Base station identification method, device, electronic device and computer readable storage medium

Abstract

The present invention provides a base station identification method, device, electronic device and computer-readable storage medium, which relate to the field of smart transportation. First, according to the mobile phone signaling data set of the specified area within the specified time period, at least one travel trajectory of each travel user is determined, and the candidate base station corresponding to each subway station in the specified area is found, and then based on the dedicated base station information, site information, land type composition information and all travel trajectories of the specified area, the feature data of each candidate base station is determined. Compared with the prior art, the present invention extracts the feature data of each candidate base station, which reflects the type attributes of the candidate base station, the change in the number of service passengers in the specified time period and the surrounding land type composition, so that finally, the effective subway base station providing services for the subway station can be identified according to the feature data of all candidate base stations, which can effectively reduce the error of subsequent subway travel identification.

Description

Technical Field

[0001] The present invention relates to the field of smart transportation, and in particular to a base station identification method, device, electronic device and computer-readable storage medium. Background Art

[0002] The identification of transportation modes is of great practical significance for understanding the changes in the travel structure of urban residents and optimizing the supply of transportation service facilities. Accurately determining the travel mode of residents is a key part of urban transportation planning and management. The proportion of green transportation is also an important indicator of urban physical examination, and the indicators in the field of urban physical examination also include the population coverage rate within 800m around the track. There is no doubt that rail transit is an important part of green travel and reducing carbon emissions from motorized travel, and rail transit has always been an important part of infrastructure construction in megacities and super-large cities. Under the current background of high-quality development and stock operation, it is particularly urgent to accurately identify the service scope of rail stations and improve the utilization efficiency of rails.

[0003] Traditional rail transit travel identification mainly relies on traffic surveys, but the current surveys have the characteristics of long cycle (once every 5 years), low frequency and high cost, and cannot reflect the latest traffic demand dynamics in a timely manner: Taking Beijing as an example, since the first comprehensive traffic survey was implemented in 1986, it has been fixed every 5 years since 2000, and only 5 major surveys have been completed so far; as for Tianjin, it conducted the first comprehensive traffic survey in the country in 1981, and conducted surveys in 1993, 2000, 2011 and 2017 respectively.

[0004] In recent years, the emerging IC (Integrated Circuit) card data has provided good support for the identification of rail transit travel volume, but this data can only identify the OD of rail travel within the track (O represents the origin and D represents the destination).

[0005] At present, with the development of mobile Internet technology, the phenomenon of using various subway apps to scan QR codes to enter and exit subway stations has become increasingly common, which has also affected the reliability of IC card data to a certain extent. For this reason, some scholars have begun to try to identify the user's complete rail travel chain information based on mobile phone signaling data and subway dedicated base stations, and have achieved certain results.

[0006] However, in actual situations, only underground subway stations often have dedicated subway base stations, while most ground subway stations (especially ground subway stations in the suburbs) do not have dedicated base stations. Therefore, existing technologies are unable to determine the service base stations used by ground subway stations, resulting in large subsequent identification errors for rail travel chains between suburbs and between suburbs and central urban areas, affecting the scientific nature of decision-making. Summary of the invention

[0007] The object of the present invention is to provide a base station identification method, device, electronic device and computer-readable storage medium to improve the problems existing in the prior art.

[0008] The embodiments of the present invention can be implemented as follows:

[0009] In a first aspect, the present invention provides a base station identification method, comprising:

[0010] Determine at least one travel trajectory of each travel user according to a mobile phone signaling data set in a specified area within a specified time period; the travel trajectory includes multiple base stations where the travel user uses a mobile phone to perform command interaction along the way;

[0011] Determine at least one candidate base station within a deployment radius around each subway station in the designated area;

[0012] Based on the dedicated base station information and site information of the designated area, all the travel trajectories and the land use type composition information of the designated area, determining the characteristic data of each candidate base station of each subway station in the designated area; the characteristic data reflects the type attribute of the candidate base station, the change of the number of service passengers in the designated time period and the surrounding land use type composition;

[0013] Based on the characteristic data of each candidate base station of each subway station in the designated area, all effective subway base stations are identified from all candidate base stations, and the subway station served by each effective subway base station is determined.

[0014] Optionally, the step of determining at least one candidate base station within a deployment radius around each subway station in the designated area includes:

[0015] Acquire the dedicated base station information and the site information; the dedicated base station information includes the location of each dedicated subway base station in the designated area and the subway station it serves; the site information includes the location of each subway station in the designated area;

[0016] Based on the location of each of the dedicated subway base stations and the location of each subway station it serves, calculating the distribution distance between each of the dedicated subway base stations and the subway stations it serves;

[0017] Based on all the distribution distances, determine the layout radius between the dedicated subway base station and the subway station it serves;

[0018] For each subway station in the designated area, all base stations within a deployment radius around the subway station are used as candidate base stations of the subway station, so as to obtain each candidate base station of each subway station in the designated area.

[0019] Optionally, the dedicated base station information includes multiple dedicated subway base stations and subway stations they serve; the site information includes the site type of each subway station in the designated area; the site types include ground stations and underground stations; the feature data includes attribute labels and comprehensive feature vectors;

[0020] The step of determining the characteristic data of each candidate base station of each subway station in the designated area based on the dedicated base station information and site information of the designated area, all the travel trajectories and the land use type composition information of the designated area comprises:

[0021] Based on the dedicated base station information and the site information, determine the attribute label of each candidate base station corresponding to each subway station in the designated area; the attribute label is one of 0, 1, 2, and 3; wherein, the attribute label of 0 represents that the subway station corresponding to the candidate base station is an underground station and there is another dedicated subway base station; the attribute label of 1 represents that the subway station corresponding to the candidate base station is an underground station, and the candidate base station is a dedicated subway base station of the corresponding subway station; the attribute label of 2 represents that the subway station corresponding to the candidate base station is a ground station, and the candidate base station is a dedicated subway base station of the corresponding subway station; the attribute label of 3 represents that there is no dedicated subway base station at the subway station corresponding to the candidate base station;

[0022] Based on all the travel trajectories and the land use type composition information, a comprehensive feature vector of each candidate base station corresponding to each subway station in the designated area is determined.

[0023] Optionally, the land use type composition information includes the locations of several points of interest in the designated area and the land use types to which they belong;

[0024] The step of determining the comprehensive feature vector of each candidate base station corresponding to each subway station in the designated area based on all the travel trajectories and the land use type composition information comprises:

[0025] Dividing the specified time period into a plurality of particle time periods according to a preset period;

[0026] For each candidate base station corresponding to each subway station, the number of people served by the candidate base station in each granular time period is counted from all travel trajectories to obtain a first feature vector of the candidate base station;

[0027] Within a deployment radius centered on the candidate base station, counting the number of points of interest of each land use type to obtain a second feature vector of the candidate base station;

[0028] The first feature vector and the second feature vector of the candidate base station are normalized and then concatenated to obtain a comprehensive feature vector of the candidate base station.

[0029] Optionally, the step of identifying all effective subway base stations from all candidate base stations based on the feature data of each candidate base station of each subway station in the designated area, and determining the subway station served by each of the effective subway base stations includes:

[0030] Generate a classification recognition model based on the comprehensive feature vectors and attribute labels of all candidate base stations whose attribute labels are 0, 1 and 2;

[0031] Using the classification and recognition model, all specific subway base stations are screened out from all candidate base stations with the attribute label 3; the specific subway base stations represent base stations that interact with terminals of users traveling at nearby subway stations;

[0032] All candidate base stations with attribute labels 1 and 2 and all specific subway base stations are regarded as effective subway base stations;

[0033] The subway station served by each candidate base station with the attribute label 1 is found out from the dedicated base station information, and the service object of each specific subway base station is set to the nearest subway station around it, so as to obtain the subway station served by each effective subway base station.

[0034] Optionally, the step of generating a classification recognition model based on the comprehensive feature vectors and attribute labels of all candidate base stations whose attribute labels are 0, 1, and 2 includes:

[0035] Constructing a verification set according to all candidate base stations with the attribute label 2, wherein the verification set includes a plurality of verification samples, each of which includes a comprehensive feature vector and an attribute label of a candidate base station;

[0036] Using all candidate base stations whose attribute labels are 0 and 1, a data set is constructed; the data set includes a plurality of training samples, each training sample includes a comprehensive feature vector and an attribute label of a candidate base station;

[0037] Dividing the data set into a training set and a test set according to a preset ratio;

[0038] The pre-constructed random forest model is trained using the training set, the validation set, and the test set to obtain the classification recognition model.

[0039] Optionally, the step of using the classification recognition model to filter out all specific subway base stations from all candidate base stations with the attribute label 3 includes:

[0040] All candidate base stations with the attribute label of 3 are taken as base stations to be identified;

[0041] Inputting the comprehensive feature vector of each base station to be identified into the classification recognition model to obtain a classification label of each base station to be identified; the classification label is 0 or 1;

[0042] All the specific subway base stations with the classification label 1 are screened out from all the base stations to be identified.

[0043] Optionally, the method further includes:

[0044] Using all effective subway base stations and subway stations they serve, each subway travel trajectory that meets preset characteristic conditions is selected from all the travel trajectories;

[0045] Determine the entry subway station and the exit subway station corresponding to each of the subway travel trajectories;

[0046] According to the entry subway station and exit subway station corresponding to each subway travel trajectory, the service scope information of each subway station in the designated area is determined, and / or the subway travel passenger flow in the designated area during the designated time period is determined, and / or the passenger flow of each subway station in the designated area during the designated time period is counted.

[0047] Optionally, the service range information includes an entry service radius and an exit service radius; the step of determining the service range information of each subway station in the designated area according to the entry subway station and the exit subway station corresponding to each subway travel trajectory includes:

[0048] For each of the subway travel trajectories, the inbound distance between the first base station of the subway travel trajectory and the inbound subway station and the outbound distance between the outbound subway station of the subway travel trajectory and the last base station are calculated to obtain the inbound distance and the outbound distance corresponding to each of the subway travel trajectories;

[0049] For each of the subway stations, respectively find a first travel trajectory and a second travel trajectory with the subway station as an entry subway station and an exit subway station from all subway travel trajectories;

[0050] The average of the entry distances corresponding to all the first travel trajectories of the subway station is used as the entry service radius of the subway station;

[0051] The mean of the exit distances corresponding to all the second travel trajectories of the subway station is used as the exit service radius of the subway station;

[0052] Each subway station is traversed to obtain the inbound service radius and the outbound service radius of each subway station.

[0053] Optionally, the service range information includes an inbound heat map and an outbound heat map;

[0054] The step of determining the service range information of each subway station in the designated area according to the entry subway station and the exit subway station corresponding to each subway travel trajectory comprises:

[0055] For each subway travel trajectory, the positions of the first base station and the last base station of the subway travel trajectory are used as the travel starting point and the travel end point;

[0056] For each of the subway stations, respectively find a first travel trajectory and a second travel trajectory with the subway station as an entry subway station and an exit subway station from all subway travel trajectories;

[0057] Taking the position of the subway station in the map as the center, and according to the position of the first base station of each first travel trajectory corresponding to the subway station in the map, generating an entry heat map of the subway station;

[0058] Taking the position of the subway station in the map as the center, and according to the position of the last base station of each second travel trajectory corresponding to the subway station in the map, generating an exit heat map of the subway station;

[0059] Traverse each of the subway stations to obtain an inbound heat map and an outbound heat map of each subway station.

[0060] In a second aspect, the present invention provides a base station identification device, comprising:

[0061] A trajectory collection module, used to determine at least one travel trajectory of each travel user according to a mobile phone signaling data set in a specified area within a specified time period; the travel trajectory includes multiple base stations where the travel user uses a mobile phone to interact with commands along the way;

[0062] A base station screening module, used to determine at least one candidate base station within a deployment radius around each subway station in the designated area;

[0063] A feature extraction module is used to determine feature data of each candidate base station of each subway station in the specified area based on the dedicated base station information and station information of the specified area, all the travel trajectories and the land use type composition information of the specified area; the feature data reflects the type attribute of the candidate base station, the change of the number of service passengers in the specified time period and the surrounding land use type composition;

[0064] The base station identification module is used to identify all effective subway base stations from all candidate base stations based on the feature data of each candidate base station of each subway station in the designated area, and determine the subway station served by each effective subway base station.

[0065] In a third aspect, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory stores a software program, and when the electronic device is running, the processor executes the software program to implement the base station identification method as described in the first aspect above.

[0066] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the base station identification method described in the first aspect is implemented.

[0067] Compared with the prior art, the embodiments of the present invention provide a base station identification method, device, electronic device and computer-readable storage medium. First, according to the mobile phone signaling data set in the specified area within the specified time period, at least one travel trajectory of each travel user is determined, and the travel trajectory includes multiple base stations where the travel user uses the mobile phone to interact with commands along the way. Then, at least one candidate base station within the radius of each subway station in the specified area is determined, and then based on the dedicated base station information and site information of the specified area, all travel trajectories and the land use type composition information of the specified area, the feature data of each candidate base station of each subway station in the specified area is determined. Finally, based on the feature data of each candidate base station of each subway station in the specified area, all effective subway base stations are identified from all candidate base stations, and the subway station served by each effective subway base station is determined. Compared with the prior art, the present invention provides feature data extracted from each candidate base station, which reflects the type attributes of the candidate base station, the change in the number of people served in the specified time period and the surrounding land use type composition, so that the effective subway base station that provides services for the subway station can be identified according to the feature data of all candidate base stations, which can effectively reduce the error of subsequent subway travel identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0069] Figure 1 One of the flowchart diagrams of a base station identification method provided by an embodiment of the present invention.

[0070] Figure 2 A frequency distribution diagram generated when determining a deployment radius is provided in an embodiment of the present invention.

[0071] Figure 3 An example diagram of a labeling provided in an embodiment of the present invention.

[0072] Figure 4 The second flowchart of a base station identification method provided by an embodiment of the present invention.

[0073] Figure 5 A scatter plot for verifying recognition effect provided by an embodiment of the present invention.

[0074] Figure 6 A line chart for verifying recognition effect provided by an embodiment of the present invention.

[0075] Figure 7 A schematic structural diagram of a base station identification device provided in an embodiment of the present invention.

[0076] Figure 8 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0077] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0078] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0080] In addition, the terms “first”, “second”, etc., if used, are merely used to distinguish between the descriptions and should not be understood as indicating or implying relative importance.

[0081] It should be noted that, in the absence of conflict, the features in the embodiments of the present invention may be combined with each other.

[0082] Please refer to Figure 1 , Figure 1 A flow chart of a base station identification method provided in an embodiment of the present invention is provided. The execution subject of the method may be an electronic device, which may be, but is not limited to, a computer device, a server, etc. The base station identification method includes the following steps S101 to S104:

[0083] S101. Determine at least one travel trajectory of each traveling user according to a mobile phone signaling data set in a specified area within a specified time period.

[0084] Optionally, the designated area may be an entire city, such as Beijing, Chengdu, etc., and the designated time period may be one day, multiple days, one week, one month, etc. In the corresponding example, the mobile phone signaling data set may include all signaling data generated by Chengdu in January 24. This example is only an example and is not limited to this in the embodiments of the present invention. A signaling data includes a user code (representing a user terminal used by the user), a timestamp, and a base station code (representing a public mobile communication base station, referred to as a base station, and each base station can query the specific longitude and latitude coordinates), so the signaling data can reflect the object (i.e., the base station) with which the user terminal performs command interaction and the interaction time.

[0085] Optionally, the acquired mobile phone signaling data set may be cleaned first, and then at least one travel trajectory of each traveling user may be extracted. The specific extraction method will not be described in detail here.

[0086] In this embodiment, a travel trajectory may include each base station for which the user uses a mobile phone to interact with commands along the way and the interaction time corresponding to each base station.

[0087] S102: Determine at least one candidate base station within a deployment radius around each subway station in a designated area.

[0088] S103: Determine feature data of each candidate base station for each subway station in the designated area based on the dedicated base station information and site information, all travel trajectories, and land use type composition information of the designated area.

[0089] In this embodiment, the characteristic data reflects the type attributes of the candidate base station, the change in the number of service users in a specified period of time, and the composition of the surrounding land use types.

[0090] S104: Based on the feature data of each candidate base station of each subway station in the designated area, identify all valid subway base stations from all candidate base stations, and determine the subway station served by each valid subway base station.

[0091] In this embodiment, the effective subway base station is a base station that has performed command interaction with the user terminal of the user traveling in the subway station. It can be understood that among all the effective subway base stations, some are dedicated subway base stations of the subway station, and the other part are base stations reused around the subway station.

[0092] The base station identification method provided by the embodiment of the present invention first determines at least one travel trajectory of each traveling user based on a mobile phone signaling data set in a specified area within a specified time period, and finds the candidate base stations corresponding to each subway station in the specified area, and then determines the feature data of each candidate base station based on the dedicated base station information, site information, land use type composition information and all travel trajectories of the specified area. Compared with the prior art, the present invention extracts the feature data of each candidate base station, and the feature data reflects the type attributes of the candidate base station, the change in the number of service passengers in the specified time period and the surrounding land use type composition. Finally, the effective subway base station providing services for the subway station can be identified based on the feature data of all candidate base stations, which can effectively reduce the error of subsequent subway travel identification.

[0093] In an optional implementation, the deployment radius needs to be obtained through analysis. Correspondingly, the sub-steps of the above step S102 may include S1021 to S1024:

[0094] S1021. Obtain dedicated base station information and site information.

[0095] In this embodiment, the dedicated base station information may include the location of each dedicated subway base station in the specified area and the subway stations it serves. The station information may include the location and station type of each subway station in the specified area. The station type may include ground stations and underground stations.

[0096] S10212. Based on the location of each dedicated subway base station and the location of each subway station it serves, calculate the distribution distance between each dedicated subway base station and the subway station it serves.

[0097] Optionally, the location of the dedicated subway base station and the location of each subway station it serves can be longitude and latitude coordinates. Calculate the distribution distance between subway station u and its corresponding dedicated subway base station t The formula can be:

[0098]

[0099] In formula (1), represents the latitude and longitude coordinates of subway station u, represents the latitude and longitude coordinates of the dedicated subway base station t, and M represents the constant coefficient.

[0100] S10213. Based on all distribution distances, determine the layout radius between the dedicated subway base station and the subway station it serves.

[0101] Optionally, based on all the distribution distances, a frequency distribution diagram of the dedicated subway base station at different distribution distances can be generated; then the breakpoints of the frequency distribution diagram can be identified, and the distribution distance corresponding to the breakpoint can be used as the layout radius.

[0102] In the optional example, if the generated frequency distribution graph is Figure 2 As shown, Figure 2 The horizontal axis representing distance can be divided into 50m segments, and the vertical axis represents frequency ( Figure 2 Left vertical axis) and cumulative frequency ( Figure 2 Right vertical axis) Figure 2 It can be seen that from 2000m, the two curves begin to flatten, and the identified breakpoint is at 2000m, so it can be determined that the deployment radius is 2km. This example is only an example, and the embodiment of the present invention is not limited to this.

[0103] S10214. For each subway station in the designated area, all base stations within a deployment radius around the subway station are used as candidate base stations of the subway station, and each candidate base station of each subway station in the designated area is obtained.

[0104] It is understandable that underground stations usually have dedicated subway base stations, but ground stations usually reuse surrounding base stations (such as base stations in nearby communities or nearby shopping malls), and few ground stations have dedicated subway base stations. Therefore, in order to more accurately identify subway travel, it is necessary to identify effective subway base stations that interact with the terminals of traveling users in subway stations. For underground stations, there is no doubt that the effective subway base stations are dedicated subway base stations, but for ground stations, it is necessary to consider how to identify effective subway base stations.

[0105] In this embodiment, for each subway base station in the specified area, it is necessary to first find each base station within the deployment radius around the subway base station as a candidate base station. There may be more than one candidate base station for a subway station. For underground stations, there will be one or more dedicated subway base stations among all the corresponding candidate base stations, but for ground stations, each corresponding candidate base station may be an effective subway base station and needs to be further identified.

[0106] In an optional implementation, the feature data may include attribute labels and comprehensive feature vectors. Correspondingly, the sub-steps of the above step S103 may include S1031~S1032:

[0107] S1031. Determine the attribute label of each candidate base station corresponding to each subway station in the specified area based on the dedicated base station information and the site information.

[0108] In this embodiment, the attribute tag of a candidate base station may be:

[0109]

[0110] The subway station corresponding to the candidate base station is an underground station, and there is another dedicated subway base station;

[0111] The subway station corresponding to the candidate base station is an underground station, and the candidate base station is a dedicated subway base station of the corresponding subway station;

[0112] condition The subway station corresponding to the candidate base station is a ground station, and the candidate base station is a dedicated subway base station of the corresponding subway station;

[0113] condition That is: there is no dedicated subway base station at the subway station corresponding to the candidate base station.

[0114] For an optional example, see Figure 3 , Figure 3 The labeling of candidate base stations of the five subway stations involved in a travel trajectory is shown in FIG. This example is only for illustration and is not intended to be limiting.

[0115] S1032. Based on all travel trajectories and land use type composition information, determine the comprehensive feature vector of each candidate base station corresponding to each subway station in the specified area.

[0116] In this embodiment, the comprehensive feature vector can represent the change in the number of people served by the candidate base station in a specified period of time and the composition of the surrounding land use types.

[0117] Optionally, the land use type composition information may include the locations of several points of interest in the specified area and the land use types they belong to. Specifically, the land use types may include six categories: scenic spots, residential, business companies, science, education, culture, health and service, and life services. Correspondingly, for each candidate base station corresponding to each subway station, the process of determining the candidate base station may include the following steps S10321~S10324:

[0118] S10321. Divide the specified time period into multiple particle time periods according to a preset period.

[0119] Optionally, the preset period can be flexibly set, such as 1 hour, 2 hours or even 1 day. For example, if the specified period is 1 working day and the preset period is 1 hour, then the specified period can be divided into 24 particle periods. This example is only an example and is not limited here.

[0120] S10322. Count the number of people served by the candidate base station in each granular time period from all travel trajectories to obtain a first feature vector of the candidate base station.

[0121] In this embodiment, for each candidate base station corresponding to each subway station, it is necessary to count the number of people served by the candidate base station in each granular time period from all travel trajectories to obtain the first feature vector of the candidate base station. For a candidate base station, it may appear once or multiple times in a travel trajectory of a travel user, but only one time is recorded when counting the number of people served.

[0122] S10323. Within a deployment radius centered on the candidate base station, count the number of points of interest of each land use type to obtain a second feature vector of the candidate base station.

[0123] S10324. Standardize the first feature vector and the second feature vector of the candidate base station and then concatenate them to obtain a comprehensive feature vector of the candidate base station.

[0124] In this embodiment, it is assumed that the first eigenvector and the second eigenvector are divided into M and N eigenvalues. In order to unify the dimensions, the first eigenvector , the second eigenvector The formulas for standardization are formula (3) and formula (4):

[0125]

[0126]

[0127] Represents the candidate base station in Number of people served in each granular time period, Represents the candidate base station within the nearby deployment radius. The number of points of interest of this land use type. After normalization, the first eigenvector , the second eigenvector They become: , .

[0128] In the optional example, if the specified time period is divided into 48 granular time periods and 6 land use types, then the comprehensive feature vector of the candidate base station is 54-dimensional, that is, it includes 54 eigenvalues. The first 48 eigenvalues ​​reflect the change in the number of people served by the candidate base station in the specified time period, and the last 6 eigenvalues ​​reflect the composition of the land use types within the deployment radius near the candidate base station. This example is only an example and is not limited here.

[0129] In an optional implementation, based on the above process, it can be clearly known that all candidate base stations with attribute labels of 0 are not valid metro base stations, and all candidate base stations with attribute labels of 1 and 2 are valid metro base stations, but it is uncertain whether the candidate base station with attribute label of 3 is a valid metro base station. Therefore, a classification recognition model can be trained using all candidate base stations with attribute labels of 0, 1, and 2, and then the classification recognition model can be used to identify all candidate base stations with attribute label of 3.

[0130] Correspondingly, the sub-steps of the above step S104 may include S1041 to S1044:

[0131] S1041. Generate a classification recognition model based on the comprehensive feature vectors and attribute labels of all candidate base stations whose attribute labels are 0, 1, and 2.

[0132] In this embodiment, the classification recognition model may adopt a decision tree or a random forest. Taking the random forest as an example, the process of obtaining the classification recognition model may include the following steps (1) to (4):

[0133] (1) Construct a validation set based on all candidate base stations with attribute label 2.

[0134] The validation set may include several validation samples (all of which are positive samples), and each validation sample includes a comprehensive feature vector and attribute label of a candidate base station.

[0135] (2) Construct a data set using all candidate base stations with attribute labels of 0 and 1.

[0136] The data set includes several training samples, each of which includes a comprehensive feature vector and attribute label of a candidate base station. The training sample with attribute label 0 is a negative sample, and the training sample with attribute label 1 is a positive sample.

[0137] (3) Divide the dataset into training set and test set according to the preset ratio.

[0138] Among them, the sample ratio in the training set and the test set can be 8:2.

[0139] (4) Use the training set, validation set, and test set to train the pre-built random forest model to obtain a classification and recognition model.

[0140] The trained classification recognition model can be used to identify whether a candidate base station is a specific subway base station that interacts with terminals of traveling users at nearby subway stations.

[0141] It should be noted that since the ratio of positive and negative samples in the data set is likely to be unbalanced, there may be fewer positive samples. In order to allow the random forest model to have a better recognition effect on positive samples and focus on identifying positive samples, the present invention can use a weighted method to evaluate the model when using a set (test set, training set or validation set) for testing during the training of the random forest model. The formula for weighted evaluation is as follows:

[0142]

[0143]

[0144] in, represents the number of positive samples in the set, Represents the number of negative samples in the set; Represents the proportion of negative samples in the set, which serves as the evaluation weight of the positive sample results; Represents the proportion of positive samples in the set, which serves as the evaluation weight of negative sample results.

[0145] Represents the depth of the random forest model of The accuracy of the test using the set of trees; The number of samples with a classification label of 1 obtained in the random forest model for the set input; The number of samples with a classification label of 0 obtained in the random forest model for the set input.

[0146] In the optional example, the inventor verified that the random forest ensemble algorithm was used to adjust the parameters to complete the sample training, and finally 35 trees were constructed, each with a depth of 5, which was optimal. The accuracy of the weighted evaluation using the test set reached 70%, and the accuracy of the validation set also reached more than 70%. This example is only an example and is not limited here.

[0147] S1042: Filter out all specific subway base stations from all candidate base stations with an attribute label of 3 using a classification recognition model.

[0148] In this embodiment, the specific subway base station represents a base station that interacts with terminals of traveling users at a nearby subway station, that is, the specific subway base station is a non-dedicated subway base station, but a base station that provides services for the subway station.

[0149] Correspondingly, the process of selecting all specific subway base stations from all candidate base stations with attribute label 3 includes the following steps (1) to (3):

[0150] (1) All candidate base stations with attribute label 3 are taken as base stations to be identified;

[0151] (2) Input the comprehensive feature vector of each base station to be identified into the classification recognition model to obtain the classification label of each base station to be identified; the classification label is 0 or 1;

[0152] (3) Filter out all specific subway base stations with a classification label of 1 from all base stations to be identified.

[0153] It can be understood that for a base station to be identified, if the classification label output by the classification recognition model is 0, then it can be considered that the base station to be identified does not provide services to the traveling users in the subway station; conversely, if the output classification label is 1, then it can be considered that the base station to be identified is a specific subway base station that interacts with the terminals of the traveling users in the nearby subway stations.

[0154] S1043. All candidate base stations with attribute labels 1 and 2 and all specific subway base stations are regarded as effective subway base stations.

[0155] S1044. Find out the subway station served by each candidate base station with attribute labels 1 and 2 from the dedicated base station information, and set the service object of each specific subway base station to the nearest subway station around it, so as to obtain the subway station served by each effective subway base station.

[0156] In this embodiment, for effective subway base stations that are professional subway base stations (attribute label is 1 or 2), the subway stations they serve can be found out from the dedicated base station information, while for effective subway base stations that are not professional subway base stations, their service objects can be set to the nearest subway station nearby.

[0157] In this way, all effective subway base stations in the specified area are found, and the service objects of each effective subway base station are obtained, which can significantly improve the accuracy of subsequent statistics related to subway travel.

[0158] In an optional implementation, by using all the effective subway base stations and their service objects in the determined designated area, the subway travel trajectories can be accurately screened out from all the travel trajectories, and then relevant data statistics can be performed.

[0159] For corresponding Figure 4 After the above step S104, the following steps may be further included: S105 to S107:

[0160] S105. Utilize all effective subway base stations and subway stations they serve to select each subway travel trajectory that meets preset characteristic conditions from all travel trajectories.

[0161] Optionally, the preset characteristic condition may be: the travel trajectory includes at least two different effective subway base stations, and the subway travel distance is above a set value (e.g., 2km, 2.5km). The subway travel distance is the distance between the subway station served by the first effective subway base station in the travel trajectory and the subway station served by the last effective subway base station.

[0162] S106. Determine the entry subway station and the exit subway station corresponding to each subway travel trajectory.

[0163] In this embodiment, for a subway travel trajectory, the service object of the first subway base station in the subway travel trajectory can be used as the entry subway station, and the service object of the last subway base station can be used as the exit subway station.

[0164] S107. Determine the service range information of each subway station in the specified area according to the entry subway station and exit subway station corresponding to each subway travel trajectory, and / or determine the subway travel passenger flow in the specified area during the specified time period, and / or count the passenger flow of each subway station in the specified area during the specified time period. Optionally, the service range information of a subway station can be presented in terms of distance or in terms of a distribution map.

[0165] Optionally, for the subway travel passenger flow in a specified area during a specified time period: it can be the number of all filtered subway travel trajectories.

[0166] Optionally, for the passenger flow of a subway station A: the number of subway travel trajectories whose entry subway station is subway station A can be counted to obtain the entry volume of subway station A in a specified period of time; then the number of subway travel trajectories whose exit subway station is subway station A can be counted to obtain the exit volume of subway station A in a specified period of time; the passenger flow of subway station A can be the sum of the entry volume and the exit volume.

[0167] Among them, the mobile phone signaling data set and occupied base station information can be provided by a single operator or by all operators. It should be noted that if the mobile phone signaling data set is supported by data provided by a single operator, then the market share of the operator in the specified area can be combined with the passenger flow identification results corresponding to the operator to infer the full passenger flow scale.

[0168] Next, we introduce two ways to determine the service range information of subway stations:

[0169] In the first optional implementation, the service range information may include an inbound service radius and an outbound service radius. Correspondingly, the process of "determining the service range information of each subway station in the specified area according to the inbound subway station and the outbound subway station corresponding to each subway travel trajectory" in step S107 may include the following sub-steps S107a to S107e:

[0170] S107a, for each subway travel trajectory, calculating the entry distance between the first base station of the subway travel trajectory and the entry subway station, and the exit distance between the exit subway station of the subway travel trajectory and the last base station, to obtain the entry distance and exit distance corresponding to each subway travel trajectory;

[0171] S107b, for each subway station, respectively find a first travel trajectory and a second travel trajectory with the subway station as the entry subway station and the exit subway station from all subway travel trajectories;

[0172] S107c, taking the average of the entry distances corresponding to all first travel trajectories of the subway station as the entry service radius of the subway station;

[0173] S107d, taking the mean of the exit distances corresponding to all the second travel trajectories of the subway station as the exit service radius of the subway station;

[0174] S107e. Traverse each subway station to obtain the inbound service radius and the outbound service radius of each subway station.

[0175] Optionally, in order to eliminate the influence of outliers, in the above step S107c or S107c, the 80% quantile corresponding to the entry distance corresponding to all the first travel trajectories of the subway station can be first determined, and then all the entry distances below the 80% quantile are averaged to obtain the mean of the entry distance; at the same time, in the above step S107c, the 80% quantile corresponding to the exit distance corresponding to all the second travel trajectories of the subway station can be selected, and then all the exit distances below the 80% quantile are averaged to obtain the mean of the exit distance. Among them, determining the 80% quantile of a set of data is a prior art and will not be described in detail here.

[0176] In a second optional implementation, the service range information may include an inbound heat map and an outbound heat map. The process of "determining the service range information of each subway station in the specified area according to the inbound subway station and the outbound subway station corresponding to each subway travel trajectory" in step S107 may include the following sub-steps S107A to S107E:

[0177] S107A, for each subway travel trajectory, taking the positions of the first base station and the last base station of the subway travel trajectory as the travel starting point and the travel end point;

[0178] S107B, for each subway station, respectively find a first travel trajectory and a second travel trajectory with the subway station as an entry subway station and an exit subway station from all subway travel trajectories;

[0179] S107C, taking the location of the subway station in the map as the center, and generating an entry heat map of the subway station according to the location of the first base station of each first travel trajectory corresponding to the subway station in the map;

[0180] S107D, taking the location of the subway station in the map as the center, and generating an exit heat map of the subway station according to the location of the last base station of each second travel trajectory corresponding to the subway station in the map;

[0181] S107E, traverse each subway station to obtain the entry heat map and exit heat map of each subway station.

[0182] It can be understood that heat maps use color to represent data density. In the exit heat map, the darker the color of a point, the more people enter the subway station from that point; the lighter the color of a point, the fewer people enter the subway station from that point. In the exit heat map, the darker the color of a point, the more people come out of the subway station to that point; the lighter the color of a point, the fewer people come out of the subway station to that point. Therefore, the combination of the entry heat map and the exit heat map can clearly show the distribution of the service population range of a subway station.

[0183] In order to verify the implementation effect of the present invention, the inventor collected a data set of mobile phone signaling for one consecutive working day and one rest day in March 2023 in Tianjin, and used the method of the present invention to identify in Tianjin during these two days: all subway travel trajectories and their entry and exit subway stations, and the passenger flow of each subway station.

[0184] In order to verify the accuracy of the recognition, the inventor also obtained the passenger flow of each subway station in Tianjin in the past two days based on the subway system's card swiping data (using IC cards, mobile phone cards, and mobile phone QR codes to enter and exit the station), and then made the following Figure 5 The scatter plot shown. Figure 5 In the figure, each point represents a subway station, the abscissa of a point represents the passenger flow counted by the subway system, and the ordinate represents the passenger flow identified by the method of the present invention.

[0185] Figure 5 In an ideal situation, the fitting execution of all points should be y=x. Figure 5 Based on the fitted straight line, the correlation coefficient r between the passenger flow identified by the signaling method of the present invention and the passenger flow statistically analyzed by the subway system is about 0.76, and the corresponding R 2It is about 0.57, and the regression line is basically on the diagonal line, so the recognition effect is significant.

[0186] Regarding the passenger flow recognition effect of a single subway station, the inventor identified the passenger flow based on the signaling of Tianjin University Town Station in the past two days ( Figure 6 Signaling recognition in the subway system) and passenger flow statistics in the subway system ( Figure 6 The total number of people swiping cards in the Figure 6 ), it can be seen that the two are basically consistent, indicating that the recognition effect of the method provided by this aspect is better.

[0187] It should be noted that the execution order of each step in the above method embodiment is not limited to that shown in the drawings, and the execution order of each step shall be based on the actual application situation.

[0188] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0189] (1) The present invention uses the mobile phone signaling data set of the specified area in the specified time period to determine at least one travel trajectory of each travel user. Then find the candidate base stations corresponding to each subway station, and then annotate the attribute labels of each candidate base station based on the dedicated base station information and the site type of each subway station. Then, the service number of each candidate base station in all travel trajectories and the land use type composition information of the specified area are integrated to construct a comprehensive feature vector for each candidate base station. Then, the comprehensive feature vectors and attribute labels of all candidate base stations with attribute labels of 0, 1 and 2 are used to train a classification recognition model, so that a specific subway base station that is not a dedicated subway base station but serves a subway station can be screened out from the candidate base stations with attribute label 3, thereby accurately identifying all effective subway base stations in the specified area (including candidate base stations with attribute labels 1 and 2 and the specific subway base station finally screened out) and their service objects.

[0190] (2) Compared with the prior art, subway travel can only be based on dedicated subway base stations, while subway stations that do not use dedicated subway base stations are ignored, resulting in inaccurate identification of subsequent subway travel trajectories. The present invention uses all the effective subway base stations in the identified designated area to accurately identify the subway travel trajectories and their entry and exit subway stations in all travel trajectories without omissions, and can then accurately count the subway travel passenger flow in the designated area and the passenger flow of each subway station.

[0191] (3) The present invention can also use each identified subway travel trajectory and its entry and exit subway stations to analyze the service scope of each subway station, thereby facilitating subsequent subway site selection and construction.

[0192] In order to execute the corresponding steps in the above method embodiment and various possible implementation modes, an implementation mode of a base station identification device is given below.

[0193] See also Figure 7 , Figure 7 The schematic diagram of the structure of the base station identification device provided by the embodiment of the present invention is shown. The base station identification device 200 comprises: a track collection module 210, a base station screening module 220, a feature extraction module 230 and a base station identification module 240.

[0194] The trajectory collection module 210 is used to determine at least one travel trajectory of each travel user according to the mobile phone signaling data set in the specified area within the specified time period; the travel trajectory includes multiple base stations where the travel user uses the mobile phone to interact with commands along the way;

[0195] A base station screening module 220 is used to determine at least one candidate base station within a deployment radius around each subway station in a specified area;

[0196] The feature extraction module 230 is used to determine the feature data of each candidate base station of each subway station in the specified area based on the dedicated base station information and station information of the specified area, all travel trajectories and the land use type composition information of the specified area; the feature data reflects the type attributes of the candidate base station, the change of the number of service passengers in the specified time period and the surrounding land use type composition;

[0197] The base station identification module 240 is used to identify all effective subway base stations from all candidate base stations based on the characteristic data of each candidate base station of each subway station in the designated area, and determine the subway station served by each effective subway base station.

[0198] Optionally, the base station identification device 200 may also include a travel statistics module 250, which can be used to: utilize all effective subway base stations and the subway stations they serve to screen out each subway travel trajectory that meets preset characteristic conditions from all travel trajectories; determine the entry subway station and exit subway station corresponding to each subway travel trajectory; determine the service range information of each subway station in the specified area based on the entry subway station and exit subway station corresponding to each subway travel trajectory, and / or determine the subway travel passenger flow in the specified area within a specified time period, and / or count the passenger flow of each subway station in the specified area within a specified time period.

[0199] Those skilled in the art can clearly understand that the various modules of the base station identification device 200 described above are used to implement the corresponding steps of the base station identification method described above. For the convenience and simplicity of description, the specific working process of the base station identification device 200 described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0200] See also Figure 8 , Figure 8 The electronic device 300 includes a processor 310 , a memory 320 , and a bus 330 , wherein the processor 310 is connected to the memory 320 via the bus 330 .

[0201] The memory 320 may be used to store software programs, for example, software programs corresponding to the base station identification device 200 provided in the embodiment of the present invention. The processor 310 executes various functional applications and data processing to implement the base station identification method provided in the embodiment of the present invention by running the software programs stored in the memory 320.

[0202] Among them, the memory 320 can be but is not limited to: RAM (Random Access Memory), ROM (Read Only Memory), FLASH (Flash Memory), PROM (Programmable Read-Only Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electric Erasable Programmable Read-Only Memory), etc.

[0203] The processor 310 may be an integrated circuit chip with signal processing capability. The processor 310 may be a general-purpose processor, including: CPU (Central Processing Unit), NP (Network Processor), etc.; it may also be: DSP (Digital Signal Processing), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0204] Understandably, Figure 8 The structure shown is for illustration only. The electronic device 300 may also include Figure 8 More or fewer components as shown, or with Figure 8 Different configurations are shown. Figure 8 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0205] The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the base station identification method disclosed in the above embodiment is implemented. The computer-readable storage medium can be, but is not limited to, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a PROM, an EPROM, an EEPROM, a FLASH disk, or an optical disk.

[0206] In summary, the embodiments of the present invention provide a base station identification method, device, electronic device and computer-readable storage medium. First, according to the mobile phone signaling data set in the specified area within the specified time period, at least one travel trajectory of each travel user is determined, and the travel trajectory includes multiple base stations where the travel user uses the mobile phone to interact with commands along the way. Then, at least one candidate base station within the radius of each subway station in the specified area is determined, and then based on the dedicated base station information and site information of the specified area, all travel trajectories and the land use type composition information of the specified area, the feature data of each candidate base station of each subway station in the specified area is determined. Finally, based on the feature data of each candidate base station of each subway station in the specified area, all effective subway base stations are identified from all candidate base stations, and the subway station served by each effective subway base station is determined. Compared with the prior art, the present invention provides feature data extracted from each candidate base station, which reflects the type attributes of the candidate base station, the change in the number of people served in the specified time period, and the surrounding land use type composition, so that the effective subway base station that provides services for the subway station can be identified according to the feature data of all candidate base stations, which can effectively reduce the error of subsequent subway travel identification.

[0207] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A base station identification method, characterized in that: include: Determine at least one travel trajectory of each travel user based on a mobile phone signaling data set in a specified area within a specified time period; The travel trajectory includes multiple base stations where the travel user uses a mobile phone to interact with commands along the way; Acquire the dedicated base station information and site information of the designated area; the dedicated base station information includes the location of each dedicated subway base station in the designated area and the subway station it serves; the site information includes the location of each subway station in the designated area; Based on the location of each of the dedicated subway base stations and the location of each subway station it serves, calculating the distribution distance between each of the dedicated subway base stations and the subway stations it serves; Based on all the distribution distances, determine the layout radius between the dedicated subway base station and the subway station it serves; For each subway station in the designated area, all base stations within a deployment radius around the subway station are used as candidate base stations of the subway station, to obtain each candidate base station of each subway station in the designated area; Based on the dedicated base station information and site information of the designated area, all the travel trajectories and the land use type composition information of the designated area, determining the characteristic data of each candidate base station of each subway station in the designated area; the characteristic data reflects the type attribute of the candidate base station, the change of the number of service passengers in the designated time period and the surrounding land use type composition; Based on the characteristic data of each candidate base station of each subway station in the designated area, all effective subway base stations are identified from all candidate base stations, and the subway station served by each effective subway base station is determined.

2. The method according to claim 1, characterized in that The station information also includes the station type of each subway station in the designated area; the station type includes ground stations and underground stations; the feature data includes attribute labels and comprehensive feature vectors; The step of determining the characteristic data of each candidate base station of each subway station in the designated area based on the dedicated base station information and site information of the designated area, all the travel trajectories and the land use type composition information of the designated area comprises: Based on the dedicated base station information and the site information, determine the attribute label of each candidate base station corresponding to each subway station in the designated area; the attribute label is one of 0, 1, 2, and 3; wherein, the attribute label of 0 represents that the subway station corresponding to the candidate base station is an underground station and there is another dedicated subway base station; the attribute label of 1 represents that the subway station corresponding to the candidate base station is an underground station, and the candidate base station is a dedicated subway base station of the corresponding subway station; the attribute label of 2 represents that the subway station corresponding to the candidate base station is a ground station, and the candidate base station is a dedicated subway base station of the corresponding subway station; the attribute label of 3 represents that there is no dedicated subway base station at the subway station corresponding to the candidate base station; Based on all the travel trajectories and the land use type composition information, a comprehensive feature vector of each candidate base station corresponding to each subway station in the designated area is determined, and the comprehensive feature vector reflects the change in the number of people served by the candidate base station during the designated period and the surrounding land use type composition.

3. The method according to claim 2, characterized in that The land use type composition information includes the locations of several points of interest in the specified area and the land use types to which they belong; The step of determining the comprehensive feature vector of each candidate base station corresponding to each subway station in the designated area based on all the travel trajectories and the land use type composition information comprises: Dividing the specified time period into a plurality of particle time periods according to a preset period; For each candidate base station corresponding to each subway station, the number of people served by the candidate base station in each granular time period is counted from all travel trajectories to obtain a first feature vector of the candidate base station; Within a deployment radius centered on the candidate base station, counting the number of points of interest of each land use type to obtain a second feature vector of the candidate base station; The first feature vector and the second feature vector of the candidate base station are normalized and then concatenated to obtain a comprehensive feature vector of the candidate base station.

4. The method according to claim 2, characterized in that: The step of identifying all effective subway base stations from all candidate base stations based on the characteristic data of each candidate base station of each subway station in the designated area, and determining the subway station served by each effective subway base station comprises: Generate a classification recognition model based on the comprehensive feature vectors and attribute labels of all candidate base stations whose attribute labels are 0, 1 and 2; Using the classification recognition model, all specific subway base stations are screened out from all candidate base stations with the attribute label 3; the specific subway base stations represent base stations that interact with terminals of users traveling at nearby subway stations; All candidate base stations with attribute labels 1 and 2 and all specific subway base stations are regarded as effective subway base stations; The subway stations served by each candidate base station with the attribute label of 1 are found out from the dedicated base station information, and the service object of each specific subway base station is set to the nearest subway station around it, so as to obtain the subway stations served by each effective subway base station.

5. The method according to claim 4, characterized in that The step of generating a classification recognition model based on the comprehensive feature vectors and attribute labels of all candidate base stations whose attribute labels are 0, 1 and 2 comprises: Constructing a verification set according to all candidate base stations with the attribute label 2, wherein the verification set includes a plurality of verification samples, each of which includes a comprehensive feature vector and an attribute label of a candidate base station; Using all candidate base stations whose attribute labels are 0 and 1, a data set is constructed; the data set includes a plurality of training samples, each training sample includes a comprehensive feature vector and an attribute label of a candidate base station; Dividing the data set into a training set and a test set according to a preset ratio; The pre-constructed random forest model is trained using the training set, the validation set, and the test set to obtain the classification recognition model.

6. The method according to claim 4, characterized in that The step of using the classification recognition model to filter out all specific subway base stations from all candidate base stations with the attribute label 3 includes: All candidate base stations with the attribute label of 3 are taken as base stations to be identified; Inputting the comprehensive feature vector of each base station to be identified into the classification recognition model to obtain a classification label of each base station to be identified; the classification label is 0 or 1; All the specific subway base stations with the classification label 1 are screened out from all the base stations to be identified.

7. The method according to claim 1, characterized in that The method further comprises: Using all effective subway base stations and subway stations they serve, each subway travel trajectory that meets preset characteristic conditions is selected from all the travel trajectories; Determine the entry subway station and the exit subway station corresponding to each of the subway travel trajectories; According to the entry subway station and exit subway station corresponding to each subway travel trajectory, the service range information of each subway station in the designated area is determined, and / or the subway travel passenger flow in the designated area during the designated time period is determined, and / or the passenger flow of each subway station in the designated area during the designated time period is counted.

8. The method according to claim 7, characterized in that The service range information includes an entry service radius and an exit service radius; the step of determining the service range information of each subway station in the designated area according to the entry subway station and the exit subway station corresponding to each subway travel trajectory includes: For each of the subway travel trajectories, the inbound distance between the first base station of the subway travel trajectory and the inbound subway station and the outbound distance between the outbound subway station of the subway travel trajectory and the last base station are calculated to obtain the inbound distance and outbound distance corresponding to each of the subway travel trajectories; For each of the subway stations, respectively find a first travel trajectory and a second travel trajectory with the subway station as an entry subway station and an exit subway station from all subway travel trajectories; The average of the entry distances corresponding to all the first travel trajectories of the subway station is used as the entry service radius of the subway station; The mean of the exit distances corresponding to all the second travel trajectories of the subway station is used as the exit service radius of the subway station; Each subway station is traversed to obtain the inbound service radius and the outbound service radius of each subway station.

9. The method according to claim 7, characterized in that: The service range information includes an inbound heat map and an outbound heat map; The step of determining the service range information of each subway station in the designated area according to the entry subway station and the exit subway station corresponding to each subway travel trajectory comprises: For each subway travel trajectory, the positions of the first base station and the last base station of the subway travel trajectory are used as the travel starting point and the travel end point; For each of the subway stations, respectively find a first travel trajectory and a second travel trajectory with the subway station as an entry subway station and an exit subway station from all subway travel trajectories; Taking the position of the subway station in the map as the center, and according to the position of the first base station of each first travel trajectory corresponding to the subway station in the map, generating an entry heat map of the subway station; Taking the position of the subway station in the map as the center, and according to the position of the last base station of each second travel trajectory corresponding to the subway station in the map, generating an exit heat map of the subway station; Traverse each of the subway stations to obtain an inbound heat map and an outbound heat map of each subway station.

10. A base station identification device, characterized in that: include: A trajectory collection module, used to determine at least one travel trajectory of each travel user based on a mobile phone signaling data set in a specified area within a specified period of time; The travel trajectory includes multiple base stations where the travel user uses a mobile phone to interact with commands along the way; A base station screening module, used to obtain the dedicated base station information and site information of the designated area; the dedicated base station information includes the location of each dedicated subway base station in the designated area and the subway station it serves; the site information includes the location of each subway station in the designated area; The base station screening module is further used to calculate the distribution distance between each of the dedicated subway base stations and the subway stations it serves based on the location of each of the dedicated subway base stations and the location of each subway station it serves; The base station screening module is further used to determine the layout radius between the dedicated subway base station and the subway station it serves based on all distribution distances; The base station screening module is further configured to, for each subway station in the designated area, use all base stations within a deployment radius around the subway station as candidate base stations for the subway station, and obtain each candidate base station for each subway station in the designated area; A feature extraction module is used to determine feature data of each candidate base station of each subway station in the specified area based on the dedicated base station information and station information of the specified area, all the travel trajectories and the land use type composition information of the specified area; the feature data reflects the type attribute of the candidate base station, the change of the number of service passengers in the specified time period and the surrounding land use type composition; The base station identification module is used to identify all effective subway base stations from all candidate base stations based on the feature data of each candidate base station of each subway station in the designated area, and determine the subway station served by each effective subway base station.

11. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a software program, and when the electronic device is running, the processor executes the software program to implement the base station identification method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the base station identification method according to any one of claims 1 to 9 is implemented.

Images