JP2013121073A - Position information analysis device and position information analysis method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system capable of appropriately acquiring a concentration area of a user of a terminal device according to an actual user distribution and capable of reducing processing load.SOLUTION: A position information analysis device 600 comprises: an input part 601 which inputs area position information and point data containing population in the area; an area extraction part 605 which extracts a predetermined number from the point data and determines the position corresponding to the area position information as a center position of a cluster; an area distribution part 606 which, on the basis of distance between the center position of the cluster and the area position information, determines the cluster to which the point data belongs and then distributes the point data to each cluster; a tabulating part 609 which tabulates the population for each cluster on the basis of the point data; and a cluster center determination part 607 which determines the center position of each cluster on the basis of the point data. The area distribution part 606 redistributes the point data to each cluster by referring to the determined center position.

Description

  The present invention relates to a position information analysis apparatus and a position information analysis method for performing user distribution analysis based on position information indicating the number of users corresponding to an area.

  2. Description of the Related Art Conventionally, an apparatus that acquires position information of a terminal device used by a user and statistically processes the position information is known. For example, the method described in Non-Patent Document 1 is a method of capturing a rough flow of a person as a flow line by statistically processing a position information history acquired from a GPS positioning terminal or the like. Specifically, position information is periodically acquired by using satellite positioning, Cell positioning, WiFi positioning, and the like, and these are converted into an action history including a POI (Point of Interest) and a flow line. As this POI, a place that is statically determined as a place where people gather such as a station or a school, a static section or a stay section is obtained from the position information history, and a place that is dynamically determined from these is set.

Shunichiro Tsuji et al. “A method for creating a flow line map using P-type Fourier descriptor”, IEICE Technical Report, Vol.108 No.397 IEICE Technical Report, January 15, 2009, P.7-P .12

  However, in the above-described conventional statistical processing method, since the area where the users gather is statically determined, the user's concentrated area cannot be obtained appropriately according to the actual user distribution. In addition, when a concentrated area is dynamically obtained from a stationary section or a stay section, it is necessary to calculate an action history based on position information of individual users, so that the processing load tends to be high.

  Therefore, the present invention has been made in view of such problems, and a position information analysis apparatus and position information that can appropriately determine the concentration area of the user of the terminal device according to the actual user distribution and reduce the processing load. The purpose is to provide an analysis method.

  In order to solve the above problem, the position information analysis apparatus of the present invention is area data including area position information indicating the position of an area and a statistical value related to the number of users in the area, and a plurality of pieces of information about a plurality of areas. An input unit for inputting the area data, and a cluster that extracts a predetermined number of area data from the input plurality of area data and collects the positions corresponding to the area position information included in the predetermined number of area data. By determining the distance between the center position of the predetermined number of clusters and the area position information by determining the distance between the area extraction unit that is determined as the center position and the area position information, each of the clusters to which the area corresponding to the plurality of area data belongs is determined. Based on an area distribution unit that distributes area data to each cluster, and a plurality of area data distributed by the area distribution unit, By determining the position where the number of users in each cluster is concentrated based on a plurality of area data distributed by the totaling unit that aggregates the statistical value for each cluster and the area distribution unit, the position is determined for each cluster. A cluster center determination unit that determines the center position, and the area distribution unit redistributes a plurality of area data to each cluster with reference to the center position determined by the cluster center determination unit.

  Alternatively, the position information analysis method of the present invention is a position information analysis method executed in the position information analysis apparatus, wherein the input unit includes area position information indicating the position of the area and a statistical value relating to the number of users in the area. An input step of inputting a plurality of area data for a plurality of areas, and an area extraction unit extracts a predetermined number of area data from the input plurality of area data, An area extraction step for determining a position corresponding to the area position information included in the area data as a center position of a cluster that groups a plurality of areas, and a distance between the center position of a predetermined number of clusters and the area position information. By determining each of the clusters to which the area corresponding to the plurality of area data belongs, An area distribution step for distributing data to each cluster, a totaling unit for totaling statistical values for each cluster based on a plurality of area data distributed by the area distribution unit, and a cluster center determining unit, A cluster center determining step for determining a position where the number of users in each cluster is concentrated based on a plurality of area data distributed by the area distributing unit to determine the position as the center position of each cluster; In the area distribution step, a plurality of area data are redistributed to each cluster with reference to the center position determined by the cluster center determination unit.

  According to such position information analysis apparatus and position information analysis method, a plurality of area data including area position information and statistical values related to the number of users in the area are input, and a predetermined number is selected from the plurality of area data. The center positions of a predetermined number of clusters are determined from the area position information extracted and included in the extracted area data. Further, by determining the distance between the center position of the predetermined number of clusters and the area position information, a plurality of area data is distributed to the predetermined number of clusters, and the statistics included in the area data for each cluster using the distribution result. Values are aggregated. Thereby, the distribution of the number of users in a plurality of areas included in each cluster can be specified for each predetermined number of clusters, and the user concentration area for each predetermined number of clusters can be appropriately obtained according to the actual user distribution. . In addition, the processing load can be reduced by processing using the statistical value for each area. Further, the center position of the cluster can be determined based on the actual concentrated area of the user, and a plurality of areas can be distributed to the cluster in accordance with the center position. As a result, it is possible to appropriately set the cluster range for grouping the areas corresponding to the user's concentrated position.

  It is also preferable that the area distribution unit repeats redistribution of the plurality of area data until the center position determined by the cluster center determination unit coincides with the previously determined center position. By adopting such a configuration, it is possible to set a cluster that further reflects the actual concentration position of the user.

  Further, the cluster center determination unit calculates the centroid position for each cluster by weighting and adding the statistical information included in the plurality of area data to the area position information included in the plurality of area data, It is also preferable to determine the center of gravity position as the center position of each cluster. In this case, the user concentration position in each cluster can be easily calculated.

  Still further, the input unit inputs the area data including density information related to the user density in the area, and the cluster center determination unit refers to the density information included in the plurality of area data, and the user density belonging to each cluster. It is also preferable to extract area position information corresponding to the largest area and determine the position corresponding to the area position information as the center position of each cluster. By adopting such a configuration, it is possible to easily calculate the concentration position of users in each cluster. In addition, a position where there is a high possibility that the user actually exists can be set as the cluster center.

  Furthermore, a cluster density calculator for calculating a user density for each cluster based on a plurality of area data distributed by the area distributor is provided, and the input unit includes density information on the user density in the area. The cluster center determination unit specifies area data corresponding to the area including the barycentric position for each cluster, and then the density information included in the area data indicates the user density for each cluster or higher. In this case, the center of gravity position is determined as the center position of each cluster, and when the density information included in the area data indicates the user density or less for each cluster, the density information included in the plurality of area data is referred to. Area information corresponding to the area with the highest user density belonging to each cluster is extracted, and this area position information is supported. Determining that position as the center position of each cluster, it is also preferred. In this way, the method for determining the center position of the cluster can be appropriately selected between the calculation of the center of gravity position and the extraction of the area based on the user density, and the concentration position of the user in each cluster. Can be calculated appropriately.

  In addition, the input unit inputs area data corresponding to a plurality of types of radio base stations via a communication network based on position registration signals for a plurality of types of radio base stations communicating at a plurality of radio frequencies. It is also preferable to further include an integration unit that integrates area data corresponding to a plurality of types of radio base stations input by the unit. In this case, even when radio base stations that handle a plurality of types of radio frequencies are mixed, the cluster corresponding to the user's concentrated location can be easily and appropriately used by using the location registration signal for those radio base stations. Can be determined.

  Furthermore, it is also preferable that the counting unit determines a range obtained by connecting the middle lines of equal distances from the center positions of the plurality of clusters as the geographical range of the plurality of clusters. In this way, even if area data corresponding to multiple types of wireless base stations that may have overlapping coverage areas can be input, cluster areas can be defined without duplication, and each cluster can be divided uniformly. An area can be created.

  ADVANTAGE OF THE INVENTION According to this invention, the concentration area of the user of a terminal device can be calculated | required appropriately according to actual user distribution, and processing load can also be reduced.

It is a system configuration figure of the communications system of this embodiment. It is a functional block block diagram of the positional information analyzer of this embodiment. It is a flowchart which shows operation | movement of the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a conceptual diagram which shows the image of the Voronoi division | segmentation by the positional information analyzer of FIG. It is a figure which shows the output example of the cluster mapped by the visualization solution unit, and those center positions. It is a functional block block diagram of the positional information analyzer of the modification of this invention. It is a flowchart which shows operation | movement of the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the positional information analysis by the positional information analyzer of FIG. It is a functional block block diagram of the positional information analyzer of another modification of this invention. It is a flowchart which shows the operation | movement at the time of cluster center calculation of the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the cluster center calculation by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the cluster center calculation by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the cluster center calculation by the positional information analyzer of FIG. It is a figure which shows the structure of the data of the process target at the time of the cluster center calculation by the positional information analyzer of FIG.

  Hereinafter, preferred embodiments of a position information analyzing apparatus and a position information analyzing method according to the present invention will be described in detail with reference to the drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted.

  FIG. 1 is a system configuration diagram of a communication system 10 according to the present embodiment. As shown in FIG. 1, the communication system 10 includes a mobile device 100, a BTS (radio base station) 200, an RNC (radio control device) 300, an exchange 400, various processing nodes 700, and a management center 500. ing. The management center 500 includes a social sensor unit 501, a petamining unit 502, a mobile demography unit 503, and a visualization solution unit 504.

  The exchange 400 collects location information of the mobile device 100 via a communication network configured by the BTS 200, the RNC 300, and the like. The RNC 300 can measure the position of the mobile device 100 using the delay value in the RRC connection request signal when communication connection is established with the mobile device 100. The exchange 400 can receive the position information of the mobile device 100 measured in this way when the mobile device 100 performs communication connection, that is, when a location registration signal is transmitted. Here, the BTS 200 includes a plurality of types of BTSs that perform wireless communication using a plurality of radio frequencies (for example, a 2 GHz band and an 800 MHz band), and the exchange 400 registers the location with the plurality of types of BTSs 200. Receive location information when The exchange 400 stores the received location information, and outputs the location information collected at a predetermined timing or in response to a request from the management center 500 to the management center 500 via the communication network. Here, in general, the RNC 300 is composed of about 1,000 pieces and is arranged throughout Japan. On the other hand, about 300 exchanges 400 are arranged in Japan.

  The various processing nodes 700 acquire the location information of the mobile device 100 through the RNC 300 and the exchange 400, perform recalculation of the location in some cases, and collect at a predetermined timing or in response to a request from the management center 500 The obtained position information is output to the management center 500.

  As described above, the management center 500 includes the social sensor unit 501, the petamining unit 502, the mobile demography unit 503, and the visualization solution unit 504. Each unit is used for position information of the mobile device 100. Perform statistical processing.

  The social sensor unit 501 is a server device that collects data including location information of the mobile device 100 from each exchange 400 and various processing nodes 700 or offline. The social sensor unit 501 receives data periodically output from the exchange 400 and the various processing nodes 700, or acquires data from the exchange 400 and the various processing nodes 700 according to a predetermined timing in the social sensor unit 501. It is configured to be able to do.

  The petamining unit 502 is a server device that converts data received from the social sensor unit 501 into a predetermined data format. For example, the petamining unit 502 performs a sorting process using a user ID as a key, or performs a sorting process for each area. Further, the petamining unit 502 aggregates the number of users for each located sector (area) of the BTS 200 and generates area data indicating the number of users for each located sector. At this time, the petamining unit 502 generates area data separately for each of a plurality of types of BTSs 200.

  The mobile demography unit 503 is a server device that performs aggregation processing on the data processed in the petamining unit 502, that is, count processing for each item. For example, the mobile demography unit 503 can count the number of users located in a certain area, and can total the distribution of the located areas. Furthermore, the mobile demography unit 503 refers to the area data, generates a cluster that collects the areas in which the mobile demography unit 503 is located, and performs a process of totaling the area data for each cluster. That is, this mobile demography unit 503 is the position information analysis apparatus of this embodiment.

  The visualization solution unit 504 is a server device that processes the data aggregated in the mobile demography unit 503 so as to be visible. For example, the visualization solution unit 504 can map the aggregated data on a map. Data processed by the visualization solution unit 504 is provided to companies, government offices or individuals, and is used for store development, road traffic surveys, disaster countermeasures, environmental countermeasures, and the like. It should be noted that the information statistically processed in this way is processed so that individuals are not specified so as not to infringe privacy.

  The social sensor unit 501, the petamining unit 502, the mobile demography unit 503, and the visualization solution unit 504 are all configured by the server device as described above, and although not shown, the basic configuration of a normal information processing device Needless to say, it includes a CPU, a RAM, a ROM, an input device such as a keyboard and a mouse, a communication device that communicates with the outside, a storage device that stores information, and an output device such as a display and a printer.

  Here, as the “area data” handled by the position information analysis apparatus of the present embodiment, the position information of the mobile device 100 is totaled for each area obtained by dividing a predetermined area (such as the whole of Japan) into a plurality of areas. It is generated as data indicating the number of users. As this area, the serving sector of the BTS 200 may be used as it is, or an area divided into a predetermined shape such as a mesh shape may be adopted, or the serving sectors that overlap each other among a plurality of types of BTS 200 may be adopted. May be adopted as different areas. Further, as the position information to be totaled in the area data, GPS positioning data obtained by GPS positioning may be used. Further, the number of users to be aggregated may be the number of location registration signals or the number of terminals themselves, or may be the number of users estimated from the weight value calculated from the transmission density of location registration signals, or the weight value. It may be the number of users obtained from itself, or may be converted into a population for each area by performing various calculations such as multiplying the number of users by various coefficients.

  For example, the estimation of the number of users from the weight value and the conversion to the population can be calculated as follows.

That is, the estimated generation density of the position information generated by the mobile device 100 with respect to data obtained by combining the position information of the mobile device 100 and the attribute information related to the mobile device 100 (hereinafter simply referred to as “position data”). A feature quantity w _ij as corresponding information (weight value) is _assigned, and the number of in-zones (number of users) is estimated by totaling the set of position data. Here, the “estimated generation density” means an estimated value of the number of signals generated per unit time around the generation time of the position information by the terminal that generated the position information.

即ち、各端末ａ_ｉの観測期間内のセクタＳの滞在時間ｔ_ｉの総和を観測期間の長さＴで除した結果を、端末数ｍとして推計する。ただし、端末ａ_ｉの観測期間内のセクタＳの滞在時間ｔ_ｉの真の値は観測不能であるが、各端末ａ_ｉは位置情報（例えば位置登録信号であり、以下では単に「信号」と言う。）を発信し、それらの信号は観測可能である。 First, the concept of terminal number estimation and the calculation method will be described. During a certain observation period (length T), n pieces of terminal _{a 1, a 2, ...,} a n passes through the sector S of BTS 200, the residence time of the sector S in the observation period of each terminal _{a i} It is assumed that t _i (0 <t _i ≦ T). At this time, the number m of terminals present in the sector S (actually the average value of the number m of terminals present in the sector S within the observation period) is expressed by the following equation (1).

That is, the result of dividing the sum of the stay times t _i of the sectors S within the observation period of each terminal a _i by the length T of the observation period is estimated as the number of terminals m. However, although the true value of the stay time t _i of the sector S within the observation period of the terminal a _i is not observable, each terminal a _i is position information (for example, a position registration signal. The signal is observable.

（ｘ_ｉは、端末ａ_ｉが観測期間内にセクタＳで発信した信号の総数）とすると、端末数の推計とは、観測された信号ｑ_ｉｊ（ｊは１以上ｘ_ｉ以下の整数）からｍの値を推計することに他ならない。 The signals transmitted by the terminal a _i from the sector S during the observation period

If x _i is the total number of signals transmitted by the terminal a _i in the sector S during the observation period, the estimation of the number of terminals is based on the observed signal q _ij (j is an integer from 1 to x _i ). It is none other than estimating the value of m.

Here, the density at which the signal q _ij is transmitted from the terminal a _i (that is, the number of signals per unit time) is p _i . At this time, if the probability that the signal is transmitted is independent of the sector, the expected value E (x _i ) of the total number x _i of signals transmitted from the terminal a _i in the sector S within the observation period is E (x _i) since a = _{t i} × _{p i,} the expected value E _{(t i)} the following expression for the residence time _{t i} of the sector S in the observation period of the terminal _{a i} (2) is satisfied.
E (t _i ) = x _i / p _i (2)
Here, when the transmission time of the signal _{q ij} was _{u ij,} density _{p ij} of the signal _{q ij} is given by the following equation (3).
p _ij = 2 / (u _{i (j + 1)} −u _{i (j−1)} ) (3)
Here, if the signal q _ij is a signal related to position data (first position data) for which the feature value w _ij is to be obtained, the signal q _{i (j−1)} is the same as the first position data. Of the position data including the identification information, the signal q _{i (j + 1)} related to the position data (second position data) immediately before the first position data is the same identification information as the first position data. Corresponds to a signal related to position data (third position data) immediately after the first position data. In the present embodiment, the transmission time _{u i} of the signal of the second position data _{q i (j-1)} transmission time _{u i} of _(j-1) and the signal _{q i} of the third position data _{(j + 1) (} The difference of _{j + 1)} , that is, (u _{i (j + 1)} −u _{i (j−1)} ) in the above equation (3) is set as the feature quantity w _ij for the first position data. Therefore, the above formula (3) is as follows. That is, the feature quantity w _ij can be calculated in association with the reciprocal of the density p _ij .
p _ij = 2 / (u _{i (j + 1)} −u _{i (j−1)} ) = 2 / w _ij (4)
Further, the feature amount w _ij is added to all the position data to be aggregated.

で与えられるため、端末数ｍの推計値Ｅ（ｍ）は以下の式（６）で計算することができる。

At this time, the density p _i is

Therefore, the estimated value E (m) of the number m of terminals can be calculated by the following equation (6).

In this way, the estimated number of terminals for each attribute for each sector is obtained using the feature value w _ij . Furthermore, by multiplying the estimated number of terminals by an expansion coefficient, the estimated number of terminals for each sector S and for each attribute can be calculated, and the sum of those can be converted into a population estimate for each sector S. Such an expansion coefficient can be set for each address such as for each prefecture, for each user's age group, for each sex, and for each time zone related to position information generation.

  For example, conversion to a population estimate using an expansion factor can be calculated as follows.

That is, it is assumed that a data set composed of four data elements each including the following data items “sector, time, age, sex, number of estimated terminals” is obtained as the estimated number of terminals for each sector and attribute. To do.
Element 1: (S001, t, 23, male, a1)
Element 2: (S001, t, 33, male, a2)
Element 3: (S001, t, 31, male, a3)
Element 4: (S001, t, 85, male, a4)

Here, it is assumed that the following information is held as the contract rate for the group corresponding to the age group and the sex at the time t at the area corresponding to the address of the corresponding user.
(20s, male): 40% (= 0.4)
(30s, male): 50% (= 0.5)
(80s, male): 5% (= 0.05)

In this case, assuming that the expansion coefficient is the reciprocal of the contract rate, the following expansion coefficient is given to each element. Then, a population estimate for each data element is obtained by multiplying the estimated number of terminals by the expansion factor.
Element 1: (S001, t, 23, male, a1): 2.5
Element 2: (S001, t, 33, male, a2): 2.0
Element 3: (S001, t, 31, male, a3): 2.0
Element 4: (S001, t, 85, male, a4): 20.0

  The area position information indicating the position of the area included in the area data may be point data including a representative point (point) representing the area in addition to the information indicating the area range. Hereinafter, a case where the number of users is calculated as a population will be described as an example.

[Configuration of location information analyzer]
Next, the position information analysis apparatus according to the present embodiment will be described. FIG. 2 shows a functional block configuration of the position information analysis apparatus 600. As shown in FIG. 2, the position information analysis apparatus 600 includes an input unit 601, an integration unit 602, an extraction unit 603, a removal unit 604, an area extraction unit 605, an area distribution unit 606, a cluster center determination unit 607, and a reintegration unit. 608, a totaling unit 609, and a cluster information database 610 are provided.

  The position information analysis apparatus 600 corresponds to the mobile demography unit 503 in FIG. 1, and the position information database 620 corresponds to the petamining unit 502 in FIG. However, a configuration corresponding to the petamining unit 502 in FIG. 1 may be employed in the configuration unit of the location information analysis apparatus 600 and the location information database 620.

  Hereinafter, functions of each unit of the position information analysis apparatus 600 of FIG. 2 will be described. The input unit 601 reads point data, which is area data over a plurality of times for a plurality of areas, from the position information database 620 and inputs the point data to the position information analysis apparatus 600. The integration unit 602 integrates point data input for each of a plurality of types of BTSs 200. The extraction unit 603 extracts data corresponding to the target time and target area to be processed from the point data integrated by the integration unit 602. Further, the removal unit 604 removes and outputs data whose population is 0 from the point data extracted by the extraction unit 603, and temporarily stores the removed data.

  The area extraction unit 605 randomly extracts data for a predetermined number of clusters set in advance from the point data output from the removal unit 604. This “cluster” is a unit of a collective area that collects a plurality of existing sectors in a target area (for example, all over Japan, the Tokyo metropolitan area, etc.). In this embodiment, the target area is centered on a population concentration point. This means a divided area that is divided into power ranges. That is, the area extraction unit 605 determines the point indicated by the extracted point data as the temporary center position of the cluster.

  The area distribution unit 606 determines a cluster to which the serving sector corresponding to the point data output from the removal unit 604 belongs, and distributes each point data for each cluster. Specifically, the distance between the temporary center position of all the clusters of the predetermined number of clusters determined by the area extraction unit 605 and the position indicated by the area position information included in the point data is determined, and the cluster having the closest distance is determined. Is a cluster to which the point data belongs. Further, the area distribution unit 606 determines the distance between the center positions of all the clusters re-determined by the cluster center determination unit 607 and the position indicated by the area position information included in the point data, and the cluster having the closest distance Is set as a cluster to which the point data belongs. Then, the area distribution unit 606 gives each point data a cluster ID for identifying the cluster to which it belongs.

  The cluster center determination unit 607 re-determines the center position of each cluster having a predetermined number of clusters based on the point data distributed by the area distribution unit 606. That is, the cluster center determination unit 607 determines a position where the population is concentrated for each cluster, and sets the position as the center position of the cluster. More specifically, the cluster center determination unit 607 performs weighting addition on the coordinate value indicated by the position information included in the point data belonging to each cluster and performs the weighted addition according to the population included in the point data, thereby The center of gravity position is calculated, and the center of gravity position is set as the center position of each cluster.

  The reintegration unit 608 incorporates the point data removed by the removal unit 604 into the point data to which the cluster ID is assigned by the area distribution unit 606, and performs cluster ID on the incorporated point data in the same manner as the area distribution unit 606. Is granted. Further, the totaling unit 609 totals the population for each cluster of a predetermined number of clusters based on the point data output from the reintegration unit 608, and indicates the total population for each cluster indicating the center position of the cluster. The information is stored in the cluster information database 610 in association with each cluster.

  [Procedure of location information analysis method]

  Hereinafter, the operation of the position information analysis apparatus 600 will be described with reference to FIGS. 3 to 13 and the position information analysis method in the position information analysis apparatus 600 will be described in detail. FIG. 3 is a flowchart showing an operation at the time of position information analysis by the position information analysis apparatus 600, and FIGS. 4 to 13 are diagrams showing a configuration of data to be processed at the time of position information analysis by the position information analysis apparatus 600.

  First, the input unit 601 of the position information analysis apparatus 600 reads point data for each of a plurality of types of BTSs 200 from the position information database 620 and inputs them into the position information analysis apparatus 600 (steps S101 and S102). In the present embodiment, point data is input for each 2G base station that handles the 2 GHz band and each 800M base station that handles the 800 MHz band based on the aggregated data in which the population in the sector is aggregated.

  4 and 5 show examples of point data of the 2G base station and the 800M base station input by the input unit 601, respectively. As shown in FIG. 4, the point data includes an “area ID” that identifies the area indicated by the point data, a “shape type” that indicates the type of the point data, and a “time” that indicates the date and time when the point data is aggregated. , “Latitude” and “Longitude” which are area position information indicating the position of the representative point, and “Population” which is a total value (statistical value) regarding the population in the area indicated by the point data are data associated with each other. is there. For example, “shape type: Point” means point data representing a sector in the BTS 200 as a representative point, “latitude: ya1, longitude: xa1” indicates a representative point of the sector, and “population: 100”. Indicates a statistical value obtained by converting the number of location registration signals to the BTS 200 into a population. The input unit 601 inputs a plurality of such point data for each 2G base station and 800M base station in correspondence with the existing sectors (areas) of a plurality of BTSs 200 and a plurality of aggregation dates.

  Next, the integration unit 602 integrates a plurality of pieces of point data for each 2G base station and 800M base station input by the input unit 601 (step S103, FIG. 6). Then, the extraction unit 603 extracts data corresponding to the target time and target area to be subjected to the cluster determination process from the plurality of integrated point data (step S104, FIG. 7). FIG. 7 shows point data obtained when data corresponding to the target time “January 1, 2011, 1 am” and the target area “within Tokyo” is extracted from the point data shown in FIG. It is an example.

Further, the removal unit 604 removes data having a statistical value “population” of “0” from the plurality of point data extracted by the extraction unit 603, and temporarily stores the removed data (step S105, FIG. 8). . This process is a measure for ensuring that the calculation process is normally executed in the calculation process of the center position of the cluster, which will be described later. Thereafter, the area extraction unit 605 provisionally determines the center position of the cluster having the predetermined number of clusters by randomly extracting a predetermined number of clusters from the plurality of point data output from the removal unit 604 (step S106, FIG. 9). FIG. 9 shows that two point data PD ₁ and PD ₂ are extracted from a plurality of point data, and two representative points “latitude: ya2, longitude included in these point data PD ₁ and PD ₂ ”. : Xa2 ”,“ latitude: yb1, longitude: xb1 ”are determined as the center positions of the two clusters identified by the cluster IDs“ c1 ”and“ c2 ”, and the determined center positions are associated with the cluster IDs. This is an example of creating cluster data CD.

Next, the area distribution unit 606 distributes each point data to each cluster by determining which cluster each point data belongs to by targeting a plurality of pieces of point data output from the removal unit 604 (step S107). , FIG. 10). FIG. 10 shows point data to which a cluster ID for identifying a distribution destination cluster among the two clusters determined by the area extraction unit 605 as a result of the processing is given. Specifically, for each point data, the area distribution unit 606 determines the position indicated by the representative point “latitude and longitude” included in the point data, and the center positions of all clusters determined in step S106 or step S108 below. And the cluster having the shortest distance from the representative point is determined as the distribution destination cluster. For example, according to the example of FIGS. 9 and 10, the representative point "latitude: ya1, Longitude: xa1" point data PD ₃ of its representative point, the central point of the cluster ID "c2""Latitude: yb1, longitude : The center point “latitude: ya2 and longitude: xa2” of the cluster ID “c1” is closer to the cluster with the cluster ID “c1” than the cluster ID “c1”.

Thereafter, based on the point data distributed to each cluster, the cluster center determination unit 607 re-determines the center position of each cluster (step S108). Specifically, the cluster center determination unit 607 applies the weights of the coordinates indicated by the representative points included in the point data to the point data to which the cluster ID of each cluster is assigned by the statistical value “population”. To calculate the coordinates of the center position of each cluster. For example, there are three point data belonging to the cluster with the cluster ID “c1”, and the combinations of the coordinates of the representative points and the population included in each of them are (X ₁ , W ₁ ), (X ₂ , W ₂ ), (X ₃ , W ₃ ), the following formula:
(W ₁ X ₁ + W ₂ X ₂ + W ₃ X ₃ ) / W ₁ + W ₂ + W ₃
To calculate the coordinates of the center position of the cluster having the cluster ID “c1”. In FIG. 11, the coordinates “latitude: yc1, longitude: xc1”, “latitude: yc2, longitude: xc2” of the center position calculated for each cluster ID “c1” and “c2” by the cluster center determination unit 607 The case where it is reflected in the cluster data CD created in step S106 is shown.

  Further, the cluster center determination unit 607 determines the relationship between the center coordinates calculated in step S108 and the center coordinates (the center coordinates determined last time) included in the cluster data CD before the calculation. Are matched (step S109). Although the position match is more preferable, the cluster center determination unit 607 calculates the distance between the center coordinates calculated in step S108 and the center coordinates included in the cluster data CD before calculation, and the distance is predetermined. You may determine whether it is below a value. As a result of the determination, if it is determined that the distance between the center coordinates exceeds a predetermined value (step S109; NO), the process returns to step S107, and the points are referred to while referring to the newly calculated center coordinates. Data distribution processing (step S107) and cluster center position determination processing (S108) are repeated. On the other hand, when the distance between the center coordinates is equal to or smaller than the predetermined value (step S109; YES), the determined center coordinates are confirmed.

Then, the reintegration unit 608 reintegrates the cluster data removed in step S105 with the point data that has been distributed to the clusters (step S110). At that time, the reintegration unit 608 distributes the incorporated cluster data to one of the clusters and assigns a cluster ID in the same manner as in step S107. In FIG. 12, the point data PD ₄ identified by the area ID “b4” removed in step S105 is incorporated into the point data generated in step S107 after the cluster ID “c1” is given. Show.

  Finally, the totaling unit 609 calculates the population for each cluster by counting the statistical value “population” for each cluster ID for the plurality of point data created in step S110. Furthermore, the totaling unit 609 determines the geographical range of each cluster based on the center position of each cluster (step S111). Then, the totaling unit 609 stores data indicating the population and the geographical range for each cluster in the cluster information database 610. For example, as shown in FIG. 14, the geographical range is determined by calculating a midline CL that is equidistant from the center position of two adjacent clusters “c1” and “c2” in all clusters, This is done by obtaining a range surrounded by a polygon formed by connecting a plurality of midlines calculated for a cluster (Voronoi division). FIG. 13 shows the result of totaling populations “population: 900” and “population: 400” for each cluster ID “c1” and “c2”. The determination of the geographical range of the cluster is not limited to when it is performed by Voronoi division, but is performed by integrating areas such as meshes and existing sectors corresponding to the point data distributed to each cluster. Also good. However, since the area distribution unit 606 belongs the point data to the cluster at the central position close to the representative point, from the viewpoint of matching the total population for each cluster with the population within the cluster range, a method by Voronoi division is used. desirable.

  The data stored in the cluster information database 610 in step S111 is referred to by the visualization solution unit 504 (FIG. 1), and the visualization solution unit 504 processes the range and center position of each cluster on the map. FIG. 15 shows the clusters c1 and c2 mapped by the visualization solution unit 504 and their center positions pc1 and pc2.

  According to the position information analysis device 600 and the position information analysis method described above, a plurality of point data including the area position information and the statistical values related to the population in the area are input, and a predetermined number of clusters is selected from the plurality of point data. Minutes are extracted, and the center position of a predetermined number of clusters is determined from the area position information included in the extracted point data. Further, a plurality of point data is distributed to each cluster by determining the distance between the center position of the cluster of a predetermined number of clusters and the area position information, and a statistical value included in the point data for each cluster using the distribution result Are counted. This makes it possible to specify the distribution of the population of a plurality of areas included in each cluster for a predetermined number of clusters, and to appropriately determine the concentration area of the population for each cluster of the predetermined number of clusters according to the actual population distribution. it can. In particular, the power range of the concentration point centered on the concentration point of the population in a desired area can be determined by a desired number. Thereby, the information of the predetermined number of population concentration points and the power range obtained can be effectively used as disaster prevention related information when planning a disaster prevention related base such as a shelter.

  Further, the processing load can be reduced by processing so as to obtain a group of the serving sectors by using the population of each serving sector of the BTS 200. Further, by obtaining the regional range of the cluster using Voronoi division, the statistical value for each cluster obtained by collecting the point data can be obtained as a numerical value close to the actual population within the cluster range.

  Further, in the position information analysis device 600, based on a plurality of point data distributed by the area distribution unit 606, the position where the population in each cluster is concentrated is determined, so that the position is the center position of each cluster. The area distribution unit 606 refers to the center position determined by the cluster center determination unit 607 and redistributes the plurality of point data to each cluster. Therefore, the area distribution unit 606 relies on the actual concentration area of the population. A center position can be determined, and a plurality of areas indicated by point data can be distributed to clusters according to the center position. As a result, it is possible to appropriately set the cluster range that groups a plurality of areas in correspondence with the population concentration position. In particular, by determining the center-of-gravity position for each cluster as the center position of each cluster, the concentration position of the population in each cluster can be easily calculated.

  Further, the area distribution unit 606 repeats the redistribution of the plurality of point data until the center position determined by the cluster center determination unit 607 matches the previously determined center position. It is possible to set a cluster that is further reflected.

  In addition, since the determination of the geographical range of the cluster is performed by Voronoi division, when the point data corresponding to the 2G base station and the 800M base station with which communication areas may overlap is input, the cluster is not overlapped. An area can be defined, and an area that divides each cluster uniformly can be created. For example, the size of the serving sector is different between the 2G base station and the 800M base station, and in the method of defining the cluster area by combining the areas of the serving sectors included in each cluster, a plurality of clusters are The part to which it belongs will occur. According to Voronoi division, such a problem does not occur, areas can be defined without duplication, and areas that divide clusters are uniquely created. In other words, a serving sector defined as belonging to one cluster does not occur in the area of the other cluster.

  In addition, this invention is not limited to embodiment mentioned above. For example, the cluster center determination unit 607 of the position information analysis apparatus 600 calculates the center-of-gravity position from the point data distributed to each cluster. However, the cluster center position may be determined by another calculation method.

  FIG. 16 shows a configuration of a position information analysis apparatus 600A which is a modification of the present invention, and FIG. 17 shows a flowchart showing an operation at the time of position information analysis by the position information analysis apparatus 600A. The only difference between the location information analysis device 600A and the location information analysis device 600 is the function of the cluster center determination unit 607A, and the processing of steps S203 to S207 and S209 to S211 shown in FIG. 17 is the step shown in FIG. It is the same as the processes of S103 to S107 and S109 to S111, and only the processes of steps S101, S102, and S208 are different. 18 to 20 are diagrams showing the structure of data to be processed during position information analysis by the position information analysis apparatus 600A.

  That is, in steps S201 and S202, the input unit 601 of the position information analysis apparatus 600A inputs the point data including the area “area area” of the area indicated by the point data, and the statistical value “population” of the point data. And information on the population density (user density) in the area calculated from the area “area area”. FIG. 18 shows an example in which “area area: 1000” and “population density: 0.1” calculated from “area area” and “population: 100” are added to the point data of area ID “a1”. Is shown. Thereafter, in step S207, the point data is distributed to each cluster (FIG. 19). Further, in step S208, the cluster center determining unit 607A refers to the population density information included in the plurality of point data belonging to each cluster, extracts the point data having the highest user density, and includes the point data in the corresponding point data. The representative point to be determined is determined as the center position of each cluster.

For example, in Figure 20, clusters ID "c1" is extracted point data PD ₆ having the highest population density "0.37" at the point in the data that has been granted, the central position "included in the point data PD ₆ Point data having the largest population density “0.25” in the point data to which the latitude: ya3 and longitude: xa3 ”are reflected in the center position of the cluster“ c1 ”in the cluster data CD and the cluster ID“ c2 ”is assigned. PD ₅ are extracted, the center position is included in the point data PD ₅ "latitude: ya2, longitude: xa2" indicates the case where it is reflected to the center position of the cluster "c2" in the cluster data CD.

  According to such position information analysis apparatus 600A, it is possible to easily calculate the concentration position of the population in each cluster. In addition, by setting a position where a user is likely to exist as a cluster center, a population concentration point can be obtained appropriately.

  FIG. 21 shows a configuration of a position information analyzing apparatus 600B which is a modification of the present invention, and FIG. 22 shows a flowchart showing an operation at the time of cluster center calculation processing by the position information analyzing apparatus 600B. The difference between the position information analysis apparatus 600B and the position information analysis apparatus 600A is that the cluster density calculation unit 611 is different from the function of the cluster center determination unit 607B. The operation at that time is different only in the operation in step S208 shown in FIG. FIG. 23 to FIG. 26 are diagrams illustrating a configuration of data to be processed during cluster center calculation processing by the cluster center determination unit 607B.

  That is, for the point data output from the area distribution unit 606, the cluster center determination unit 607B calculates the center of each cluster by calculating the center of gravity in the same manner as the cluster center determination unit 607 of the position information analysis device 600. The position is determined (step S301). Further, the cluster center determining unit 607B specifies point data corresponding to the serving sector (area) included in the determined cluster center position for each cluster (step S302). In FIG. 23, the point data “a2” and “b1” corresponding to the existing sector (area) included in the center positions of the clusters “c2” and “c1” are specified, and the flag “FI” is assigned to each of them. Shows the case.

  Next, the cluster density calculation unit 611 calculates the population “cluster population” and the area “cluster area area” for each cluster by adding the point data distributed by the area distribution unit 606 for each cluster. By dividing the “cluster population” by the “cluster area area”, the population density (user density) “cluster population density” for each cluster is calculated (step S303). In FIG. 24, “cluster population: 900”, “cluster population density: 0.23”, and “cluster area area: 3976” are calculated in association with the corresponding cluster ID for the cluster with cluster ID “c1”. An example of primary storage is shown.

  Thereafter, the cluster center determination unit 607B compares, for each cluster, the population density (sector population density) regarding the point data of the area included in the cluster center position and the cluster population density corresponding to each cluster (step S304). ). As a result of the comparison, when it is determined that the sector population density is equal to or higher than the cluster population density (step S304; YES), the cluster center determination unit 607B determines the center of gravity determined in step S301 as the center position of each cluster. (Step S305). On the other hand, if it is determined as a result of the comparison that the sector population density is less than the cluster population density (step S304; NO), the cluster center determination unit is similar to the cluster center determination unit 607A of the location information analysis device 600A. By 607B, point data having the largest population density is extracted from a plurality of point data belonging to each cluster, and a representative point included in the corresponding point data is determined as the center position of each cluster (step S306). In FIG. 24 to FIG. 26, for the cluster with cluster ID “c1”, the sector population density “0.1” is smaller than the cluster population density “0.23”, so the population density is the largest in the point data. The representative point “latitude: ya3, longitude: xa3” of the point data “a3” is determined as the center position of the cluster. For the cluster with the cluster ID “c2”, the sector population density “0.25” is the cluster population density “0”. Since it is larger than .22 ”, an example is shown in which the center of gravity“ latitude: yc2, longitude: xc2 ”that has already been calculated is determined as the center position of the cluster.

  According to such position information analysis apparatus 600B, a method for determining the center position of the cluster is appropriately selected between the calculation based on the center of gravity position and the area extraction based on comparison between areas of population density. The population concentration point in each cluster can be calculated appropriately.

  600, 600A, 600B: position information analyzer, 601: input unit, 602 ... integration unit, 603 ... extraction unit, 604 ... removal unit, 605 ... area extraction unit, 606 ... area distribution unit, 607, 607A, 607B ... cluster Center determining unit, 608... Reintegration unit, 609... Aggregation unit, 610. Cluster information database, 611.

Claims (8)

Area data including area position information indicating the position of the area, and statistics regarding the number of users in the area, an input unit for inputting a plurality of the area data for a plurality of areas;
A predetermined number of area data is extracted from the plurality of input area data, and a position corresponding to the area position information included in the predetermined number of area data is determined as a center position of a cluster that collects the plurality of areas. An area extractor;
By determining the distance between the center position of the predetermined number of the clusters and the area position information, the cluster to which the area corresponding to the plurality of area data belongs is determined, and the plurality of area data is assigned to each cluster. An area distribution unit that distributes to
Based on a plurality of the area data distributed by the area distribution unit, a totaling unit that totals the statistical values for each cluster;
Cluster center determination for determining the position where the number of users in each cluster is concentrated based on the plurality of area data distributed by the area distribution unit, and determining the position as the center position of each cluster With
The area distribution unit refers to the center position determined by the cluster center determination unit and redistributes the plurality of area data to each cluster.
A position information analyzing apparatus characterized by that. The area distribution unit repeats redistribution of the plurality of area data until the center position determined by the cluster center determination unit matches the center position determined last time.
The position information analyzer according to claim 1. The cluster center determination unit, for the area position information included in a plurality of the area data, is weighted by the statistical information included in the plurality of area data, and added to the center of gravity position for each cluster. Calculating the center of gravity position as the center position of each cluster,
The position information analyzing apparatus according to claim 1 or 2, characterized in that The input unit inputs the area data including density information related to a user density in the area,
The cluster center determination unit refers to the density information included in the plurality of area data, extracts the area position information corresponding to the area having the highest user density belonging to each cluster, and extracts the area position information as the area position information. Determining the corresponding position as the center position of each cluster;
The position information analyzing apparatus according to claim 1 or 2, characterized in that Based on the plurality of area data distributed by the area distribution unit, further comprising a cluster density calculation unit for calculating a user density for each cluster,
The input unit inputs the area data including density information related to a user density in the area,
When the cluster center determination unit identifies the area data corresponding to the area including the barycentric position for each cluster, and the density information included in the area data indicates a user density or more for each cluster The center of gravity position is determined as the center position of each cluster,
When the density information included in the area data indicates a user density equal to or lower than the user density for each cluster, an area having the highest user density belonging to each cluster is referred to by referring to the density information included in the plurality of area data. The area position information corresponding to is extracted, and the position corresponding to the area position information is determined as the center position of each cluster.
The position information analyzer according to claim 3. The input unit inputs area data corresponding to a plurality of types of radio base stations via a communication network based on location registration signals for a plurality of types of radio base stations communicating at a plurality of radio frequencies,
An integration unit that integrates area data corresponding to the plurality of types of radio base stations input by the input unit;
The position information analyzer according to any one of claims 1 to 5, wherein The aggregation unit determines a range obtained by connecting equidistant middle lines from the center positions of the plurality of clusters as a geographical range of the plurality of clusters.
The position information analyzer according to claim 6. A location information analysis method executed in a location information analyzer,
An input unit is area data including area position information indicating the position of the area and a statistical value related to the number of users in the area, and an input step for inputting a plurality of the area data for a plurality of areas;
The area extraction unit extracts a predetermined number of area data from the plurality of input area data, and a position corresponding to the area position information included in the predetermined number of area data is a cluster of the plurality of areas. An area extraction step to determine the center position;
The area distribution unit determines a cluster to which the area corresponding to a plurality of the area data belongs by determining a distance between the center position of the predetermined number of the clusters and the area position information. An area distribution step for distributing area data to each cluster;
A tabulation step in which the tabulation unit tabulates the statistical value for each cluster based on the plurality of area data distributed by the area distribution unit;
A cluster center determination unit determines a position where the number of users in each cluster is concentrated based on the plurality of area data distributed by the area distribution unit, thereby determining the position of the center of each cluster. And a cluster center determining step for determining as
In the area distribution step, with reference to the center position determined by the cluster center determination unit, a plurality of the area data are redistributed to each cluster.
A location information analysis method characterized by the above.

Images