A weak coverage area grid clustering method and apparatus
By slicing and probabilistically calculating the raster map, the regions to be clustered are selected, achieving efficient clustering of weakly covered areas, improving real-time performance and update speed, and solving the problems of large computational load and poor real-time performance in existing technologies.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA MOBILE GROUP DESIGN INST
- Filing Date
- 2021-04-30
- Publication Date
- 2026-07-10
AI Technical Summary
Existing technologies suffer from high computational cost and poor real-time performance when clustering weakly covered areas in massive datasets. Furthermore, updating raster coverage data requires traversing the entire map, which also impacts real-time performance.
By tiling the rasterized map, calculating the weak coverage probability of each map tile, filtering the map tiles to be clustered based on the probability, performing clustering processing, and finally updating the rasterized map.
It reduces the consumption of computing resources, improves the real-time performance and update speed of positioning in weak coverage areas, and solves the problems of high computing power requirements and high resource consumption for real-time updates.
Smart Images
Figure CN115272731B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of communication technology, and more specifically to a method and apparatus for grid clustering of weak coverage areas. Background Technology
[0002] Currently, with the continuous advancement of mobile communication network construction, the location of weak coverage areas is crucial to the quality of network construction. Rasterized maps, as an effective map data digitization method, can effectively support the clustering and location of weak coverage areas. Rapidly locating weak coverage areas from massive amounts of raster coverage data is an important task. However, current clustering operations for weak coverage areas suffer from high computational load and poor real-time performance. This is because, based on existing data resources, weak coverage areas are identified, extracted, and delineated according to certain standards to obtain their outlines. However, existing raster feature extraction and clustering algorithms mainly focus on traversing the entire rasterized map to find problem raster areas and complete clustering, which places enormous demands on computing power when dealing with massive amounts of data. Furthermore, since raster coverage data is not static but constantly updated, generating new clustering results often requires traversing the entire rasterized map; a single coverage data update may necessitate a complete traversal, impacting the real-time performance of obtaining weak coverage results. Summary of the Invention
[0003] In view of the above problems, the present invention is proposed to provide a weak coverage area grid clustering method and apparatus to overcome or at least partially solve the above problems.
[0004] According to one aspect of the present invention, a weak coverage area raster clustering method is provided, comprising:
[0005] The rasterized map is sliced to obtain multiple map tiles;
[0006] For each map tile, calculate the probability of weak coverage of that map tile based on the coverage data corresponding to that map tile;
[0007] Based on the weak coverage probability of multiple map tiles, map tiles to be clustered are selected from multiple map tiles, and the rasters in the map tiles to be clustered are clustered to obtain weak coverage clusters.
[0008] Collect updated coverage data corresponding to multiple map tiles, and update the rasterized map based on the updated coverage data and the weak coverage clusters.
[0009] According to another aspect of the present invention, a weak coverage area grid clustering device is provided, comprising:
[0010] The tiling module is used to tile rasterized maps to obtain multiple map tiles;
[0011] The calculation module is used to calculate the weak coverage probability of each map tile based on the coverage data corresponding to that map tile.
[0012] The clustering module is used to filter map tiles to be clustered from multiple map tiles based on the weak coverage probability of multiple map tiles, and to perform clustering processing on the raster in the map tiles to be clustered to obtain weak coverage clusters.
[0013] The update module is used to collect update coverage data corresponding to multiple map tiles, and update the raster map based on the update coverage data and the weak coverage clusters.
[0014] According to another aspect of the present invention, a computing device is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus;
[0015] The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the weak coverage area grid clustering method described above.
[0016] According to another aspect of the present invention, a computer storage medium is provided, the storage medium storing at least one executable instruction that causes a processor to perform an operation corresponding to the weak coverage area grid clustering method described above.
[0017] According to a method and apparatus for weak coverage area raster clustering of the present invention, multiple map tiles are obtained by slicing a rasterized map; for each map tile, the weak coverage probability of the map tile is calculated based on the coverage data corresponding to that map tile; based on the weak coverage probabilities of multiple map tiles, map tiles to be clustered are selected from the multiple map tiles, and the rasters in the map tiles to be clustered are clustered to obtain weak coverage clusters; updated coverage data corresponding to multiple map tiles are collected, and the rasterized map is updated based on the updated coverage data and the weak coverage clusters. The present invention classifies the rasterized map by slicing the rasterized map and calculating the weak coverage probability of each map tile, and selects map tiles to be clustered, achieving tiled management and differentiated processing. It avoids the need for complete traversal and clustering of the entire rasterized map, thereby saving computing resources, effectively improving the update speed of the rasterized map, and solving the problems of high computing power requirements for current clustering operations and high resource consumption for real-time map updates.
[0018] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description
[0019] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:
[0020] Figure 1 A flowchart of a weak coverage area grid clustering method provided by an embodiment of the present invention is shown;
[0021] Figure 2 This diagram illustrates the structure of a weak coverage area grid clustering device provided in an embodiment of the present invention.
[0022] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present invention is shown. Detailed Implementation
[0023] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
[0024] Figure 1 A flowchart of an embodiment of weak coverage area grid clustering according to the present invention is shown, as follows: Figure 1 As shown, the method includes the following steps:
[0025] Step S110: Perform tile processing on the rasterized map to obtain multiple map tiles.
[0026] In this step, the entire rasterized map is sliced evenly according to the number of graticules, so that each map slice has the same number of graticules. Those skilled in the art can set the size of the map slices according to actual needs, for example, each map slice may include 25 graticules. The sliced rasterized map can also be marked with a special color, and graticules outside the scope of the rasterized map can be marked with "Null".
[0027] Step S120: For each map tile, calculate the weak coverage probability of the map tile based on the coverage data corresponding to the map tile.
[0028] In an optional approach, step S120 further includes: analyzing the coverage data corresponding to the map tile to obtain antenna data within a preset range of the map tile and building occlusion data corresponding to the map tile; and using a trained weak coverage probability recognition model to process the antenna data and building occlusion data to obtain the weak coverage probability of the map tile.
[0029] Specifically, for each map tile, the coverage data corresponding to the map tile is analyzed to obtain the antenna data within a preset range from the map tile. Specifically, the center position of the map tile, as well as the antenna positions and number of antennas within a preset distance range from the center position, can be read. Furthermore, the building occlusion data corresponding to the map tile can be obtained, where the building occlusion data can be obtained based on empirical values. Combined with the wireless propagation loss model, a trained weak coverage probability identification model can be obtained by training based on sample data (such as sample antenna data and sample building occlusion data). Using this weak coverage probability identification model, the antenna data and building occlusion data are processed to obtain the weak coverage probability of the map tile. A two-dimensional variable (li, bj) can be established through the antenna data and building occlusion data, where, for antennas within a preset distance L from the center position of the map tile, li represents the distance between the antenna and the center position of the map tile, and i represents the antenna number; bj represents the building occlusion data, which can be represented by the following formula (1):
[0030] bj = L(lj, sj); (1)
[0031] Among them, for buildings within a preset distance L from the center of the map tile, j represents the building number, bj can be obtained from experience, bj is positively correlated with the distance lj between the building and the center location on the propagation path, the occlusion area sj, etc., and the above function (1) is constructed.
[0032] During the training process of the weak coverage probability identification model, two-dimensional variables (li, bj) corresponding to the sample data can be obtained based on the sample data. Combined with the wireless propagation loss model, a two-dimensional variable probability distribution function F is constructed. Through continuous training, the function configuration and function variables are fitted to obtain the weak coverage probability identification model. After obtaining the weak coverage probability identification model, when it is necessary to calculate the weak coverage probability of a certain map tile, two-dimensional variables (li, bj) can be constructed based on the antenna data and building obstruction data corresponding to that map tile. These two-dimensional variables are then input into the weak coverage probability identification model, which processes the data and outputs the weak coverage probability of that map tile.
[0033] Step S130: Based on the weak coverage probability of multiple map tiles, select map tiles to be clustered from multiple map tiles, and perform clustering processing on the rasters in the map tiles to be clustered to obtain weak coverage clusters.
[0034] In an optional manner, step S130 further includes: classifying map tiles with a weak coverage probability less than or equal to a first preset threshold into a first category, classifying map tiles with a weak coverage probability greater than the first preset threshold and less than a second preset threshold into a second category, and classifying map tiles with a weak coverage probability greater than or equal to the second preset threshold into a third category; wherein the second preset threshold is greater than the first preset threshold; and using map tiles in the second category and map tiles in the third category as map tiles to be clustered.
[0035] Specifically, to reduce the computational load of clustering and thus save computing resources, a first preset threshold A and a second preset threshold B are set, where A is the threshold for distinguishing whether an area is weakly covered, and B is the threshold for distinguishing the probability of weak coverage, with B > A. Map tiles are divided into three categories: map tiles with a weak coverage probability less than or equal to A are identified as non-weakly covered areas and assigned to the first category; map tiles with a weak coverage probability greater than A and less than B are identified as areas with low weak coverage probability and assigned to the second category; and map tiles with a weak coverage probability greater than or equal to B are identified as areas with high weak coverage probability and assigned to the third category. Since map tiles in the second and third categories may both be weakly covered areas, they are identified as map tiles to be clustered. Map tiles in the first category are non-weakly covered areas and therefore do not require clustering.
[0036] In an optional manner, step S130 further includes: for each map tile to be clustered, selecting an initial access raster from the map tile to be clustered, accessing other rasters in the map tile to be clustered starting from the initial access raster, and designating rasters with coverage data lower than the coverage threshold data as weak coverage rasters; and using a density clustering algorithm to cluster the accessed rasters to obtain weak coverage clusters.
[0037] In one alternative approach, before clustering the rasters in the map tiles to be clustered to obtain weakly covered clusters, the method further includes: merging the rasters in the map tiles of the third category according to a preset merging rule.
[0038] Specifically, to further reduce computational load, map tiles in the second category (areas with low coverage probability) are clustered based on their original raster size; while map tiles in the third category (areas with high coverage probability) are raster-merged according to a preset merging rule before being clustered. The specific preset merging rule is as follows: map tiles in the third category are re-rasterized, with the side length of the newly formed merged raster being two or three times that of the original raster. That is, the new raster consists of 4 or 9 original raster cells, ensuring that the signal received strength of each raster in the newly formed merged raster of the map tiles in the third category is the average of the signal received strength values of the multiple original raster cells that make up the new raster.
[0039] Furthermore, the initial access grid is accessed from the map tiles of the second and third categories. Starting from the initial access grid, other grids in the map tiles to be clustered are accessed. Grids with coverage data below the coverage threshold are designated as weak coverage grids. The initial access grid can be selected as the grid at the center point of the line connecting multiple antenna locations or base station locations. If there are no antennas or base stations in the map tile area, the midpoint of the line connecting the intersection of the four vertices and the diagonal of the map tile is selected as the initial access grid. When accessing other grids in the map tile from the initial access grid, the grids that have been accessed are marked as "accessed". For grids with coverage data below the coverage threshold, they are marked as "weak coverage grids". The specific marking method can be achieved by changing the color of the grid. For the grids marked as weak coverage, density clustering and other clustering operations are performed to obtain the weak coverage clusters of this cluster.
[0040] Currently, the main clustering algorithms used include K-Nearest Neighbor (KNN) classification algorithm, weighted K-Nearest Neighbor (WKNN) algorithm, density-based spatial clustering of applications with noise (DBSCAN) algorithm, and hierarchical clustering algorithm. The general clustering process starts with the selection of the initial clustering grid, and performs clustering operations on the grids with covered data on the raster map using relevant algorithms. After the entire raster map is processed, clusters that meet the criteria of the problem area are obtained. In this embodiment, the density-based clustering algorithm mentioned above is applied. Taking the DBSCAN clustering algorithm mentioned above as an example, the clustering operation of weakly covered grids is performed. Several important definitions are as follows (1)-(5):
[0041] (1) The two initial parameters are E (neighborhood radius) and minPts (minimum number of points in neighborhood E).
[0042] (2) Core Raster: A given raster is a core raster if the number of weakly covered rasters within its E-neighborhood is greater than or equal to minPts. Boundary Raster: A raster is a boundary raster if it contains fewer than minPts of weakly covered rasters within its neighborhood radius E, but falls within the neighborhood of a core raster. Noise Raster: A raster is a noise raster if it is neither a core raster nor a boundary raster.
[0043] (3) Direct density reachability: For a sample set D, if grid q is within the E neighborhood of grid p and p is the core grid, then grid q is said to be directly density reachable from grid p.
[0044] (4) Density reachable: For a sample set D, given a sequence of samples p1, ..., p i , ..., p n Where p = p1, q = p n If the grid p i From p i If the density of -1 is directly attainable, then the density of grid q is attainable from the density of grid p.
[0045] (5) Density-connected: If there exists o in the sample set D, and grid o is density-reachable to grid p and grid q, then p and q are density-connected. The purpose of DBSCAN is to find the largest set of density-connected grids.
[0046] The basic steps for DBSCAN clustering operations on rasters marked as weakly covered are as follows (1)-(4):
[0047] The minPts setting can be set to half the standard number of rasters in a weakly covered region that are identified as weakly covered rasters.
[0048] (1) Starting with the initial access raster, the raster is accessed and a core raster is found according to the core raster definition. A cluster A is formed according to the set E and minPts, and the raster in the cluster is added to the candidate set Ca. If the initial access raster is a boundary raster or a noisy raster, its adjacent raster is selected until a new cluster is established or the traversal loop is exited. The traversed raster is marked as "accessed". For rasters with coverage data below the coverage threshold, they are marked as "weakly covered raster". Based on this marking, the raster that has undergone clustering operation is marked as "processed" again, and the raster that has not undergone clustering operation is marked as "unprocessed". The specific marking method is not described in detail.
[0049] (2) Process any unvisited raster x in the candidate raster set Ca, and check whether its neighborhood E contains minPts weakly covered rasters. If it does, add these rasters to Ca for further processing. If x does not belong to any cluster, add x to cluster A.
[0050] (3) Repeat step (2) until all elements in Ca have been processed.
[0051] (4) Repeat steps (1)-(3) until all grids in the entire map tile are processed to form corresponding clusters. Merge objects according to the density reachability rule and merge clusters.
[0052] In an alternative approach, the method further includes: if the direct density of weak cover raster in the weak cover cluster at the boundary of any map tile to be clustered is as high as that of the weak cover cluster at the boundary of a neighboring map tile, then the weak cover cluster at the boundary of the map tile to be clustered is merged with the weak cover cluster at the boundary of a neighboring map tile.
[0053] Specifically, for a map tile that has undergone clustering, if the direct density of weakly covered raster cells in a weakly covered cluster at the boundary of a map tile is equal to that of a weakly covered cluster at the boundary of a neighboring map tile, then the weakly covered cluster at the boundary of the map tile to be clustered is merged with the weakly covered cluster at the boundary of the neighboring map tiles to obtain a new cluster. At this point, the first traversal of the entire rasterized map is complete, and all weakly covered clusters are determined.
[0054] In an alternative approach, for all the rasters within the identified weakly covered clusters, the DBSCAN clustering operation is performed again with a number of rasters greater than 1 / 2 minPts to obtain a more detailed boundary profile.
[0055] Step S140: Collect update coverage data corresponding to multiple map tiles, and update the raster map based on the update coverage data and weak coverage clusters.
[0056] In an alternative approach, step S140 further includes: re-clustering the rasters in the weak coverage clusters and the rasters in the neighboring map tiles of the weak coverage clusters based on the updated coverage data; and updating the rasterized map based on the clustering results.
[0057] Since the overlay data within the rasterized map may be updated at any time, updated overlay data corresponding to multiple map tiles are collected periodically. When the overlay data is updated, based on the updated overlay data, it is only necessary to re-cluster the rasters in the weak overlay clusters and the rasters in the neighboring map tiles of the weak overlay clusters; and then add the updated labeling results to the entire rasterized map.
[0058] The method provided in this embodiment improves the real-time performance of locating weak coverage areas in rasterized maps by slicing the rasterized map and calculating the weak coverage probability of each slice. By classifying the rasterized map and selecting slices to be clustered, it achieves slice management and differentiated processing, eliminating the need for complete traversal and clustering of the entire rasterized map, thus saving computing resources and effectively improving the update speed of the rasterized map. This solves the problems of high computing power requirements for current clustering operations and high resource consumption for real-time map updates.
[0059] Figure 2 A schematic diagram of an embodiment of a weak coverage area grid clustering device of the present invention is shown. Figure 2 As shown, the device includes: a slicing module 210, a calculation module 220, a clustering module 230, and an update module 240.
[0060] The tiling module 210 is used to perform tiling processing on the rasterized map to obtain multiple map tiles.
[0061] The calculation module 220 is used to calculate the weak coverage probability of each map tile based on the coverage data corresponding to that map tile.
[0062] In an alternative approach, the calculation module 220 is further configured to: analyze the coverage data corresponding to the map tile to obtain antenna data within a preset range of the map tile and building occlusion data corresponding to the map tile; and use a trained weak coverage probability recognition model to process the antenna data and building occlusion data to obtain the weak coverage probability of the map tile.
[0063] The clustering module 230 is used to filter map tiles to be clustered from multiple map tiles based on the weak coverage probability of multiple map tiles, and to perform clustering processing on the rasters in the map tiles to be clustered to obtain weak coverage clusters.
[0064] In an optional manner, the clustering module 230 is further configured to: classify map tiles with a weak coverage probability less than or equal to a first preset threshold into a first category, classify map tiles with a weak coverage probability greater than the first preset threshold and less than a second preset threshold into a second category, and classify map tiles with a weak coverage probability greater than or equal to the second preset threshold into a third category; wherein the second preset threshold is greater than the first preset threshold; and use the map tiles in the second category and the map tiles in the third category as map tiles to be clustered.
[0065] In an alternative embodiment, the clustering module 230 is further configured to: for each map tile to be clustered, select an initial access raster from the map tile to be clustered, access other rasters in the map tile to be clustered starting from the initial access raster, and designate rasters with coverage data lower than the coverage threshold data as weak coverage rasters; and use a density clustering algorithm to cluster the accessed rasters to obtain weak coverage clusters.
[0066] In an alternative approach, the clustering module 230 is further configured to: merge the rasters in the map tiles of the third category according to a preset merging rule.
[0067] In an optional manner, the clustering module 230 is further configured to: if the direct density of the weak cover raster in the weak cover cluster at the boundary of any map tile to be clustered can reach the weak cover cluster at the boundary of the neighboring map tile, then merge the weak cover cluster at the boundary of the map tile to be clustered with the weak cover cluster at the boundary of the neighboring map tile.
[0068] The update module 240 is used to collect update coverage data corresponding to multiple map tiles, and update the raster map based on the update coverage data and weak coverage clusters.
[0069] In one alternative approach, based on the updated coverage data, the rasters in the weak coverage clusters and the rasters in the neighboring map tiles of the weak coverage clusters are re-clustered; and the rasterized map is updated based on the clustering results.
[0070] The apparatus provided in this embodiment improves the real-time performance of locating weak coverage areas in rasterized maps by slicing the rasterized map and calculating the weak coverage probability of each slice. By classifying the rasterized map and selecting slices to be clustered, it achieves slice management and differentiated processing, eliminating the need for complete traversal and clustering of the entire rasterized map, thus saving computing resources and effectively improving the update speed of the rasterized map. This solves the problems of high computing power requirements for current clustering operations and high resource consumption for real-time map updates.
[0071] This invention provides a non-volatile computer storage medium storing at least one executable instruction that can execute the weak coverage area raster clustering method in any of the above method embodiments.
[0072] Executable instructions can specifically be used to cause the processor to perform the following operations:
[0073] The rasterized map is sliced to obtain multiple map tiles;
[0074] For each map tile, calculate the probability of weak coverage for that map tile based on the coverage data corresponding to that map tile;
[0075] Based on the weak coverage probability of multiple map tiles, map tiles to be clustered are selected from multiple map tiles, and the rasters in the map tiles to be clustered are clustered to obtain weak coverage clusters.
[0076] Collect updated coverage data corresponding to multiple map tiles, and update the raster map based on the updated coverage data and weak coverage clusters.
[0077] Figure 3 The diagram shows a structural schematic of an embodiment of the computing device of the present invention. The specific embodiments of the present invention do not limit the specific implementation of the computing device.
[0078] like Figure 3 As shown, the computing device may include:
[0079] Processor, Communications Interface, Memory, and Communications Bus.
[0080] The processor, communication interface, and memory communicate with each other via a communication bus. The communication interface is used to communicate with other network elements, such as clients or other servers. The processor executes programs, specifically the relevant steps described in the weak coverage area raster clustering method embodiment.
[0081] Specifically, the program may include program code, which includes computer operation instructions.
[0082] The processor may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The server may include one or more processors of the same type, such as one or more CPUs; or it may include processors of different types, such as one or more CPUs and one or more ASICs.
[0083] Memory is used to store programs. Memory may include high-speed RAM, and may also include non-volatile memory, such as at least one disk drive.
[0084] Specifically, the program can be used to cause the processor to perform the following operations:
[0085] The rasterized map is sliced to obtain multiple map tiles;
[0086] For each map tile, calculate the probability of weak coverage for that map tile based on the coverage data corresponding to that map tile;
[0087] Based on the weak coverage probability of multiple map tiles, map tiles to be clustered are selected from multiple map tiles, and the rasters in the map tiles to be clustered are clustered to obtain weak coverage clusters.
[0088] Collect updated coverage data corresponding to multiple map tiles, and update the raster map based on the updated coverage data and weak coverage clusters.
[0089] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, the embodiments of the present invention are not directed to any particular programming language. It should be understood that the content of the invention described herein can be implemented using various programming languages, and the above description of specific languages is for the purpose of disclosing the best mode of implementation of the invention.
[0090] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.
[0091] Similarly, it should be understood that, in order to simplify the invention and aid in understanding one or more of the various inventive aspects, features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the above description of exemplary embodiments of the invention. However, this disclosure should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the following claims, inventive aspects lie in fewer than all features of a single foregoing disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into this detailed description, wherein each claim itself is a separate embodiment of the invention.
[0092] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.
[0093] Furthermore, those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not others, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments can be used in any combination.
[0094] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components according to the embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.
[0095] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.
Claims
1. A raster clustering method for weakly covered areas, characterized in that, include: The rasterized map is sliced uniformly according to the number of grid cells to obtain multiple map slices; Each map tile consists of multiple grids; For each map tile, calculate the probability of weak coverage for that map tile based on the coverage data corresponding to that map tile; Based on the weak coverage probability of multiple map tiles, map tiles to be clustered are selected from multiple map tiles, and the rasters in the map tiles to be clustered are clustered to obtain weak coverage clusters. Collect updated coverage data corresponding to multiple map tiles, and update the rasterized map based on the updated coverage data and the weak coverage clusters; The step of calculating the weak coverage probability of each map tile based on the coverage data corresponding to that map tile further includes: Analyze the coverage data corresponding to the map tile to obtain antenna data within a preset range of the map tile and building occlusion data corresponding to the map tile; The weak coverage probability identification model is trained to process the antenna data and the building occlusion data to obtain the weak coverage probability of the map tile. In the model training process of the weak coverage probability identification model, corresponding two-dimensional variables are obtained based on the sample antenna data and sample building occlusion data. Based on the wireless propagation loss model, a two-dimensional variable probability distribution function is constructed. The function configuration and function variables are fitted through training to obtain the weak coverage probability identification model.
2. The method according to claim 1, characterized in that, The step of selecting map tiles to be clustered from multiple map tiles based on the weak coverage probability of multiple map tiles further includes: Map tiles with a weak coverage probability less than or equal to a first preset threshold are classified into a first category; map tiles with a weak coverage probability greater than the first preset threshold and less than a second preset threshold are classified into a second category; and map tiles with a weak coverage probability greater than or equal to the second preset threshold are classified into a third category; wherein, the second preset threshold is greater than the first preset threshold. Map tiles from the second category and map tiles from the third category are selected as map tiles to be clustered.
3. The method according to claim 1, characterized in that, The process of clustering the raster cells in the map tile to be clustered to obtain weakly covered clusters further includes: For each map tile to be clustered, an initial access grid is selected from the map tile to be clustered. Starting from the initial access grid, other grids in the map tile to be clustered are accessed. Grids with coverage data lower than the coverage threshold data are regarded as weak coverage grids. Density clustering algorithm is used to cluster visited rasters to obtain weakly covered clusters.
4. The method according to claim 2, characterized in that, Before performing clustering processing on the rasters in the map tiles to be clustered to obtain weakly covered clusters, the method further includes: According to the preset merging rules, the grids in the map tiles of the third category are merged.
5. The method according to any one of claims 1-4, characterized in that, The method further includes: If the density of weak-coverage rasters in a weak-coverage cluster at the boundary of any map tile to be clustered can reach the density of a weak-coverage cluster at the boundary of a neighboring map tile, then the weak-coverage cluster at the boundary of the map tile to be clustered is merged with the weak-coverage cluster at the boundary of the neighboring map tile.
6. The method according to any one of claims 1-4, characterized in that, The step of updating the rasterized map based on the updated coverage data and the weak coverage clusters further includes: Based on the updated coverage data, the rasters in the weak coverage clusters and the rasters in the neighboring map tiles of the weak coverage clusters are re-clustered; The rasterized map is updated based on the clustering results.
7. A grid clustering device for weakly covered areas, characterized in that, include: The slicing module is used to slice the rasterized map evenly according to the number of grid cells, resulting in multiple map slices. Each map tile consists of multiple grids; The calculation module is used to calculate the weak coverage probability of each map tile based on the coverage data corresponding to that map tile. The clustering module is used to filter map tiles to be clustered from multiple map tiles based on the weak coverage probability of multiple map tiles, and to perform clustering processing on the raster in the map tiles to be clustered to obtain weak coverage clusters. The update module is used to collect update coverage data corresponding to multiple map tiles, and update the raster map based on the update coverage data and the weak coverage clusters; The calculation module is further used for: Analyze the coverage data corresponding to the map tile to obtain antenna data within a preset range of the map tile and building occlusion data corresponding to the map tile; The weak coverage probability identification model is trained to process the antenna data and the building occlusion data to obtain the weak coverage probability of the map tile. In the model training process of the weak coverage probability identification model, corresponding two-dimensional variables are obtained based on the sample antenna data and sample building occlusion data. Based on the wireless propagation loss model, a two-dimensional variable probability distribution function is constructed. The function configuration and function variables are fitted through training to obtain the weak coverage probability identification model.
8. A computing device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform the operation corresponding to the weak coverage area grid clustering method as described in any one of claims 1-6.
9. A computer storage medium, characterized in that, The storage medium stores at least one executable instruction that causes the processor to perform the operation corresponding to the weak coverage area grid clustering method as described in any one of claims 1-6.