusters, and select the cluster with the largest number of positioning coordinates as the optimal cluster; Obtaining the average value of all positioning coordinates in the optimal cluster; The positioning coordinates closest to the average value are used as the positioning coordinates of the target to be positioned.

6. The base station selection method based on clustering algorithm as claimed in claim 5, characterized in that: In the clustering according to the optimal number of clusters k, the coordinate mean of the cluster with the largest number of positioning coordinates in the cluster matches the nearest positioning coordinate, and the positioning coordinates of the target to be positioned and the corresponding positioning base station combination are obtained according to the positioning coordinates, further comprising: The arrangement and combination of the initial base stations corresponding to the positioning coordinates are obtained according to the mapping relationship.

7. A base station selection system based on clustering algorithm, characterized in that: include: The base station filtering module is configured to obtain an initial base station set based on the position of the target to be positioned, and use an isolation forest algorithm to filter isolated base stations in the initial base station set according to the strength of the reference signal of the initial base station, and obtain positioning coordinates and corresponding positioning base station combinations if there are no redundant base stations; if there are still redundant base stations, execute subsequent steps; A permutation and combination module, based on a preset number of base stations in each group h, where the value range of the number of base stations is [3, M], obtains the corresponding initial base stations for permutation and combination according to different values ​​of the number of base stations h; Obtaining a sum W of permutation and combination types obtained by different numbers of base stations h; obtaining positioning coordinates for each permutation and combination based on the target to be located, and establishing a mapping relationship between the positioning coordinates and the permutation and combination; The coordinate clustering module selects the first initial cluster center in the positioning coordinate set according to the density peak value; calculates the distance from each of the remaining positioning coordinates in the positioning coordinate set to the initial cluster center and the sum of the distances; takes the positioning coordinate farthest from the initial cluster center as the next initial cluster center, the value range of the number of clusters k is [2, T], and assigns each positioning coordinate to the nearest cluster; T is the smallest natural number greater than the root result of W; The clustering results are evaluated for classification accuracy, and the clustering result with the best evaluation result is selected as the final clustering result through iteration, and the value of the number of clusters k corresponding to the final clustering result is regarded as the optimal number of clusters; The positioning coordinate module matches the coordinate mean of the cluster with the largest number of positioning coordinates in the cluster according to the clustering of the optimal number of clusters k with the nearest positioning coordinate, and obtains the positioning coordinate of the target to be positioned and the corresponding positioning base station combination according to the positioning coordinate.

8. A base station selection device based on a clustering algorithm, characterized in that: include: processor; a memory storing executable instructions of the processor; The processor is configured to execute the steps of the base station selection method based on the clustering algorithm in any one of claims 1 to 6 by executing the executable instructions.

9. A computer-readable storage medium for storing a program, characterized in that: When the program is executed by a processor, the steps of the base station selection method based on the clustering algorithm described in any one of claims 1 to 6 are implemented.

Images