Method for determining mapping relation between community and building and related equipment

Abstract

The invention discloses a method and device for determining a mapping relation between a community and a building, computing equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: determining the similarity between the name of a target building and the name of a building intended to be covered by an indoor cell in a target indoor cell cluster; determining a relative position relationship between the target building and the target indoor cell cluster; and when the similarity and the relative position relationship meet a preset mapping relationship establishment condition, establishing a first mapping relationship between the target building and the indoor cells in the target indoor cell cluster.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, device, computer-readable storage medium, and computer program product for determining the mapping relationship between a cell and a building. Background Technology

[0002] In the field of communications, "cell-to-building mapping" refers to establishing a precise association between communication cells and specific buildings and their internal spaces, clarifying which base station coverage area corresponds to which building or buildings, and which communication cell matches different areas within a building. The core purpose is to optimize communication network coverage and management.

[0003] Currently, determining the mapping relationship between residential communities and buildings largely relies on manual data entry and on-site surveys. Manual data entry is prone to errors due to human negligence, such as mistakenly associating an indoor community corresponding to "XX Business Building" with "XX Commercial Building." On-site surveys, on the other hand, suffer from inefficiency and high costs. In urban core areas with dense high-rise buildings and numerous indoor communities, a comprehensive survey often takes several weeks and is difficult to adapt to the dynamic adjustments required for community layouts. Therefore, there is an urgent need for an efficient and accurate method to determine the mapping relationship between residential communities and buildings to address the shortcomings of existing technologies.

[0004] How to efficiently and accurately determine the mapping relationship between communication cells and buildings is an urgent problem to be solved. Summary of the Invention

[0005] This application provides a method for determining the mapping relationship between a cell and a building, in order to solve the problem in the prior art of how to efficiently and accurately determine the mapping relationship between a communication cell and a building.

[0006] This application also provides an apparatus, device, computer-readable storage medium, and computer program product for determining the mapping relationship between a community and a building.

[0007] The embodiments of this application adopt the following technical solutions: A method for determining the mapping relationship between residential areas and buildings, comprising: Determine the similarity between the name of the target building and the name of the building that the indoor cells in the target indoor cell cluster intend to cover; Determine the relative positional relationship between the target building and the target indoor cell cluster; When the similarity and the relative positional relationship satisfy the preset mapping relationship establishment conditions, a first mapping relationship is established between the target building and the indoor cells in the target indoor cell cluster.

[0008] A device for determining the mapping relationship between residential areas and buildings, comprising: The similarity determination module is used to determine the similarity between the name of the target building and the name of the building that the indoor cells in the target indoor cell cluster intend to cover; A location relationship determination module is used to determine the relative location relationship between the target building and the target indoor cell cluster; The first mapping establishment module is used to establish a first mapping relationship between the target building and the indoor cells in the target indoor cell cluster when the similarity and the relative position relationship meet the preset mapping relationship establishment conditions.

[0009] A computing device includes: a memory and a processor, wherein, The memory is used to store computer programs; The processor, coupled to the memory, is used to execute the computer program stored in the memory for performing the methods described above.

[0010] A computer-readable storage medium storing a computer program that, when executed by a computer, enables the implementation of the above-described method.

[0011] A computer program product storing instructions that, when executed by a computer, cause the computer to perform the method described above.

[0012] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: By combining the dual dimensions of "name" and "location," the mapping relationship between buildings and communities can be accurately and automatically established, solving the problems of low efficiency and large errors in traditional methods. Attached Figure Description

[0013] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A flowchart illustrating the specific implementation of a method for determining the mapping relationship between a residential community and a building, provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating the implementation process of the method for extracting building names from cell names using an LSTM network in this embodiment of the application. Figure 3 This is a schematic diagram illustrating the implementation process of a rule-based method for extracting building names from community names. Figure 4This is a schematic diagram illustrating the use of ray discrimination to determine whether a target indoor cell cluster is located inside (or outside) the target building in an embodiment of this application. Figure 5 This is a schematic diagram illustrating the use of ray discriminant analysis to determine whether a target indoor cell cluster is located inside (or outside) a target building in a specific example. Figure 6 A schematic diagram of the specific structure of a device for determining the mapping relationship between a residential area and a building, provided in an embodiment of this application; Figure 7 This is a schematic diagram of the specific structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0015] As will be known to those skilled in the art, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0016] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements, but may include other elements not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0017] This application provides a method for determining the mapping relationship between a cell and a building, in order to solve the problem in the prior art of how to efficiently and accurately determine the mapping relationship between a communication cell and a building.

[0018] The subject executing this method can be any computing device capable of implementing the method, such as servers, mobile phones, personal computers, smart wearable devices, smart robots, drones, IoT devices, network element devices in mobile communication networks, etc.

[0019] Different steps of this method can be implemented by the same execution entity or by different execution entities. This application does not limit which execution entity is used to implement the method.

[0020] Furthermore, the embodiments of this application do not limit the execution order of different steps. When using the method provided in the embodiments of this application, the execution order of different steps can be adjusted according to actual needs.

[0021] For ease of description, the following uses a network element device in a mobile communication network as the execution subject of this method to provide a detailed description of the method provided in the embodiments of this application.

[0022] like Figure 1 The diagram shown is a flowchart illustrating a method for determining the mapping relationship between a residential community and buildings, as provided in this application embodiment. The method includes the following steps: Step 11: Determine the similarity between the name of the target building and the name of the building that the indoor cells in the target indoor cell cluster intend to cover; The target building is the building to which a mapping relationship needs to be established with a communication cell, but it is unclear which communication cells it should be mapped to.

[0023] The building mentioned in the embodiments of this application may refer to a specific building and / or the interior space of a specific building. For example, a specific building may be Financial Center Tower A, or Financial Center Tower A-1F.

[0024] A target indoor cell cluster refers to one or more clusters within "at least one predetermined indoor cell cluster." An indoor cell cluster is a set of indoor cells. The method for determining indoor cell clusters will be explained in detail later.

[0025] Currently, for the same building, there may be both outdoor and indoor base stations covering the building. Outdoor base stations refer to communication base stations deployed in the outdoor environment; indoor base stations generally refer to communication base stations deployed indoors to solve signal coverage problems inside buildings. In this application embodiment, the cell of the outdoor base station is referred to as an outdoor cell, and the cell of the indoor base station is referred to as an indoor cell.

[0026] In this embodiment of the application, indoor cell clusters can be obtained by clustering.

[0027] For example, in one alternative implementation, the indoor cell cluster can be obtained by the following steps: Step ①: First, obtain the names of each indoor cell; Through research on related technologies, the inventors discovered that, in order to distinguish indoor communities covering different buildings and to standardize community names, the naming rules for different indoor communities usually have certain similarities.

[0028] Specifically, while the order of information in the names of indoor cells located in different geographical areas and using different communication standards may vary, they generally include the following information: province, city, area where the building is located, name of the building covered by the cell, base station type, main equipment manufacturer, network standard and frequency band, and base station number within the system. For example, a typical indoor cell name might be: "Guangdong Guangzhou Tianhe District CITIC Building Macro Base Station Huawei 5G-2.6GHz 001 Cell", or "Huawei 5G-2.6GHz 001 Cell Guangdong Guangzhou Tianhe District CITIC Building Macro Base Station".

[0029] Based on the above research, it is possible to cluster several indoor cells that converge at the same base station / latitude and longitude and cover the same building, from a reasonable perspective, into the same indoor cell cluster.

[0030] Therefore, this application proposes to obtain the name of the indoor cell so that clustering of the indoor cells can be achieved based on the name of the indoor cell.

[0031] In one optional implementation, taking a network element device of different standards such as 4G or 5G in a mobile communication network as an example, the network element device can obtain the indoor cell name from the network management platform, such as "Beijing Chaoyang Guomao Mall Huawei 5G-2.6GHz056" or "Shanghai Pudong Lujiazui World Financial Center ZTE 4G-LTE023", etc.

[0032] The network management platform mentioned here refers to the core support system in mobile communication networks used for centralized management, configuration, and data synchronization of different network standards such as 4G / 5G. Its full name is usually "Network Management Platform." It is the "data hub and control core" of the mobile communication network, capable of storing basic configuration data for the entire network. For example, it can uniformly collect core information of all network elements (such as 4G eNodeB and 5G gNodeB) and cells (including indoor and outdoor macro cells), including cell name, cell identifier (CI), coverage area, frequency band configuration, and associated network elements.

[0033] In one alternative implementation, for example, one or more of the following keywords can be used in the indoor cell name to query the indoor cell names stored on the network management platform, thereby obtaining indoor cell names containing keywords such as "Beijing Chaoyang Guomao Mall Huawei 5G-2.6GHz056" and "Beijing Chaoyang Guomao Mall Building B Huawei 5G-2.6GHz002".

[0034] Step 2: Based on the names of each indoor cell obtained, determine the names of the buildings that each indoor cell is intended to cover; In one alternative implementation, the obtained name of the indoor cell can be segmented to extract the name of the building covered by the indoor cell.

[0035] In this embodiment of the application, the building name that indicates whether the indoor cell is planned to be covered or actually covered, such as the name of the building extracted from the name of the indoor cell, is referred to as the "name of the building to be covered" of the indoor cell.

[0036] In one optional implementation, the embodiments of this application may employ a building name extraction method that "primarily uses LSTM network extraction and secondarily uses rules." That is, a neural network algorithm is preferentially used to extract building names from the names of indoor communities, and a rule-based method is applied for extraction only if the neural network algorithm fails.

[0037] The two methods will be introduced below.

[0038] (I) Extraction using neural network algorithms In an alternative implementation, a Long Short-Term Memory (LSTM) network can be used, for example, to extract building names.

[0039] The method of extracting building names from cell names using LSTM networks is implemented as follows: Figure 2 As shown. The process mainly includes the following steps, where steps 21 to 23 are for training the LSTM network, which can be done offline, while steps 24 to 25 are for using the trained LSTM network to extract building names from cell names: Step 21: Construct the cell annotation dataset; In the current technology, there is no conventional labeled dataset for community names. Therefore, it is necessary to design a word segmentation and labeling method based on community names to extract the target buildings that are covered, so as to obtain a community labeled dataset as a sample.

[0040] In this embodiment of the application, for the obtained cell name (which can be an indoor cell name), the cell name can first be divided into labeling levels, and then based on the labeling levels, the entities in the cell name can be labeled using the BIOES-based sequence labeling method (referred to as BIOES labeling method).

[0041] BIOES is an abbreviation for five English words: Beginning, Inside, Outside, End, and Single. BIOES is an extension of the traditional BIO annotation method, adding the "End" and "Single" tags to address the issue of ambiguous boundaries for long entities in BIO annotation.

[0042] The core logic of the BIOES annotation method is to use the above five types of tags to annotate each element in the sequence (such as characters and words in text), clarifying whether it belongs to a named entity and its position and role in the entity, and finally reconstructing the entity boundary and type through the tag sequence. The definition and division of these five types of tags are shown in Table 1 below.

[0043] Table 1:

[0044] In this embodiment of the application, it is assumed that the labeling hierarchy of cell names is divided as shown in Table 2 below: Table 2:

[0045] In Table 2, the contents of the "Type" column are all information that may appear in the cell name, such as province, city, district, etc. These types are the possible labeling levels in the cell name, with each type corresponding to one labeling level. The contents of the "Identifier" column represent the symbols that can be labeled at the corresponding labeling level. For example, the symbol (identifier) ​​that can be labeled for "province" is "pro", and the identifier for network standard and frequency band is "O", and so on.

[0046] Based on the above annotation levels and corresponding identifiers, after segmenting the cell names to obtain individual words, these words can be further refined and layered and annotated using the BIOES annotation method shown in Table 1, thereby obtaining the cell annotation dataset.

[0047] For example, a specific example of a labeled cell dataset obtained using this annotation method is as follows: ["Zheng","Zhou","Shi","Jin","Yuan","Da","Xia","-","H","L","H","-","4"],{"tokens":"ner_tags":["B-city","I-city","E-city","B-build","I-build","I-build","E-build", "O", "O", "O", "O", "O", "S"]}. Step 22: Vectorize the labeled cell dataset; Considering that the Chinese characters in the labeled cell dataset cannot be directly understood by the LSTM network, in step 22, each character in the labeled cell dataset is vectorized and converted into a feature vector of fixed dimension d. Here, d is a positive integer, which can be set according to the actual data scale and model performance requirements, preferably d∈[100,512]. By vectorizing the characters, the text data can be transformed into an input format that the LSTM network can accept, while preserving the semantic information and sequence structure features of the text.

[0048] In one alternative implementation, a pre-trained word vector mapping method can be used to vectorize the labeled cell dataset. Specifically: You can choose an open-source Chinese pre-trained word vector model (such as the Chinese pre-trained models of Word2Vec, GloVe, and FastText, or the word embedding layer output of the Chinese BERT model). The model must support vector queries for a single Chinese character, and the dimension of the pre-trained vector must be consistent with the target dimension d (if they are inconsistent, the dimension of the pre-trained vector can be adjusted to d through linear transformation). For each character in the character-index mapping dictionary D, query the pre-trained word vector model to obtain the pre-trained vector corresponding to the character, which is used as the initial feature vector of the character; For out-of-vocabulary (OOV) characters, a vector of dimension d is generated by random initialization (the random values ​​follow a uniform distribution in the interval [-0.01, 0.01], or a normal distribution with a mean of 0 and a variance of 0.01), and this vector is used as the feature vector of the OOV character. Establish a mapping table of "character index - feature vector", traverse each text sequence in the labeled cell dataset, and retrieve the corresponding feature vector from the mapping table according to the index of the character to form a sequence feature matrix of dimension [L,d] (L is the set uniform sequence length) to complete the vectorization of a single text.

[0049] Step 23: Input the vectorized labeled cell dataset into the LSTM network to be trained, and train the LSTM network to obtain the trained LSTM network. In an optional implementation, to ensure the LSTM network obtains an accurate inference result, when performing inference based on character vectors, it is advisable to consider combining the contextual information of a single character within the cell name. Therefore, in this embodiment, a sliding window is specifically set up to simultaneously input the character vectors within the window into the LSTM network to be trained. For example, the size of the sliding window can be set to 5. Then, according to this sliding window, the first two and last two character vectors of a given character vector can be concatenated and used as a single input into the LSTM network to be trained.

[0050] The following explains the specific training process of the LSTM network: The core of the LSTM network lies in the transmission of cell states. The gate structure determines whether information is added or deleted from the cell state. The gate structure design is implemented using a sigmoid neural layer and pointwise multiplication operations. The LSTM network uses three types of gate structures—input gate, forget gate, and output gate—to protect and control information. The relevant descriptions of these three gates are as follows: 1) Forget Gate: Determines how much information passed from the previous character is discarded from the cell state; in other words, it's responsible for deleting useless old information from the cell state. The forget gate uses the Sigmoid function to output a value between 0 and 1, where 1 represents "completely retain" and 0 represents "completely discard." The calculation method is as follows:

[0051] in, It is the output of the forget gate. It is the Sigmoid activation function. and These are the weights and biases of the forget gate, respectively. This indicates the output of the previous character's hidden layer. It is the current input to the cell.

[0052] 2) Input Gate: Determines how much information is incorporated into the current cell state, including filtering and adding new useful information. The input gate consists of two parts: a sigmoid layer that determines which information needs to be updated, and a tanh layer that generates a candidate vector to be added to the current cell state. The calculation method is as follows:

[0053]

[0054] in, It is the output of the input gate. and yes Door bias term, It is the state of the candidate memory unit.

[0055] The forget gate and the input gate work together to achieve the following: t Unit state update for each character:

[0056] in, for No. t The updated cell state of the nth character, i.e., the nth... t Time (processing the first) t Cell state (when there are 1 character).

[0057] 3) Output gate: Determines which information to output, that is, from the updated cell state (unit state), it filters out information useful for the current task and generates the hidden state at the current moment. The output gate does not change the cell state itself; it only determines "which information in the cell state should be output or transmitted".

[0058] The output value is also scaled using tanh, calculated as follows:

[0059]

[0060] in, This represents the dot product operation.

[0061] After calculating the hidden state Afterwards, based on Predict the task outcome (e.g., what the next character will be) and calculate the error between the predicted and actual values.

[0062] Then, the model will backpropagate the forward propagation error (such as the difference between the predicted character and the actual character) and calculate each parameter (the weights of the three gates). and bias terms The "contribution" (i.e., gradient) to the error.

[0063] For example, if the forget gate fails to remove useless old information, causing an error in predicting the next character, backpropagation will calculate that "a certain weight of the forget gate should be adjusted so that its forgetting weight for this type of old information is closer to 1 (i.e., more deletion)"; another example is that if the input gate stores irrelevant character information into the cell state, causing the error to increase, backpropagation will adjust the parameters of the input gate so that its filtering weight for this type of information is closer to 0 (i.e., filtering).

[0064] Finally, by using optimization algorithms (such as SGD or Adam), and based on the gradients calculated through backpropagation, all parameters of the three gates are adjusted. / / ), / / (etc.) — For example, increase the weight of the forget gate that "leads to insufficient forgetting" and decrease the bias term of the input gate that "leads to redundant input information".

[0065] Repeat the above process of "forward calculation of output → backward calculation of error → optimization of gate parameters" until the parameters of the three gates are stable and the gate decisions can minimize the error of the LSTM network (e.g., the accuracy of character prediction reaches the target). The training is then complete, and the trained LSTM network is obtained.

[0066] In this embodiment of the application, the essence of training the LSTM network is to optimize the parameters of the three gates through backpropagation, so that they learn "when to forget, when to store, and when to output", and finally enable the LSTM network to accurately capture sequence dependencies (such as contextual associations in text such as "cell name").

[0067] Step 24: Perform label inference based on the trained LSTM network; First, for a community name, after word segmentation and vectorization, it is input into a trained LSTM network to obtain the output. To form a total score matrix .in: n The total number of characters contained in the community name; k The total number of tag categories in the preset tag set; matrix elements. This indicates that the trained LSTM network predicts the first... i characters Corresponding to the j The probability score of each label ( , ), and satisfy for any i , (That is, the sum of the probability scores of a single character across all tags is 1). Based on a matrix. P Preliminary results can be obtained n Each character's label predicts the candidate sequence. ,in For the first i A set of candidate labels of 1 character.

[0068] Then, a label transition score matrix is ​​introduced to optimize the inference results: in order to capture the contextual dependencies between labels (avoiding logical contradictions in the label sequence caused by isolated predictions), a pre-defined label transition score matrix is ​​introduced. Matrix elements Indicates the number of tags in the label sequence. i Class tags are directly transferred to the first j probability score of class label ( , The parameters of this matrix are determined by the statistical regularity of the label sequences in the training data or by an additional sequence labeling training process.

[0069] Then, combine the label probability score matrix P With label transition score matrix A The final label score for each character is calculated using a sequence labeling inference algorithm (such as the Viterbi algorithm), as shown in the following formula:

[0070] in, This represents the final overall score when the i-th character is labeled as the j-th category. For example, That is, the initial score of the first character is equal to its label probability score. m Indicates the first A tag category of 1 character.

[0071] Based on the final composite score The optimal label for each character is selected to obtain the final label prediction sequence. ,in For the first i The final label of 1 character.

[0072] Step 25: Determine the building name based on the obtained final label prediction sequence.

[0073] For example, a predefined target label set (T = {"B-build", "I-build", "E-build") is defined, where "B-build" represents the starting character label of the building name, "I-build" represents the middle character label of the building name, and "E-build" represents the ending character label of the building name. The final label prediction sequence is then iterated through. Y Extract a continuous subsequence of characters that meets the following conditions: it starts with the label "B-build", contains only the label "I-build" (optional) in the middle, and ends with the label "E-build".

[0074] The extracted character subsequence is the building name extracted from the community name.

[0075] (II) Extraction based on rules The method for extracting building names from community names based on rules, and its specific implementation process is as follows: Figure 3 As shown. The process mainly includes the following steps: Step 31: First, segment the cell name according to the basic identifier to obtain the segmentation fields contained in the cell name; For example, the cell name can be split into strings based on basic identifiers (including but not limited to "-", "_", etc.) to obtain each first split field.

[0076] Here, the segmentation field refers to the independent semantic unit formed after the string is segmented, and each unit carries specific information (such as building name, frequency band identifier, serial number, etc.). For ease of distinction and description, the segmentation field obtained by performing step 31 is called the first segmentation field.

[0077] For example, in the indoor community name "Guangdong Shenzhen-Futian District-Xinghe Building-Huawei-5G-n78-001", if "-" is used as the delimiter, the resulting "Guangdong Shenzhen", "Futian District", "Xinghe Building", "Huawei", "5G", and "n78-001" are all the first delimiter fields.

[0078] Step 32: Determine whether each first segmentation field contains Chinese characters; for the first segmentation field determined to contain Chinese characters, execute step 32-1 "Keep the first segmentation field", and then execute step 33; for the first segmentation field determined not to contain Chinese characters, execute step 32-2 "Remove the first segmentation field", and then execute step 33. By executing steps 32, 32-1, and 32-2, the first segmentation field containing Chinese characters (e.g., the name of the building intended to be covered) can be retained. The first segmentation field containing only the station type, network standard and frequency band, and base station number within the system can be removed.

[0079] Using the previous example, after removing the first segmentation field that does not contain Chinese characters from "Guangdong Shenzhen", "Futian District", "Xinghe Building", "Huawei", "5G", and "n78-001", the remaining fields are "Guangdong Shenzhen", "Futian District", "Xinghe Building", and "Huawei".

[0080] It should be noted that, considering that the manufacturers of indoor stations (such as Huawei) are generally relatively fixed and limited in number, in one implementation, a database can be pre-established to store the "names of indoor station manufacturers (such as Huawei)". Therefore, by comparing the remaining first segmentation field with the name of this database, it can be determined whether the remaining first field contains the "name of indoor station manufacturer (such as Huawei)". If it does, such a first field can be further removed.

[0081] Step 33: For the first segmentation field retained after performing the above steps, arrange it sequentially according to the original arrangement position of the first segmentation field in the cell name to obtain a new string; identify the geographic information identifier contained in the new string, and segment the new string according to the geographic information identifier to obtain the second segmentation field; Geographic information identifiers are standardized identifiers or strings used to uniquely identify and locate geographic spatial entities (such as buildings, communities, and regions) and to carry geographic-related attribute information.

[0082] Following the previous example, by segmenting the new string "Guangdong Shenzhen Futian District Xinghe Building (with "name of indoor station manufacturer (such as Huawei)" removed) according to the geographic information identifier, we can obtain the second segmentation field "Guangdong", "Shenzhen", "Futian District", and "Xinghe Building".

[0083] Step 34: Determine whether each second segmentation field contains a geographic information identifier of a specified type; for second segmentation fields that are determined not to contain geographic information identifiers of a specified type, execute step 34-1 "retain the second segmentation field", and then execute step 35; for second segmentation fields that are determined to contain geographic information identifiers of a specified type, execute step 34-2 "remove the second segmentation field", and then execute step 35. Taking geographic information identifiers applicable to China's administrative divisions as an example, the specified types of geographic information identifiers mentioned here include those representing provinces, prefectures, cities, districts, and counties. These provinces, prefectures, cities, districts, and counties all refer to areas significantly larger than the buildings intended to be covered by the indoor cell.

[0084] Using the previous example, for the second dividing field "Guangdong", "Shenzhen", "Futian District" and "Xinghe Building", by executing steps 34, 24-1 and 24-2, we can finally get "Futian District" and "Xinghe Building" remaining.

[0085] Step 35: For the second segmentation field retained after performing the above steps, arrange it sequentially according to the original arrangement position of the second segmentation field in the community name to obtain the final string, which is used as the building name extracted from the community name.

[0086] Using the previous example, by arranging the remaining "Futian District" and "Xinghe Building" in sequence, we can obtain "Futian District Xinghe Building", which can be used as the building name extracted from the community name, that is, the building name that the indoor community intends to cover.

[0087] Step 3: Based on the names of the buildings that each indoor cell intends to cover, cluster the indoor cells that intend to cover the same buildings into the same indoor cell cluster to obtain at least one indoor cell cluster.

[0088] After obtaining at least one indoor cell cluster, at least one cell cluster can be selected from it as the target indoor cell cluster.

[0089] In step ③, indoor cell clustering can be achieved using the following method: First, the names of the buildings that each indoor cell intends to cover are converted into feature vectors based on TF-IDF weights. Specifically, a corpus containing the names of all buildings that the indoor cells intend to cover can be constructed, and each name can be segmented. Second, the weight of each keyword in the corpus is calculated based on the TF-IDF formula to generate the TF-IDF feature vector of each name. Then, calculate the pairwise similarity between feature vectors (such as cosine similarity). Finally, by merging indoor cells with a similarity greater than a preset threshold into a cluster, at least one indoor cell cluster is obtained. For each indoor cell cluster, the name of the building extracted from the name of any indoor cell in that indoor cell cluster is used as the name of that indoor cell cluster.

[0090] In this embodiment of the application, whenever an indoor cell is clustered into an indoor cell cluster, and the indoor cell cluster is named in the manner described above, the name of the building that all indoor cells in the cluster intend to cover is the name of the same building, which is also the name of the indoor cell cluster.

[0091] In this embodiment of the application, after obtaining and naming each indoor cell cluster, in an optional implementation, the buildings surrounding the geographical location of the indoor cell cluster can be determined as target buildings.

[0092] Specifically, in one optional implementation, assuming any indoor cell cluster is the target indoor cell cluster, at least one building whose distance from the first geographical location is within a preset distance range can be selected as the target building based on the geographical location of the target indoor cell cluster (hereinafter referred to as the first geographical location).

[0093] In a specific example, consider the Financial Center Block A residential cluster, whose primary geographical location has latitude and longitude coordinates of (116.403874°, 39.914885°). Assuming a preset distance range of 50 meters, the location data of buildings within a 50-meter radius centered on this primary geographical location can be retrieved using a Geographic Information System (GIS). Buildings within this area can then be identified as target buildings corresponding to the target indoor residential cluster. For instance, after filtering, only the geographical location of "Financial Center Block A" (116.403921°, 39.914912°) meets the criteria, and it is thus identified as the target building corresponding to the target indoor residential cluster.

[0094] Once the target building is identified, the operation of "determining the similarity between the name of the target building and the name of the building that the indoor cells in the target indoor cell cluster intend to cover" can be performed.

[0095] For example, similar to the description above, the name of the target building and the name of the building that the indoor cells in the target indoor cell cluster intend to cover can both be converted into feature vectors based on TF-IDF weights; then, the similarity between each pair of feature vectors (such as cosine similarity) can be calculated.

[0096] Step 12: Determine the relative positions of the target building and the target indoor cell cluster; In this embodiment of the application, the relative positional relationship includes: a relative positional relationship that characterizes whether indoor cells in the target indoor cell cluster cover the target building. For example, the relative positional relationship may include: the target indoor cell cluster is located inside (or outside) the target building.

[0097] In one alternative implementation, a ray discrimination method can be used to determine whether the target indoor cell cluster is located inside (or outside) the target building, as follows: First, take the location of the target indoor cell cluster as the measurement point P, and draw rays in any direction from P as the endpoint—see the instruction manual for details. Figure 4 ; Then, count the number of valid intersections between the ray and the outline of the target building; Finally, the number of valid intersection points determines whether the target indoor cell cluster is located inside the target building. If the number of valid intersection points is odd, the target indoor cell cluster is located inside the target building; if the number of valid intersection points is even, the target indoor cell cluster is located outside the target building.

[0098] A specific example is as follows: Step 1: Obtain basic geographic data First, retrieve the first geographic location data of the target indoor cell cluster through GIS. Use the average latitude and longitude coordinates of all indoor cells in the cluster as the reference point P, denoted as (lon1, lat1). Example: P (116.403874°, 39.914885°). Simultaneously, the geographic boundary data of the target building is acquired. This data is a set of polygonal outline vertices of the target building in GIS, with each vertex represented by latitude and longitude coordinates (London, Latitude, and Longitude). i lat i (i=1,2,...,n, where n is the total number of vertices). Example: The boundary vertices of the target building are A (116.403921°, 39.914912°), B (116.404056°, 39.914908°), C (116.404052°, 39.914789°), and D (116.403917°, 39.914793°), forming a quadrilateral boundary.

[0099] Step 2: Constructing Ray and Coordinate Standardization Starting from the reference point P, a ray L with a fixed direction is constructed. In this embodiment, the direction of ray L is set to due north (i.e., parallel to the Earth's meridian, azimuth angle 0°). It can also be preset to due east (azimuth angle 90°) or other fixed directions according to the actual application scenario to ensure the consistency of the judgment standard. To simplify calculations, the latitude and longitude coordinates of the reference point P and all vertices of the target building boundary are converted to coordinates (x, y) in a Cartesian coordinate system. The conversion uses the Gauss-Kruger projection algorithm, and the projection zone is selected as the 3° zone corresponding to the target area (e.g., 116°E corresponds to zone 39). Examples of the converted values: P (5231892.15m, 3856721.32m), A (5231905.48m, 3856732.16m), B (5231928.73m, 3856731.24m), C (5231927.96m, 3856701.58m), D (5231894.71m, 3856702.45m).

[0100] Step 3: Determining the intersection point of the ray and the building boundary Traverse each edge of the target building boundary (consisting of two adjacent vertices, such as edge AB, edge BC, edge CD, and edge DA), and determine whether ray L intersects with that edge. The rules for determining the intersection point are as follows: Let the two endpoints of the boundary edge be Q1(x1, y1) and Q2(x2, y2), and the equation of ray L be y = y_P (since the ray points due north, x can increase infinitely, and y is fixed as the y coordinate of the reference point P). If the y - coordinates of both endpoints of the edge Q1Q2 are greater than y_P or both are less than y_P, then the ray L has no intersection with this edge; If the y - coordinates of the two endpoints of the edge Q1Q2 are on both sides of y_P (including the case where the y - coordinate of one of the endpoints is equal to y_P), then calculate the intersection coordinates (x_inter, y_P) of the ray L and the edge Q1Q2. The intersection x_inter is calculated by the linear interpolation formula: x_inter = x1+(y_P - y1)×(x2 - x1) / (y2 - y1); Only retain the intersections where x_inter>x_P (i.e., the intersections in the due - north extension direction of the ray L), and exclude the invalid intersections where x_inter≤x_P.

[0101] In the example, the y - coordinate of the reference point P is 3856721.32m. The endpoints A (y = 3856732.16m) and B (y = 3856731.24m) of the edge AB are both greater than y_P, so there is no intersection; for the edge BC, the endpoint B (y = 3856731.24m)>y_P and C (y = 3856701.58m)<y_P. The calculated intersection x_inter = 5231918.65m>x_P (5231892.15m), which is a valid intersection; for the edges CD and DA, the y - coordinates of their endpoints are both less than y_P, so there is no intersection. Finally, the number of valid intersections of the ray L and the boundary of the target building is 1.

[0102] Step 4: Determination of relative position relationship According to the number of valid intersections and the position relationship between the boundary of the target building and the reference point, determine the relative position relationship between the target building and the target indoor cell cluster: If the number of valid intersections is odd, it is determined that the target building is within the coverage range of the ray L, that is, the relative position between the target building and the target indoor cell cluster is "overlapping association", indicating that the spatial distance between the two is relatively close; if the number of valid intersections is even (including 0), it is determined that the target building is outside the coverage range of the ray L, that is, the relative position is "non - overlapping association", indicating that the spatial distance between the two is relatively far.

[0103] Through the ray discrimination method combined with coordinate projection and geometric calculation in the embodiments of this application, the spatial association relationship between the target building and the target indoor cell cluster can be accurately determined, avoiding the misjudgment problem of adjacent buildings caused by only judging through the straight - line distance, and providing a reliable position basis for the subsequent establishment of the first mapping relationship.

[0104] Step 13: When the similarity between the name of the target building and the name of the building that the indoor cell in the target indoor cell cluster intends to cover, and the relative positional relationship between the target building and the target indoor cell cluster, satisfy the preset mapping relationship establishment conditions, establish the first mapping relationship between the target building and the indoor cell in the target indoor cell cluster.

[0105] In one optional implementation, the preset mapping relationship establishment conditions may include: 1. The similarity between the name of the target building and the name of the building that the indoor cells in the target indoor cell cluster intend to cover is greater than the preset similarity threshold; 2. The "relative positional relationship between the target building and the target indoor cell cluster" is "the target indoor cell cluster is located inside the target building".

[0106] If both 1 and 2 above are true, then a first mapping relationship can be established between the target building and the indoor cells in the target indoor cell cluster.

[0107] In this embodiment of the application, establishing a first mapping relationship may include, but is not limited to, the following operations: Store the "unique identifier and / or name of the target building" and the "unique identifier and / or name of the target indoor cell cluster" in a corresponding manner or associate them in other ways.

[0108] The method provided in this application combines the dual dimensions of "name" and "location" to achieve accurate and automated establishment of the mapping relationship between buildings and communities, solving the problems of low efficiency and large error in traditional methods.

[0109] As mentioned above, for the same building, there may be both outdoor and indoor stations covering the building. Based on the preceding description, a first mapping relationship can be established between the building and the indoor station's cells. Therefore, for the case where the building is also covered by an outdoor station, this application further provides the following scheme to establish a second mapping relationship between the building and the outdoor station's cells.

[0110] Specifically, in an optional implementation, the method provided in this application embodiment may further include: establishing a second mapping relationship between the target building and the outdoor cell accessed by the terminal device at the second geographical location based on the changes in the communication signal received by the terminal device at the second geographical location of the target building.

[0111] Among these, "terminal equipment" generally refers to user equipment, or UE, such as a user's mobile phone, wearable device, or smart device; "secondary geographical location" refers to the geographical location of the target building. "Communication signals" include, but are not limited to, one or more of the following: signals transmitted by indoor stations, signals transmitted by outdoor stations, and signals transmitted by Wi-Fi hotspots.

[0112] In one optional implementation, a second mapping relationship is established based on changes in the received communication signals. Specifically, this may include: if, based on changes in the received communication signals, it is determined that an event representing a sudden change in the outdoor cell signal reception strength has occurred at a second geographical location, then a second mapping relationship is established.

[0113] In one optional implementation, the event characterizing a sudden change in outdoor cell signal reception strength may include, but is not limited to, one or more of the following events: Within the first specified time period, the terminal device switched between outdoor and indoor cells; Within a second specified time period, the terminal device switched between a mobile communication network and a wireless local area network (e.g., a Wi-Fi network); Within a third specified time period, the difference between the outdoor cell reference signal received power (RSRP) and the indoor cell RSRP of the terminal device is greater than a preset difference threshold.

[0114] The first specified time length, the second specified time length, and the third specified time length can be the same or different.

[0115] It should be noted that the above events are all directly related to changes in the communication network characteristics of the environment in which the terminal device is located: the switching between outdoor and indoor cells, and the switching between mobile communication networks and WIFI networks, are essentially the terminal moving from an environment with no indoor network coverage (or weak indoor network signal) to a scenario where the indoor network signal can be received stably. This scenario usually corresponds to the target building area where both outdoor cell signals and indoor cell (or WIFI) signals are covered. When the difference between the terminal's outdoor cell RSRP and indoor cell RSRP is greater than a preset threshold, it often means that the terminal has approached or entered an area with a strong indoor network signal, and this area usually also has outdoor cell signal coverage, which is consistent with the network coverage characteristics of the target building.

[0116] Considering the aforementioned characteristics of events that characterize sudden changes in outdoor cell signal reception strength, embodiments of this application propose using whether such an event has occurred at a second geographical location of the terminal device as the basis for establishing a second mapping relationship. This ensures that the established second mapping relationship is accurate and reasonable.

[0117] In a specific example, assuming the data is sorted from earliest to latest according to timestamp, and the relevant data of the cell accessed by the UE and the changes in the network source are captured, the data is shown in Table 3: Table 3:

[0118] Taking the first row of data as an example, R1,1,t0 and S1,1,t0 represent the RSRP and SINR values ​​collected by the user terminal with ID "User_Encrypted_ID1" when it accesses cell Cell_ID1 at time t0, respectively.

[0119] Set the threshold for the cell handover interval for user terminals. The following conditions can be used to determine whether a user terminal has experienced an event that indicates a sudden change in the outdoor cell signal reception strength: 1) The user terminal's access cell switches from an outdoor cell cluster to an indoor cell cluster (or from an indoor cell cluster to an outdoor cell cluster), and the adjacent time does not exceed [a certain value]. The adjacent time mentioned here refers to the time difference between the cell handover trigger time and the cell handover completion time.

[0120] 2) The network source of the user terminal connection is switched from mobile communication network to WIFI (or from WIFI to mobile communication network), and the adjacent time does not exceed [the specified time]. The meaning of adjacent times here is the same as above.

[0121] 3) When the RSRP value received by the user terminal from the currently accessing cell differs from the RSRP value of the outdoor cell by more than 15, and the adjacent time intervals do not exceed [a certain value], the user terminal is considered to be in a state of emergency. The meaning of adjacent times here is the same as above.

[0122] If any of the above conditions are met, it can be determined that the aforementioned event has occurred on the user terminal.

[0123] For the data in other rows, their meaning and how to determine whether the user terminal has experienced the aforementioned event based on the data can be found in the above text, and will not be repeated here.

[0124] The following is a specific embodiment, which is used to illustrate the implementation process of establishing the first and second mapping relationships in a real-world scenario using the method provided in this application.

[0125] This embodiment uses a mobile communication network in a core urban business district as the application scenario. This area contains 10 high-rise buildings, corresponding to 25 indoor cells. The indoor cells achieve signal coverage inside the buildings through a distributed antenna system. The name of each indoor cell includes the name of the building it intends to cover, as shown in Table 4 below. Table 4:

[0126] For the above application scenarios, the first mapping relationship can be established using the following process: (I) Clustering process of indoor cell clusters 1) First, the name information of the above 25 indoor communities is obtained through the network management system, and the core fields representing the buildings in the name of each indoor community are extracted using a string parsing algorithm.

[0127] For example, after analyzing the indoor community name "CBD-Financial Center A-1F" and removing the area identifier "CBD" and the floor identifier "1F", it is determined that the building name it intends to cover is "Financial Center Tower A".

[0128] 2) Subsequently, clustering is performed based on the extracted building names to group indoor cells that intend to cover the same building into the same indoor cell cluster; the name of the building that any indoor cell in the same indoor cell cluster intends to cover is named the name of the indoor cell cluster.

[0129] Specifically, in this embodiment, IC1, IC2, and IC3 are clustered into the same indoor cell cluster and named "Financial Center Block A Cell Cluster"; IC4 and IC5 are clustered into the same indoor cell cluster and named "Financial Center Block B Cell Cluster"; IC6 and IC7 are clustered into the same indoor cell cluster and named "Science and Technology Building Cell Cluster"...

[0130] In this way, 10 indoor cell clusters can be obtained, each cluster corresponding to a building, and each indoor cell cluster obtained by the above clustering can be selected as the target indoor cell cluster.

[0131] (II) The process of selecting the target building 1) First, obtain the first geographical location of each target indoor cell cluster.

[0132] For example, taking the Financial Center Block A residential cluster as an example, its primary geographical location has latitude and longitude coordinates of (116.403874°, 39.914885°).

[0133] 2) Then, assuming the preset distance range is 50 meters, the location data of buildings in the area with the first geographical location as the center and a radius of 50 meters are retrieved through GIS, and the buildings located in this area are taken as the target buildings corresponding to the target indoor community cluster.

[0134] Using the previous example, for instance, after screening, only the geographical location of "Financial Center Block A" (116.403921°, 39.914912°) meets the conditions, and it is identified as the target building corresponding to the target indoor community cluster.

[0135] (III) Calculation process of name similarity The cosine similarity algorithm is used to calculate the similarity between the target building name and the building names that the target indoor cell cluster intends to cover.

[0136] The formula for calculating cosine similarity is: Here, A and B are the word vectors corresponding to the two names respectively. These two names are: the name of the target building and the name of the building that the target indoor cell cluster intends to cover, where the latter is also the name of the target indoor cell cluster.

[0137] Based on the cosine similarity calculation formula, the similarity between "Financial Center A" (the target building name) and the building name "Financial Center A" intended to be covered by the Financial Center A cluster is calculated. After converting the two names into word vectors, the similarity is calculated to be 1.0. If, due to data entry error, the building name intended to be covered by an indoor cluster is mistakenly written as "Financial Center A", the calculated similarity is 0.92.

[0138] In this embodiment, a preset similarity threshold of 0.85 can be used. It can be seen that the similarities calculated above are all greater than this preset similarity threshold.

[0139] (iv) Determination of relative positional relationships and establishment of the first mapping relationship First, the ray discrimination method is used to determine whether the target indoor cell cluster is located inside (or outside) the target building.

[0140] For example, the ray discriminant method can be used to determine whether the cluster of buildings in Financial Center A is located within the target building "Financial Center A". The specific determination process can be found in the previous description and will not be repeated here.

[0141] Assuming that the cluster of residential units in Financial Center A is located in the target building "Financial Center A", the names of each indoor unit in Financial Center A cluster and the name of the target building "Financial Center A" can be stored accordingly, thereby establishing the first mapping relationship.

[0142] Similarly, the first mapping relationship between the remaining 9 indoor cell clusters and their corresponding target buildings can be established.

[0143] (v) Acquisition of communication signal data One hundred terminal devices in normal communication status were selected. The terminal devices support fourth-generation mobile communication technology (4G) and wireless local area network (WLAN). Real-time communication signal data of the terminal devices in the second geographical location of Financial Center A (i.e., the entrance and exit of Financial Center A and the surrounding area, with latitude and longitude range of 116.4037°-116.4041°, 39.9147°-39.960°) were collected, including cell access status, reference signal received power (RSRP), network type, etc.

[0144] (vi) Judgment of signal mutation events and establishment of the second mapping relationship The preset first specified time length is 5 seconds, the second specified time length is 3 seconds, and the third specified time length is 2 seconds, with a preset difference threshold of 20dBm.

[0145] Event 1 (Indoor / Outdoor Cell Handover): When a terminal device moves from the outdoor area of ​​Building A of the Financial Center to the indoor area, it completes the handover from the outdoor cell OC1 (frequency 1800MHz, cell identifier 12345) to the indoor cell IC1 within 4 seconds. This handover event meets the first specified time length condition, indicating that the signal reception strength of the outdoor cell OC1 changes abruptly, and it is determined that OC1 is associated with Building A of the Financial Center.

[0146] Event 2 (Mobile Network to WLAN handover): When another terminal device stopped at the entrance of Building A of the Financial Center, it completed the handover from the mobile communication network (accessing outdoor cell OC2) to WLAN (SSID: Financial Center A-WLAN) within 2 seconds, which met the second specified time length condition, indicating a sudden change in the signal strength of OC2, and confirming that OC2 and Building A of the Financial Center are associated.

[0147] Event 3 (RSRP Difference Exceeds Limit): When another terminal device is near a window in Building A of the Financial Center, the RSRP of outdoor cell OC3 is -105dBm within 1.5 seconds, while the RSRP of indoor cell IC2 is -82dBm. The difference between the two is 23dBm, which is greater than the preset difference threshold of 20dBm. This meets the third specified time length condition, indicating a sudden change in the signal strength of OC3, and confirming a correlation between OC3 and Building A of the Financial Center. Based on the above three events, a second mapping relationship is established between Building A of the Financial Center and outdoor cells OC1, OC2, and OC3, respectively.

[0148] The application of the method provided in this application in a real-world scenario demonstrates that the method can achieve the following: by clustering indoor cells, the amount of matching computation is reduced; by combining name similarity and relative positional relationship to determine the first mapping relationship, the matching accuracy is high; at the same time, by determining the second mapping relationship based on the terminal communication signal mutation event, the accurate association between outdoor cells and buildings is achieved.

[0149] This method is fully automated and requires no manual intervention. It significantly shortens the time required to establish the mapping relationship between cells and buildings from several weeks, greatly reducing network operation and maintenance costs and providing reliable data support for the efficient optimization of mobile communication networks.

[0150] To address the problem of how to efficiently and accurately determine the mapping relationship between communication cells and buildings in existing technologies, and based on the same inventive concept as the above embodiments of this application, this application also provides a device for determining the mapping relationship between cells and buildings.

[0151] A schematic diagram of the specific structure of the device is shown below. Figure 6 As shown, it includes the following functional modules: The similarity determination module 61 is used to determine the similarity between the name of the target building and the name of the building that the indoor cells in the target indoor cell cluster intend to cover; The positional relationship determination module 62 is used to determine the relative positional relationship between the target building and the target indoor cell cluster; The first mapping establishment module 63 is used to establish a first mapping relationship between the target building and the indoor cells in the target indoor cell cluster when the similarity and the relative position relationship meet the preset mapping relationship establishment conditions.

[0152] In an optional implementation, the apparatus provided in this application embodiment may further include: The building coverage determination module is used to determine the names of the buildings that each indoor cell intends to cover, based on the names of each indoor cell obtained, before the similarity determination module determines the similarity. The cell clustering module is used to cluster indoor cells that intend to cover the same building into the same indoor cell cluster based on the names of the buildings that each indoor cell intends to cover, thereby obtaining at least one indoor cell cluster.

[0153] The target indoor cell cluster is one or more indoor cell clusters among the at least one indoor cell clusters obtained by clustering.

[0154] In an optional implementation, the apparatus provided in this application embodiment may further include: The target building selection module is used to select at least one building whose distance from the first geographical location is within a preset distance range, based on the first geographical location of the target indoor cell cluster, before the similarity determination module determines the similarity, as the target building.

[0155] In an optional implementation, the apparatus provided in this application embodiment may further include: The second mapping establishment module is used to establish a second mapping relationship between the target building and the outdoor cell accessed by the terminal device at the second geographical location based on the changes in the communication signal received by the terminal device at the second geographical location of the target building.

[0156] In one optional implementation, the second mapping establishment module is specifically used to: if, based on the changes in the received communication signals, it is determined that the terminal device has experienced an event representing a sudden change in the outdoor cell signal reception strength at the second geographical location, then the second mapping relationship is established.

[0157] In one alternative implementation, the event includes at least one of the following: Within a first specified time period, the terminal device switched between outdoor and indoor cells; Within a second specified time period, the terminal device switched between a mobile communication network and a wireless local area network; Within a third specified time period, the difference between the outdoor cell reference signal received power and the indoor cell reference signal received power of the terminal device is greater than a preset difference threshold.

[0158] The device described in this application combines the dual dimensions of "name" and "location" to achieve accurate and automated establishment of the mapping relationship between buildings and communities, thus solving the problems of low efficiency and large error in traditional methods.

[0159] Based on the same inventive concept as the foregoing embodiments of this application, this application provides a computing device to solve the problem of how to efficiently and accurately determine the mapping relationship between communication cells and buildings in the prior art.

[0160] like Figure 7As shown, the computing device includes a memory 71 and a processor 72. The memory 71 can be configured to store various other data to support operation on the electronic device. Examples of this data include instructions for any application or method used to operate on the electronic device. The memory 71 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0161] The processor 72, coupled to the memory 71, is used to execute the computer program stored in the memory 71 to perform a method for determining the mapping relationship between a cell and a building as described in Embodiment 1 of this application.

[0162] When the processor 72 executes the computer program in the memory 71, in addition to the functions described above, it can also perform other functions, as detailed in the descriptions of the preceding embodiments.

[0163] Furthermore, such as Figure 7 As shown, the computing device also includes other components such as a display 74, a communication component 73, a power supply component 75, and an audio component 76. Figure 7 The diagram only shows some components and does not mean that the computing device includes only these components. Figure 7 The components shown.

[0164] Accordingly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a computer, can implement the methods provided in the above embodiments.

[0165] Accordingly, this application also provides a computer program product, which stores instructions that, when executed by a computer, cause the computer to implement the methods provided in the above embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0166] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0167] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for determining the mapping relationship between residential areas and buildings, characterized in that, include: Determine the similarity between the name of the target building and the name of the building that the indoor cells in the target indoor cell cluster intend to cover; Determine the relative positional relationship between the target building and the target indoor cell cluster; When the similarity and the relative positional relationship satisfy the preset mapping relationship establishment conditions, a first mapping relationship is established between the target building and the indoor cells in the target indoor cell cluster.

2. The method as described in claim 1, characterized in that, Before determining the similarity, the method further includes: Based on the names of each indoor cell obtained, determine the names of the buildings that each indoor cell is intended to cover; Based on the names of the buildings that each indoor cell intends to cover, the indoor cells that intend to cover the same building are clustered into the same indoor cell cluster to obtain at least one indoor cell cluster. The target indoor cell cluster is one or more indoor cell clusters among the at least one indoor cell cluster obtained.

3. The method as described in claim 1 or 2, characterized in that, Before determining the similarity, the method further includes: Based on the first geographical location of the target indoor cluster, at least one building whose distance from the first geographical location is within a preset distance range is selected as the target building.

4. The method as described in claim 1, characterized in that, The method further includes: Based on the changes in the communication signals received by the terminal device at the second geographical location of the target building, a second mapping relationship is established between the target building and the outdoor cell accessed by the terminal device at the second geographical location.

5. The method as described in claim 4, characterized in that, Based on the changes in the received communication signals, the second mapping relationship is established, including: If, based on the changes in the received communication signals, it is determined that an event characterized by a sudden change in the outdoor cell signal reception strength has occurred at the second geographical location of the terminal device, then the second mapping relationship is established.

6. The method as described in claim 5, characterized in that, The event includes at least one of the following: Within a first specified time period, the terminal device switched between outdoor and indoor cells; Within a second specified time period, the terminal device switched between a mobile communication network and a wireless local area network; Within a third specified time period, the difference between the outdoor cell reference signal received power and the indoor cell reference signal received power of the terminal device is greater than a preset difference threshold.

7. A device for determining the mapping relationship between residential areas and buildings, characterized in that, include: The similarity determination module is used to determine the similarity between the name of the target building and the name of the building that the indoor cells in the target indoor cell cluster intend to cover; A location relationship determination module is used to determine the relative location relationship between the target building and the target indoor cell cluster; The first mapping establishment module is used to establish a first mapping relationship between the target building and the indoor cells in the target indoor cell cluster when the similarity and the relative position relationship meet the preset mapping relationship establishment conditions.

8. A computing device, characterized in that, include: Memory and processor, among which, The memory is used to store computer programs; The processor, coupled to the memory, is configured to execute the computer program stored in the memory for performing the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, which, when executed by a computer, enables the implementation of the method described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product stores instructions that, when executed by a computer, cause the computer to perform the method described in any one of claims 1 to 6.

Images