Method and device for identifying base station co-coverage relationship, storage medium and electronic device

Abstract

The present disclosure relates to the technical field of communication, and provides a base station co-coverage relationship identification method, a base station co-coverage relationship identification device, a computer storage medium and an electronic device, wherein the base station co-coverage relationship identification method comprises: obtaining index measurement data of at least two base stations and corresponding basic configuration parameters of each base station; the basic configuration parameters contain name information of the base station, the name information contains position description information of the base station, and the position description information contains the height of the base station from a preset reference surface; performing information mining on the name information of each base station to obtain deep hidden information; inputting the index measurement data, the basic configuration parameters and the deep hidden information into a trained co-coverage relationship identification model, and determining whether the at least two base stations have a co-coverage relationship according to the output of the co-coverage relationship identification model. The present disclosure can identify whether indoor base stations have a co-coverage relationship, and has a wide range of applications.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of communication technology, and in particular, to a method for identifying a base station co-coverage relationship, an apparatus for identifying a base station co-coverage relationship, a computer storage medium, and an electronic device. BACKGROUND

[0002] 5G (5th Generation Mobile Communication Technology) base stations have a low actual load, and often need to be deeply hibernated at a low business load to reduce energy consumption. However, while reducing energy consumption, the user experience needs to be ensured, and therefore, the co-coverage relationship between a 5G base station and a nearby 4G (the 4th generation mobile communication technology) base station needs to be determined to ensure user service acceptance during hibernation.

[0003] Currently, the co-coverage relationship between base stations is generally determined based on manually set rules, MR data (Measurement Report), and base station configuration parameters. However, the above scheme can only be applied to the identification of the co-coverage relationship between outdoor base stations, and cannot identify the co-coverage relationship between indoor base stations.

[0004] Therefore, there is an urgent need in the art to develop a new method and apparatus for identifying a base station co-coverage relationship.

[0005] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present disclosure. SUMMARY

[0006] The purpose of the present disclosure is to provide a method for identifying a base station co-coverage relationship, an apparatus for identifying a base station co-coverage relationship, a computer storage medium, and an electronic device, thereby at least partially overcoming the technical problem that the co-coverage relationship between indoor base stations cannot be identified due to the limitations of related technologies.

[0007] Other characteristics and advantages of the present disclosure will become apparent from the following detailed description, or will be learned by practice of the present disclosure.

[0008] According to a first aspect of the present disclosure, a method for identifying base station co-coverage relationship is provided, comprising: obtaining index measurement data of at least two base stations and corresponding basic configuration parameters of each base station; the basic configuration parameters include name information of the base station, the name information includes location description information of the base station, and the location description information includes the height of the base station from a preset reference surface; performing information mining on the name information of each base station to obtain deep hidden information; inputting the index measurement data, the basic configuration parameters and the deep hidden information into a trained co-coverage relationship identification model, and determining whether the at least two base stations have a co-coverage relationship according to the output of the co-coverage relationship identification model.

[0009] In an example embodiment of the present disclosure, the information mining on the name information of each base station to obtain deep hidden information comprises: extracting the location description information of the base station from the name information of each base station; extracting word vector features corresponding to the location description information by using a word vector extraction model; and determining the word vector features as the deep hidden information.

[0010] In an example embodiment of the present disclosure, the information mining on the name information of each base station to obtain deep hidden information further comprises: performing semantic similarity recognition on the location description information of the at least two base stations by using a similarity recognition model to obtain semantic similarity features; and determining the word vector features and the semantic similarity features as the deep hidden information.

[0011] In an example embodiment of the present disclosure, the information mining on the name information of each base station to obtain deep hidden information further comprises: statistically obtaining word frequency features corresponding to the location description information by using a bag-of-words model; and determining the word vector features, the semantic similarity features and the word frequency features as the deep hidden information.

[0012] In an example embodiment of the present disclosure, before obtaining the index measurement data of the at least two base stations and the corresponding basic configuration parameters of each base station, the method further comprises: obtaining training samples; and iteratively training a co-coverage relationship identification model to be trained by using the training samples to obtain the trained co-coverage relationship identification model.

[0013] In the example embodiment of the present disclosure, the training sample is obtained by: obtaining index measurement data of each initial base station in the initial base station pair set and basic configuration parameters corresponding to each initial base station; the basic configuration parameters include name information of the initial base station, the name information includes position description information of the initial base station, and the position description information includes height of the initial base station from a preset reference surface; filtering the initial base station pair set according to the index measurement data and / or the basic configuration parameters to obtain a sample base station pair set; performing information mining on the name information of each sample base station in the sample base station pair set to obtain deep hidden information; and determining the basic configuration parameters of each sample base station, the index measurement data of each sample base station and the deep hidden information as the training sample.

[0014] In the example embodiment of the present disclosure, the initial base station pair set includes at least two initial base station pairs, and each initial base station pair includes at least two initial base stations; the filtering of the initial base station pair set according to the index measurement data and / or the basic configuration parameters to obtain a sample base station pair set includes: traversing each initial base station pair in the initial base station pair set; if a difference value of the index measurement data between any two initial base stations in the initial base station pair is greater than a preset difference value, filtering the initial base station pair from the initial base station pair set; and / or if a position deviation between any two initial base stations in the initial base station pair is greater than a preset deviation value, filtering the initial base station pair from the initial base station pair set; and traversing the initial base station pair set to obtain the sample base station pair set.

[0015] According to a second aspect of the present disclosure, a device for identifying base station co-coverage relationship is provided, including: an obtaining module configured to obtain index measurement data of at least two base stations and basic configuration parameters corresponding to each base station; the basic configuration parameters include name information of the base station, the name information includes position description information of the base station, and the position description information includes height of the base station from a preset reference surface; an information mining module configured to perform information mining on the name information of each base station to obtain deep hidden information; and an identifying module configured to input the index measurement data, the basic configuration parameters and the deep hidden information into a trained co-coverage relationship identification model, and determine whether the at least two base stations have a co-coverage relationship according to an output of the co-coverage relationship identification model.

[0016] According to a third aspect of the present disclosure, a computer storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the method for identifying base station co-coverage relationship according to the first aspect.

[0017] According to a fourth aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the base station co-coverage relationship identification method of the first aspect described above via execution of the executable instructions.

[0018] From the above technical solutions, the base station co-coverage relationship identification method, the base station co-coverage relationship identification device, the computer storage medium and the electronic device in the example embodiments of the present disclosure at least have the following advantages and positive effects:

[0019] In the technical solutions provided in some embodiments of the present disclosure, on the one hand, by setting the location description information of the base station in the name information, and setting the height of the base station from the preset reference surface in the location description information, the height and other factors of the base station can be taken into account, so that the present disclosure can identify whether the base stations set on the indoor building have a co-coverage relationship, solving the technical problem in the related art that only the base stations set outdoors can be identified for the co-coverage relationship, and the co-coverage relationship of indoor base stations cannot be identified. Further, by information mining on the name information of each base station, deep hidden information is obtained, and related parameters can be analyzed and processed in detail. After the training sample is used to train the co-coverage relationship identification model to be trained, the model has more fine-grained data analysis capability, thereby improving the accuracy of the output result of the model. On the other hand, by using the trained co-coverage relationship identification model to identify whether at least two base stations have a co-coverage relationship, the technical problem of low intelligent degree caused by relying on artificial rules to identify the co-coverage relationship between base stations in the related art can be solved, and the intelligent degree and identification efficiency of the identification process are improved.

[0020] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0021] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present disclosure and, together with the specification, serve to explain the principles of the present disclosure. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0022] Figure 1 A flowchart showing how to identify the co-coverage relationship between outdoor base stations in the related art is shown;

[0023] Figure 2 A flowchart showing the base station co-coverage relationship identification method in the embodiments of the present disclosure is shown;

[0024] Figure 3 A flowchart illustrating how to obtain the trained common coverage relationship identification model in the embodiments of the present disclosure is shown.

[0025] Figure 4 A flowchart illustrating how to obtain the training sample in the embodiments of the present disclosure is shown.

[0026] Figure 5 A flowchart illustrating how to obtain the sample base station pair set according to the initial base station pair set in the embodiments of the present disclosure is shown.

[0027] Figure 6 A structural diagram of the Skip-gram model in the embodiments of the present disclosure is shown.

[0028] Figure 7 A structural diagram of the Sentence-Bert model in the embodiments of the present disclosure is shown.

[0029] Figure 8 A flowchart illustrating the overall process of the base station common coverage relationship identification method in the embodiments of the present disclosure is shown.

[0030] Figure 9 A structural diagram of the base station common coverage relationship identification apparatus in the exemplary embodiments of the present disclosure is shown.

[0031] Figure 10 A structural diagram of the electronic device in the exemplary embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0032] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations can be implemented in any

[0033] The terms "one", "a", "an", and "the" are used to mean one or more, unless otherwise indicated. The terms "including", "includes", "have", "has", and "having" are used inclusively and not exclusivity, so that they permit the presence of zero or more of the referenced elements. The terms "first" and "second" are used to denote a difference, not a quantity, so that they are used to distinguish between two elements, regardless of the number of elements.

[0034] In addition, the accompanying drawings are merely schematic and are not intended to be drawn to scale. Identical reference numerals denote similar or identical parts throughout the several views, so that repeated description will be omitted. Some of the block diagrams shown in the drawings are functional entities, which do not necessarily have to correspond to physically or logically independent entities.

[0035] 5G base stations have a relatively low actual load, and often need to be deeply hibernated during a low traffic load period to reduce energy consumption. Enterprises need to determine the co-coverage relationship between 5G base stations and nearby 4G base stations to ensure user service acceptance after hibernation and wake-up, so as to determine whether to enter / exit the hibernation state, thereby reducing energy consumption costs and ensuring user experience.

[0036] Reference Figure 1 , Figure 1 A flowchart is shown to illustrate how to identify the co-coverage relationship between outdoor base stations in the related art, which includes the following steps S101-S102:

[0037] In step S101, measurement reports and basic configuration parameters of the base stations are obtained.

[0038] In step S102, the co-coverage relationship between outdoor base stations is identified based on manually set matching rules.

[0039] However, the above scheme has the following defects:

[0040] First, the above scheme can only focus on the surface data, and the data granularity is large, so it can only identify whether the base stations set outdoors have a co-coverage relationship, and cannot identify the co-coverage relationship of indoor base stations with altitude and floor differences, and cannot be applied to multiple situations;

[0041] Second, the above scheme relies too much on MR data, and the data quality of MR data cannot be guaranteed, so the identification accuracy cannot be guaranteed;

[0042] Third, the above scheme relies on manual rules and does not take advantage of big data, so the degree of intelligence is low. In the embodiments of the present disclosure, a method for identifying the co-coverage relationship of base stations is first provided, which at least partially overcomes the defect that the co-coverage relationship between indoor base stations cannot be identified in the related art.

[0043] Figure 2 A flowchart of a method for identifying a base station co-coverage relationship in an embodiment of the present disclosure is shown. The execution subject of the method for identifying a base station co-coverage relationship can be a server for identifying a co-coverage relationship.

[0044] Reference Figure 2 The method for identifying a base station co-coverage relationship according to one embodiment of the present disclosure includes the following steps:

[0045] In step S210, index measurement data of at least two base stations and corresponding basic configuration parameters of each base station are obtained. The basic configuration parameters include name information of the base station, the name information includes location description information of the base station, and the location description information includes a height of the base station from a preset reference surface.

[0046] In step S220, information mining is performed on the name information of each base station to obtain deep hidden information.

[0047] In step S230, the index measurement data, the basic configuration parameters, and the deep hidden information are input into a trained co-coverage relationship identification model. Whether the at least two base stations have a co-coverage relationship is determined according to an output of the co-coverage relationship identification model.

[0048] In Figure 2 The technical solution provided in the embodiment shown can achieve the following effects. On the one hand, by setting the location description information of the base station in the name information and setting the height of the base station from the preset reference surface in the location description information, the height of the base station and other factors can be taken into account, so that the present disclosure can identify whether the base stations set on indoor buildings have a co-coverage relationship, thereby solving the technical problem in the related art that only the base stations set outdoors can be identified for the co-coverage relationship, and the co-coverage relationship of indoor base stations cannot be identified. Further, by performing information mining on the name information of each base station to obtain deep hidden information, the related parameters can be analyzed and processed in detail. After the training sample is used to train the co-coverage relationship identification model to be trained, the model has a more detailed data analysis capability, thereby improving the accuracy of the output result of the model. On the other hand, whether the at least two base stations have a co-coverage relationship is identified by the trained co-coverage relationship identification model, which can solve the technical problem of low intelligence degree caused by relying on artificial rules to identify the co-coverage relationship between the base stations in the related art, thereby improving the intelligence degree and the identification efficiency of the identification process.

[0049] The specific implementation process of each step in Figure 2 is described in detail as follows:

[0050] The base station is a public mobile communication base station, which is an interface device for mobile devices to access the Internet, and is also a form of radio station. It is a wireless radio station that transmits and receives information between mobile communication exchanges and mobile phone terminals in a certain wireless coverage area.

[0051] It should be noted that before step S210, the present disclosure can pre-train a common coverage relationship identification model, which is used to identify whether at least two base stations have a common coverage relationship, so that when any one of the at least two base stations enters a dormant state, the user service of the base station can be taken over by another base station, thereby ensuring user experience.

[0052] The process of obtaining the trained common coverage relationship identification model in the present disclosure will be described first. Referring to FIG. 3, Figure 3 Figure 3 A flowchart showing how to obtain the trained common coverage relationship identification model in the embodiment of the present disclosure is shown, including steps S301-S302:

[0053] In step S301, the training sample is obtained.

[0054] In this step, the training sample can be obtained. Specifically, refer to Figure 4 Figure 4 A flowchart showing how to obtain the training sample in the embodiment of the present disclosure is shown, including steps S401-S404:

[0055] In step S401, the index measurement data of each initial base station in the initial base station pair set and the corresponding basic configuration parameters of each initial base station are obtained.

[0056] In this step, the initial base station pair set can be obtained from the big data platform. The initial base station pair set can include at least two initial base station pairs, and each initial base station pair can include at least two initial base stations. The at least two initial base stations in each initial base station pair can have a common coverage relationship or can not have a common coverage relationship, i.e., it cannot be determined whether the initial base station pair is valid.

[0057] The index measurement data can be the MR data of each initial base station collected within a predetermined time period (for example, within 14 days before the current time, which can be set by the actual situation, and the present disclosure does not make special limitations on this). MR data is report data used to reflect the true situation of network quality. For example, the MR data can include measurement values of multiple network quality indicators, and can also include the number of user terminals corresponding to each signal strength level, so that the common coverage relationship of the base station can be identified in combination with the signal strength of the terminal, providing more basis for identifying the common coverage relationship of the base station.​​

[0058] The basic configuration parameter can include distance information between at least two initial base stations included in each initial base station pair, and can further include basic information of each initial base station, such as name information, number information, and base station type of each initial base station, and can further include performance parameters of each initial base station, such as stability, functionality, and appearance, and can further include hardware configuration of each initial base station, such as antenna height, antenna gain, antenna half-power angle, antenna azimuth angle, and elevation angle, which can be set as needed, and the present disclosure does not make special limitations thereon.

[0059] It should be noted that the name information can include position description information of the initial base station. For example, the position description information can include longitude and latitude information, an altitude at which the initial base station is located, a name and number of a building at which the initial base station is located (for example, which floor of which building), and a height (for example, a floor) of the initial base station from a preset reference surface (for example, a horizontal surface or the ground).

[0060] By setting the position description information of the initial base station in the name information, and setting the height of the initial base station from the preset reference surface in the position description information, the height and other factors of the base station can be taken into account, so that the present disclosure can identify whether the base stations arranged on indoor buildings have a co-coverage relationship, thereby solving the technical problem that in the related art, only whether the base stations arranged outdoors have a co-coverage relationship can be identified, and the co-coverage relationship of indoor base stations cannot be identified.

[0061] After the initial base station pair set is obtained, the initial base station pair set can be filtered according to the index measurement data and / or the basic configuration parameter to obtain a sample base station pair set in step S402.

[0062] In this step, the reference Figure 5 , Figure 5 A flowchart showing how the sample base station pair set is obtained from the initial base station pair set in the embodiment of the present disclosure is shown, including steps S501-S504:

[0063] In step S501, each initial base station pair in the initial base station pair set is traversed.

[0064] In this step, each initial base station pair in the initial base station pair set can be traversed. For example, the traversal order can be from front to back, from back to front, or random order, which can be set as needed, and the present disclosure does not make special limitations thereon.

[0065] In step S502, if the difference value of the index measurement data between any two initial base stations in the initial base station pair is greater than the preset difference value, the initial base station pair is filtered from the initial base station pair set.

[0066] In this step, in view of the index measurement data of each initial base station in the above basic configuration parameter, the difference value of the index measurement data between at least two initial base stations in the initial base station pair can be compared according to any one or more indexes in the above index measurement data. For example, assuming that the initial base station pair "initial base station A-initial base station B" is included in the initial base station pair set, and the difference value between the index measurement data of the initial base station A and the index measurement data of the initial base station B is greater than the preset difference value, it can be determined that the initial base station A and the initial base station B do not have a co-coverage relationship, and thus the initial base station pair "initial base station A-initial base station B" can be filtered from the initial base station pair set.

[0067] In step S503, if the position deviation between any two initial base stations in the initial base station pair is greater than the preset deviation value, the initial base station pair is filtered from the initial base station pair set.

[0068] In this step, the position deviation can be a distance deviation, a longitude deviation, a latitude deviation, etc., which can be set according to actual conditions, and the present disclosure does not make special limitations thereon.

[0069] Taking the distance deviation as an example of the position deviation, the preset deviation value can be set to 5 kilometers (which can be set according to actual conditions, and the present disclosure does not make special limitations thereon). Thus, assuming that the initial base station pair "initial base station C-initial base station D" is included in the initial base station pair set, and the distance between the initial base station C and the initial base station D is detected to be 8 kilometers, which is greater than the preset deviation value of 5 kilometers, the initial base station pair "initial base station C-initial base station D" can be filtered from the initial base station pair set.

[0070] Taking the longitude deviation as an example of the position deviation, the preset deviation value can be set to 0.1 degrees (which can be set according to actual conditions, and the present disclosure does not make special limitations thereon). Assuming that the initial base station pair "initial base station E-initial base station F" is included in the initial base station pair set, and the longitude deviation between the initial base station E and the initial base station F is detected to be 10 degrees, which is greater than the preset deviation value of 0.1 degrees, the initial base station pair "initial base station E-initial base station F" can be filtered from the initial base station pair set.

[0071] Taking the latitude deviation as an example, the preset deviation value can be set to 0.1 degrees (which can be set by the actual situation, and the present disclosure does not make special limitations on this), assuming that the initial base station pair set contains the initial base station pair "initial base station G-initial base station H", and the longitude difference between the initial base station G and the initial base station H is detected to be 10 degrees, which is greater than the preset deviation value 0.1 degrees, so that the initial base station pair "initial base station G-initial base station H" can be filtered from the initial base station pair set.

[0072] In step S504, until the initial base station pair set is traversed, the sample base station pair set is obtained.

[0073] In this step, the sample base station pair set can be obtained after the initial base station pair set is traversed.

[0074] After obtaining the sample base station pair set, the name information of each sample base station contained in the sample base station pair set can be information mined in step S403 to obtain deep hidden information.

[0075] In this step, the sample base station pair set can contain multiple sample base station pairs, and each sample base station pair contains at least two sample base stations.

[0076] In an optional embodiment, the position description information (i.e., the above-mentioned latitude and longitude information, altitude, name and number of buildings, floor, etc.) of each sample base station can be extracted from the name information of the sample base station based on a regular expression, and then the word vector feature corresponding to the position description information is extracted using a word vector extraction model, and the word vector feature is determined as the deep hidden information.

[0077] The word vector extraction model can be a Skip-gram model, which is a model for predicting the context probability according to the current word, and is an algorithm for predicting m words in the context according to the center word, where m is a prediction window size defined by the user. For example, referring to Figure 6 , Figure 6 The structure diagram of the Skip-gram model in the embodiment of the present disclosure is shown as follows: Figure 6 As shown in the figure, in the case of inputting the center word w(t), assuming that the given window is 2, then the Skip-gram model can output two words w(t-1) and w(t-2) before w(t), and output two words w(t+1) and w(t+2) after w(t).

[0078] The Skip-gram model is essentially a Word2Vec model. Word2Vec can represent a word as a word embedding, which is low-dimensional and dense. Specifically, the descriptive information at each position is first segmented into words. Then, each segment is one-hot encoded to obtain the latent features corresponding to that word. Afterward, the Skip-gram model can perform two linear transformations and one softmax process on these latent features to obtain the word vector features corresponding to each segment (e.g., a 300-dimensional word vector).

[0079] In another optional implementation, after extracting the word vector features corresponding to the location description information using the word vector extraction model, the semantic similarity of the location description information of at least two base stations can be identified using the similarity recognition model to obtain semantic similarity features. The word vector features and the semantic similarity features are then identified as deep hidden information.

[0080] For example, the similarity recognition model described above could be the Sentence-Bert model. The Sentence-Bert model uses a Siamese network or triplet network structure to generate sentence vectors, allowing semantic similarity of sentences to be measured using cosine similarity. (See reference) Figure 7 , Figure 7 The diagram shows the structure of the Sentence-Bert model in an embodiment of this disclosure, as follows: Figure 7 As shown, taking the example of "sample base station X - sample base station Y", the location description information of sample base station X can be regarded as sentence A, and the location description information of sample base station Y can be regarded as sentence B. Sentence A and sentence B are input into the BERT model so that the BERT model performs pooling operations (also known as undersampling or downsampling, mainly used for feature dimensionality reduction, compressing the amount of data and parameters, reducing overfitting, and improving the fault tolerance of the model) on sentence A and sentence B respectively, to obtain the corresponding vectors u and v of sentence A and sentence B respectively. Then, the cosine similarity of vectors u and v (the range of cosine similarity is [-1,1]) is calculated. This cosine similarity is the semantic similarity feature mentioned above.

[0081] The cosine similarity, also known as the cosine similarity, is to evaluate the similarity of two vectors by calculating the cosine value of the included angle. The cosine similarity draws vectors according to coordinate values into a vector space, such as the most common two-dimensional space. The cosine value of 0 degrees is 1, while the cosine value of any other angle is not greater than 1, and its minimum value is -1. Thus, the cosine value of the angle between two vectors determines whether the two vectors point in the same direction.

[0082] In still another optional embodiment, after obtaining the semantic similarity feature, the word frequency feature corresponding to the location description information can also be obtained by using the bag-of-words model. The word vector feature, the semantic similarity feature and the word frequency feature are determined as the deep hidden information.

[0083] For example, the bag-of-words model can first segment the location description information of each sample base station. After segmentation, the number of occurrences of each word in the text is counted, and according to the number of occurrences of multiple segmented words, the word frequency feature corresponding to the above location description information can be obtained.

[0084] In step S404, the basic configuration parameters of each sample base station, the index measurement data of each sample base station and the deep hidden information are determined as training samples.

[0085] In this step, the basic configuration parameters of each sample base station, the index measurement data of each sample base station and the deep hidden information are determined as training samples.

[0086] Since the training samples of the present disclosure not only contain the basic configuration parameters of each sample base station and the index measurement data of each sample base station, but also contain the word vector feature extracted by the word vector extraction model, the semantic similarity feature mined by the similarity recognition model, and the word frequency feature counted by the bag-of-words model, the present disclosure can perform fine-grained analysis and processing on related parameters. After training the to-be-trained co-coverage relationship recognition model using the training samples, the model has finer-grained data analysis capability, thereby improving the accuracy of the model output result.

[0087] After obtaining the training samples, the following steps are referred to Figure 3 In step S302, the training samples are used to iteratively train the to-be-trained co-coverage relationship recognition model to obtain a trained co-coverage relationship recognition model.

[0088] In this step, the above training samples can be input into the to-be-trained co-coverage relationship recognition model to iteratively train the above to-be-trained co-coverage relationship recognition model until the loss function of the above to-be-trained co-coverage relationship recognition model converges, and a trained co-coverage relationship recognition model is obtained.

[0089] The to-be-trained common coverage relationship identification model can be a random forest model, a decision tree model, a logistic regression model, an XGBoost model (eXtreme Gradient Boosting), or the like, which can be set as needed, and the present disclosure does not make special limitations on this.

[0090] Reference is made to Figure 2 In step S210, index measurement data of at least two base stations and basic configuration parameters corresponding to each base station are obtained.

[0091] In this step, after the common coverage relationship identification model is trained, when it is needed to determine whether there is a common coverage relationship between at least two base stations, the index measurement data of the at least two base stations and the basic configuration parameters corresponding to each base station can be obtained first.

[0092] The index measurement data can be MR data of each base station in the at least two base stations collected in a preset time period (for example, 14 days before the current time, which can be set as needed, and the present disclosure does not make special limitations on this). The MR data is report data for reflecting the true situation of network quality. For example, the MR data can include measurement values of multiple network quality indicators, and can also include the number of user terminals corresponding to each signal strength level, so that the model can combine the signal strength of the terminal to identify the common coverage relationship of the base station, and provide more basis for identifying the common coverage relationship of the base station.

[0093] The basic configuration parameters can include basic information of each base station in the at least two base stations, such as name information, number information, and base station type of each base station. The basic configuration parameters can also include performance parameters of each base station, such as stability, functionality, and appearance. The basic configuration parameters can also include hardware configurations of each base station, such as antenna height, antenna gain, antenna half-power angle, antenna azimuth angle, and elevation angle, which can be set as needed, and the present disclosure does not make special limitations on this.

[0094] It should be noted that the name information can include location description information of the base station. For example, the location description information can include latitude and longitude information, an altitude at which the base station is located, a name and number of a building in which the base station is located (for example, which floor of which building), and a height (for example, a floor) of the base station from a preset reference surface (for example, a horizontal surface or the ground).

[0095] By setting the location description information of the base station in the name information, and setting the height of the base station from the preset reference surface in the location description information, the height and other factors of the base station can be taken into account, so that the disclosure can identify whether the base stations set on the indoor building have a co-coverage relationship, solving the technical problem that in the related art, only whether the base stations set outdoors have a co-coverage relationship can be identified, and the co-coverage relationship of indoor base stations cannot be identified.

[0096] In step S220, the name information of each base station is information mined to obtain deep hidden information.

[0097] In this step, the name information of each base station in the above at least two base stations can be information mined to obtain deep hidden information, as explained in step S403.

[0098] In step S230, the index measurement data, the basic configuration parameters and the deep hidden information are input into the trained co-coverage relationship identification model, and according to the output of the co-coverage relationship identification model, it is determined whether the at least two base stations have a co-coverage relationship.

[0099] In this step, the index measurement data, the basic configuration parameters and the deep hidden information of each base station in the above at least two base stations can be input into the above trained co-coverage relationship identification model, so that the above trained co-coverage relationship identification model can predict whether the above at least two base stations have a co-coverage relationship based on the above information.

[0100] Overall, the disclosure mines the third-dimensional information such as altitude, floor, and building number in the base station name, proposes a new scheme for co-coverage relationship identification of indoor base stations, improves the identification accuracy of the co-coverage relationship between indoor (especially high floor) base stations, and reduces the dependence on artificial rules.

[0101] Reference Figure 8 , Figure 8 The overall flowchart of the identification method of the co-coverage relationship of the base station in the embodiment of the disclosure is shown, which includes steps S801-S810:

[0102] In step S801, an initial base station pair set is obtained;

[0103] In step S802, coarse screening is performed to obtain a sample base station pair set;

[0104] In step S803, the name information of each sample base station in the sample base station pair set is extracted into a word vector;

[0105] In step S804, a word vector feature is obtained;

[0106] In step S805, the name information of each sample base station in the sample base station pair set is subjected to word frequency statistics;

[0107] In step S806, the word frequency feature is obtained;

[0108] In step S807, the name information of each sample base station in the sample base station pair set is subjected to semantic similarity recognition;

[0109] In step S808, the semantic similarity feature is obtained;

[0110] In step S809, the index measurement data, the basic configuration parameter, the word vector feature, the semantic similarity feature and the word frequency feature of each sample base station are used as training samples to train the random forest model, and a trained common coverage relationship recognition model is obtained;

[0111] In step S810, whether the indoor base stations have a common coverage relationship is identified by using the trained common coverage relationship recognition model.

[0112] Based on the above technical solutions, the present disclosure has at least the following technical effects:

[0113] First, the recognition accuracy of the common coverage relationship between indoor base stations is high. By introducing the word vector feature, the semantic similarity feature and the floor number and other information, the present disclosure can analyze and process related parameters in a fine-grained manner. After training the common coverage relationship recognition model to be trained using the training samples, the model has a finer-grained data analysis capability, thereby improving the accuracy of the model output result;

[0114] Second, the machine learning algorithm is introduced instead of the artificial rule. According to the data distribution of each region / province / city in the training sample, potential rules are learned and mined, and the artificial one-size-fits-all recognition mode is avoided.

[0115] The present disclosure also provides a base station common coverage relationship recognition device, Figure 9 The structure of the base station common coverage relationship recognition device in the exemplary embodiment of the present disclosure is shown in the schematic diagram; as Figure 9 As shown, the base station common coverage relationship recognition device 900 can include an acquisition module 910, an information mining module 920 and an identification module 930. Among them:

[0116] The acquisition module 910 is configured to acquire index measurement data of at least two base stations and corresponding basic configuration parameters of each base station; the basic configuration parameters include name information of the base station, the name information includes location description information of the base station, and the location description information includes a height of the base station from a preset reference surface;

[0117] The information mining module 920 is configured to mine information from the name information of each base station to obtain deep hidden information.

[0118] The recognition module 930 is configured to input the index measurement data, the basic configuration parameters and the deep hidden information into a trained common coverage relationship recognition model, and determine whether the at least two base stations have a common coverage relationship according to an output of the common coverage relationship recognition model.

[0119] In an example embodiment of the present disclosure, the information mining module 920 mines information from the name information of each base station to obtain deep hidden information, including: extracting location description information of the base station from the name information of each base station; extracting a word vector feature corresponding to the location description information by using a word vector extraction model; and determining the word vector feature as the deep hidden information.

[0120] In an example embodiment of the present disclosure, the information mining module 920 mines information from the name information of each base station to obtain deep hidden information, and further includes: performing semantic similarity recognition on the location description information of the at least two base stations by using a similarity recognition model to obtain a semantic similarity feature; and determining the word vector feature and the semantic similarity feature as the deep hidden information.

[0121] In an example embodiment of the present disclosure, the information mining module 920 mines information from the name information of each base station to obtain deep hidden information, and further includes: counting a word frequency feature corresponding to the location description information by using a bag-of-words model; and determining the word vector feature, the semantic similarity feature and the word frequency feature as the deep hidden information.

[0122] In an example embodiment of the present disclosure, before the index measurement data of the at least two base stations and the basic configuration parameters corresponding to each base station are acquired, the acquisition module 910 is configured to:

[0123] acquire training samples; and perform iterative training on a to-be-trained common coverage relationship recognition model by using the training samples to obtain the trained common coverage relationship recognition model.

[0124] In the example embodiment of the present disclosure, the obtaining module 910 obtains the training sample, including: obtaining index measurement data of each initial base station in the initial base station pair set and the basic configuration parameter corresponding to each initial base station; the basic configuration parameter contains name information of the initial base station, the name information contains position description information of the initial base station, and the position description information contains height of the initial base station from a preset reference surface; filtering the initial base station pair set according to the index measurement data and / or the basic configuration parameter to obtain a sample base station pair set; performing information mining on the name information of each sample base station in the sample base station pair set to obtain deep hidden information; and determining the basic configuration parameter of each sample base station, the index measurement data of each sample base station and the deep hidden information as the training sample.

[0125] In the example embodiment of the present disclosure, the initial base station pair set contains at least two initial base station pairs, and each initial base station pair contains at least two initial base stations; the obtaining module 910 filters the initial base station pair set according to the index measurement data and / or the basic configuration parameter to obtain a sample base station pair set, including: traversing each initial base station pair in the initial base station pair set; if a difference value of the index measurement data between any two initial base stations in the initial base station pair is greater than a preset difference value, filtering the initial base station pair from the initial base station pair set; and / or, if a position deviation between any two initial base stations in the initial base station pair is greater than a preset deviation value, filtering the initial base station pair from the initial base station pair set; until the initial base station pair set is traversed to obtain the sample base station pair set.

[0126] The specific details of each module in the above base station co-coverage relationship identification device have been described in detail in the corresponding base station co-coverage relationship identification method, and therefore will not be described here.

[0127] It should be noted that although several modules or units of the device for action execution are mentioned in the above detailed description, such division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into embodied by multiple modules or units.

[0128] Moreover, although individual steps of the methods in the present disclosure are described in a particular order in the figures, this is not required or implied as to the order of execution of the steps, nor is it required that all of the steps be performed to achieve the desired result. Additionally or alternatively, certain steps can be omitted, multiple steps can be combined into a single step, a single step can be broken into multiple steps, etc.

[0129] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.) or a network, and includes a number of instructions to make a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) execute the methods according to the embodiments of the present disclosure.

[0130] The present application also provides a computer readable storage medium, which can be included in the electronic device described in the above embodiments, or can exist separately without being assembled into the electronic device.

[0131] The computer readable storage medium may, for example, be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this disclosure, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0132] The computer readable storage medium can send, propagate, or transmit programs for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer readable storage medium can be transmitted by any suitable medium, including, but not limited to, wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0133] The computer readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to implement the methods described in the above embodiments.

[0134] Further, an electronic device capable of implementing the above method is also provided in the embodiments of the present disclosure.

[0135] Those skilled in the art can understand that each aspect of the present disclosure can be implemented as a system, a method or a program product. Therefore, each aspect of the present disclosure can be specifically implemented as a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as "circuitry", "module" or "system" here.

[0136] The electronic device 1000 according to this embodiment of the present disclosure will be described below with reference to Figure 10 Figure 10 The electronic device 1000 shown is merely an example and should not impose any limitation on the functions and application scope of the embodiments of the present disclosure.

[0137] As shown in Figure 10 The components of the electronic device 1000 can include, but are not limited to, the at least one processing unit 1010, the at least one storage unit 1020, a bus 1030 connecting different system components (including the storage unit 1020 and the processing unit 1010), and a display unit 1040.

[0138] The storage unit stores program codes which can be executed by the processing unit 1010, so that the processing unit 1010 performs the steps according to various exemplary embodiments of the present disclosure described in the "Exemplary Method" section of the present specification. For example, the processing unit 1010 can perform the steps shown in Figure 2 S210, obtaining index measurement data of at least two base stations and corresponding basic configuration parameters of each base station; the basic configuration parameters include name information of the base station, the name information includes position description information of the base station, and the position description information includes the height of the base station from a preset reference surface; S220, performing information mining on the name information of each base station to obtain deep hidden information; S230, inputting the index measurement data, the basic configuration parameters and the deep hidden information into a trained common coverage relationship identification model, and determining whether the at least two base stations have a common coverage relationship according to the output of the common coverage relationship identification model.

[0139] The storage unit 1020 can include a readable medium in the form of a volatile storage unit, such as a random access memory (RAM) 10201 and / or a cache memory 10202, and can further include a read-only memory (ROM) 10203. ​

[0140] The storage unit 1020 can also include a program / utility 10204 having a set of programs / modules 10205, each of which performs one or more of the operations / acts as described herein. These programs / modules 10205 include, but are not limited to, one or more of: an operating system. one or more application programs, other program modules, and program data, each of which or a combination of which can include implementation of a networking environment.

[0141] The bus 1030 can represent one or more of several types of bus structures, including a storage bus or bus controller, a peripheral bus, a graphics bus, a processor or local bus using any of a variety of bus architectures.

[0142] The electronic device 1000 can also communicate with one or more external devices 900 such as a keyboard or pointing device, a Bluetooth device, etc.; other devices such as a storage device or an external effects device (not shown); and / or one or more devices that enable a user to interact with the electronic device 1000 in some way (not shown) and / or one or more devices that enable the electronic device 1000 to communicate with one or more other computing devices. Such communication can be via input / output (I / O) interface(s) 1050. Still yet, the electronic device 1000 can communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or the Internet) through a network adapter 1060. As depicted, the network adapter 1060 communicates with the other components of the electronic device 1000 via the bus 1030. It should be appreciated that although not shown, other hardware and / or software components could be used in conjunction with the electronic device 1000. These include, but are not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.

[0143] Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the features disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the disclosure being indicated by the following claims.

Claims

1. A method for identifying co-coverage relationship of base stations, characterized in that, The method comprises the following steps: obtaining index measurement data of at least two base stations and corresponding basic configuration parameters of each base station; the basic configuration parameters include name information of the base station, the name information includes location description information of the base station, and the location description information includes building identification information of a building where the base station is located and a height of the base station from a preset reference surface; information mining is performed on the name information of each base station to obtain deep hidden information; the index measurement data, the basic configuration parameters and the deep hidden information are input into a trained common coverage relationship identification model, and whether the at least two base stations have a common coverage relationship is determined according to an output of the common coverage relationship identification model.

2. The method of claim 1, wherein, The information mining performed on the name information of each base station to obtain deep hidden information comprises the following steps: extracting the location description information of the base station from the name information of each base station; extracting word vector features corresponding to the location description information by using a word vector extraction model; determining the word vector features as the deep hidden information.

3. The method of claim 2, wherein, The information mining performed on the name information of each base station to obtain deep hidden information further comprises the following steps: performing semantic similarity recognition on the location description information of the at least two base stations by using a similarity recognition model to obtain semantic similarity features; determining the word vector features and the semantic similarity features as the deep hidden information.

4. The method of claim 3, wherein, The information mining performed on the name information of each base station to obtain deep hidden information further comprises the following steps: statistically obtaining word frequency features corresponding to the location description information by using a bag-of-words model; determining the word vector features, the semantic similarity features and the word frequency features as the deep hidden information.

5. The method according to any one of claims 1 to 4, characterized in that, Before the index measurement data of the at least two base stations and the corresponding basic configuration parameters of each base station are obtained, the method further comprises the following steps: obtaining training samples; iteratively training a common coverage relationship identification model to be trained by using the training samples to obtain the trained common coverage relationship identification model.

6. The method of claim 5, wherein, The obtaining of the training samples comprises the following steps: obtaining index measurement data of each initial base station in an initial base station pair set and corresponding basic configuration parameters of each initial base station; the basic configuration parameters include name information of the initial base station, the name information includes location description information of the initial base station, and the location description information includes a height of the initial base station from a preset reference surface; filtering the initial base station pair set according to the index measurement data and / or the basic configuration parameters to obtain a sample base station pair set; performing information mining on the name information of each sample base station in the sample base station pair set to obtain deep hidden information; determining the basic configuration parameters of each sample base station, the index measurement data of each sample base station and the deep hidden information as the training samples.

7. The method of claim 6, wherein, The initial base station pair set includes at least two initial base station pairs, and each initial base station pair includes at least two initial base stations. The filtering of the initial base station pair set according to the index measurement data and / or the basic configuration parameter to obtain a sample base station pair set comprises: traversing each initial base station pair in the initial base station pair set; if the difference value of the index measurement data between any two initial base stations in the initial base station pair is greater than a preset difference value, filtering the initial base station pair from the initial base station pair set; and / or, if the position deviation between any two initial base stations in the initial base station pair is greater than a preset deviation value, filtering the initial base station pair from the initial base station pair set; until the initial base station pair set is traversed to obtain the sample base station pair set.

8. A device for identifying the co-coverage relationship of base stations, characterized in that, comprise: an acquisition module configured to acquire index measurement data of at least two base stations and basic configuration parameters corresponding to each of the base stations; the basic configuration parameters include name information of the base stations, the name information includes position description information of the base stations, and the position description information includes building identification information where the base stations are located and height of the base stations from a preset reference surface; an information mining module configured to mine the name information of each of the base stations to obtain deep hidden information; an identification module configured to input the index measurement data, the basic configuration parameters and the deep hidden information into a trained common coverage relationship identification model, and determine whether the at least two base stations have a common coverage relationship according to an output of the common coverage relationship identification model.

9. A computer storage medium having stored thereon a computer program, characterized in that The computer program is executed by a processor to implement the base station common coverage relationship identification method in any one of claims 1-7.

10. An electronic device, comprising: comprise: a processor; and a memory configured to store executable instructions of the processor; wherein the processor is configured to implement the base station common coverage relationship identification method in any one of claims 1-7 by executing the executable instructions.

Images