JP2013034103A - Database server, system, program, and method for identifying target area from position information including positioning error - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a database server and the like capable of identifying a target area on the basis of position information on a large number of mobile terminals, even when position information including a positioning error is acquired only a few number of times from each mobile terminal.SOLUTION: A database server, which has position information storage means for storing position information of a mobile terminal and identifies a target area, comprises: clustering means for classifying a plurality of pieces of position information within a predetermined position range into one or more clusters on the basis of similarity having a base station identifier of a connection destination/adjacent base station and a radio quality value for each of the base station as elements; area extraction means for extracting, for each cluster, a confidence oval area based on a predetermined confidence level value from a plurality of pieces of position information included in the cluster.

Description

  The present invention relates to a technique for specifying a target area based on position information of a mobile terminal.

  The wireless communication quality of a mobile terminal in a wireless communication system is greatly affected by the surrounding communication environment such as a shield with the base station. Conventionally, a researcher of a telecommunications carrier had to visit each area in order to measure the quality of wireless communication with a base station. At this time, in order for the communication carrier to manage the wireless communication quality and the position information in association with each other, the accuracy of the position information is extremely important.

  Conventionally, there are various techniques for calculating position information and determining the accuracy of position information.

  There is a technique for accurately determining the current position of a vehicle for a system for calculating the current position (see, for example, Patent Documents 1 and 2). According to this technique, a highly reliable candidate point such as a branch point is calculated by using the traveling direction, travel distance, and road data of the vehicle. There is also a technique for calculating candidate points by further using relative displacement (see, for example, Patent Documents 3 and 4). Furthermore, there is a technique for calculating a candidate point with high reliability based on the map matching based on the determination result of whether or not the vehicle has turned a predetermined angle or more (see, for example, Patent Document 5).

  Further, there is a map matching technique for estimating the current position by measuring the electric field strength from the base station of the mobile phone system a plurality of times and identifying the moving direction of the mobile phone from the smoothed measured electric field strength (for example, patents) Reference 6). According to this technique, the current position is estimated based on electric field intensity data using electric field-coordinate conversion and a moving trace based on the moving direction.

  Furthermore, there is a technique for determining the reliability in GPS (Global Positioning System) positioning and correcting the vehicle position (see, for example, Patent Document 7). According to this technology, the travel distance of a predetermined section obtained by self-contained navigation is compared with the distance of a section corresponding to the predetermined section obtained by GPS, and the reliability of GPS positioning is calculated according to the difference value. .

  Furthermore, there is a technique for determining the reliability of positioning data for a map display device (see, for example, Patent Document 8). According to this technique, the fact that the position at the same reference voltage changes with time due to crustal movement or the like is used. The reliability of a plurality of past positioning data is determined depending on whether or not it is moving in a certain direction along with the change in the acquisition time. Further, the reliability of the newly acquired positioning data is determined depending on whether or not the measured data values in the past match a certain direction.

  Furthermore, there is a technique for displaying the current position of the vehicle on the ring road with certainty even while traveling on the ring road (see, for example, Patent Document 9). According to this technique, the current position is calculated using the link data of the link formed on the ring road and the measured values of the travel distance and traveling direction of the vehicle.

  Further, regarding the position information detection system, there is a technique for displaying the size of a point to be displayed as a size corresponding to the accuracy of the position information when displaying the position information of the mobile terminal on the map screen (for example, patents). Reference 10). According to this technology, position information obtained from a plurality of types of mobile radio communication systems such as a mobile phone network, a wireless LAN, and GPS is used. Then, a correction frequency distribution is created from the weight based on the estimation accuracy of various position information and the frequency distribution of the position information. The average position of this correction frequency distribution is determined as position information of the mobile terminal. Then, the accuracy of the position information is determined based on the correction frequency distribution.

  Furthermore, there is a technique for determining the walking behavior of a pedestrian and estimating the position of the pedestrian using a distance sensor, an orientation sensor, an altitude sensor, or the like (for example, see Patent Document 11).

  According to the techniques described in Patent Documents 1 to 11 described above, it is possible to detect a location where wireless communication quality has deteriorated by managing the positional information obtained by positioning and the wireless communication quality in association with each other. As another technique, there is a technique for estimating a significant position of a user who has a mobile terminal (see, for example, Non-Patent Documents 2 to 4). The “significant position” refers to a position range that is meaningful to the user, such as “the neighborhood of the home and the nearest station”. According to this technology, continuous / discontinuous position information in a mobile terminal is acquired, and clustering is performed based on the proximity of the acquisition times. As a result, the behavior of the user is modeled and traced, and an area range (significant position) with a relatively long stay time is extracted.

  In contrast to the technique described above, there is a technique in which a mobile terminal transmits location information to a server to measure a degradation point (dead zone) of wireless communication quality (see, for example, Patent Documents 12 and 14). According to this technique, when the mobile terminal detects that the mobile terminal is out of the service area of the base station, the mobile terminal performs positioning by the GPS function and stores the position information. Thereafter, when the mobile terminal detects again that it has entered the service area of the base station, it transmits the already stored location information to the server via the network. As a result, the telecommunications carrier that operates the server can collect the location information of the location where the wireless communication quality is degraded.

  There is also a technique for creating an area map based on wireless communication quality (see, for example, Patent Document 13). According to this technique, it is possible to reduce the number / amount of information transmission from the mobile terminal and the network load by changing the range of wireless communication quality transmitted from the mobile terminal to the server according to the type of the communication system. . For example, for a wide area system such as a cellular system, a mobile terminal acquires wireless communication quality and location information when wireless communication quality deteriorates below a predetermined quality. Also, for a narrow area system such as a wireless LAN, when the wireless communication quality exceeds a predetermined quality, the wireless communication quality and position information are acquired. The mobile terminal transmits the acquired wireless communication quality and position information to the server, and the server creates area maps in different communication systems.

JP 08-292050 A Japanese Patent Laid-Open No. 08-334336 Japanese Patent Laid-Open No. 08-334367 Japanese Patent Laid-Open No. 08-334370 JP 08-334372 A Japanese Patent Laid-Open No. 2000-213948 JP 2003-279362 A JP 2004-144711 A JP 2006-292928 A JP 2009-272742 A JP 2009-229204 A Japanese Patent Laying-Open No. 2005-210530 JP 2008-306240 A JP 2010-088074 A JP 2009-284495 A

“Hybrid Positioning In CDMA Networks, Invitational Workshop on Opportunistic RF Localization for Next Generation Wireless Devices”, [online], [Search January 6, 2011], Internet <URL: http://www.cwins.wpi.edu/ workshop08 / pres / tech_3.pdf> "A User-Centered Location Model, Personal and Ubiquitous Computing 2002" Toyama Midori, Hattori Takashi, Kanno Tatsuya, "Learning Significant Positions Using Mobile Phone Positioning Functions", [online], [January 7, 2011], Internet <URL: http: //www.tom .sfc.keio.ac.jp / ~ next / papers / 2005-ipsj.pdf> Shigeki Kurokawa, Hiroyuki Yokoyama, Kazuyoshi Yoshii, Hideki Aso, “Examination of Significant Location Extraction Method from Sparse Location Information History”, 13th Information Theory Learning Theory Workshop (IBIS 2010), [online], [ January 7, 2011], Internet <http://ibisml.org/ibis2010/preview/P4-6.pdf> Introduction to statistics, “9.4 Multivariate case”, [online], [Search January 7, 2011], Internet <URL: http://www.snap-tck.com/room04/c01/stat/stat09 /stat0904.html>

  According to the techniques described in Patent Documents 12 to 14 described above, an area where radio degradation of a predetermined threshold or more has occurred in a wide area radio communication system or an area where radio communication can be performed in a narrow area radio communication system is roughly specified. be able to.

  However, any positioning method that can be realized by the mobile terminal, such as the GPS positioning method, the multiple base station positioning method, and the hybrid positioning method, causes a positioning error. For this reason, even if degradation of wireless communication quality or the like is detected, it is necessary to allow the position information to include an error, and it is relatively difficult to acquire an accurate position. For example, the positioning error based on these positioning methods may range from several tens of meters to several hundreds of meters (see, for example, Non-Patent Document 1).

  The main reason for this is that for wide-area wireless communication systems such as mobile phone networks, wireless communication quality degradation occurs indoors or behind buildings, and for narrow-area systems such as wireless LANs, This is because they are often indoors. In any positioning method such as GPS positioning, multiple base station positioning, and hybrid positioning, when positioning is performed indoors, the positioning error is larger than usual.

  In particular, according to the techniques described in Patent Documents 12 to 14, the purpose is to acquire position information of a location where wireless communication quality is deteriorated, and how much error is included in the position information. Is extremely difficult to estimate.

  Further, according to the technique described in Patent Document 10, a correction frequency distribution is created using a plurality of pieces of position information obtained from a plurality of types of mobile radio communication systems such as a mobile phone network, a wireless LAN, and a GPS. Then, according to the accuracy of the position information based on the corrected frequency distribution, the size of the spot displayed on the map screen is changed to allow the user to understand the relative accuracy.

  FIG. 1 is a map showing position information notified from a mobile terminal when radio deterioration equal to or greater than a predetermined threshold occurs.

  For a wide area radio communication system, the mobile terminal detects radio degradation (decrease in radio communication quality) equal to or greater than a predetermined threshold. And a mobile terminal transmits the positional information at the time of radio | wireless degradation to a database server via a base station. Thereby, the database server can plot the position of the radio degradation of the mobile terminal on the map.

  According to FIG. 1, the position of the radio degradation of the measured mobile terminal is represented by ●, and the number in it indicates the identifier of the mobile terminal. That is, according to FIG. 1, each position where radio degradation is detected is detected by another mobile terminal. Moreover, according to FIG. 1, the position of the base station 3 communicating with the mobile terminal is also shown, and base station identifiers A to C are given.

  FIG. 2 is a map showing a place where radio degradation of a predetermined threshold value or more actually occurs.

According to FIG. 2, radio deterioration of a predetermined threshold or more occurs at three locations a to c.
Mobile terminals 1 to 9: Experienced radio degradation of a predetermined threshold value or more for the radio degradation location a.
Mobile terminals 10 to 15: Experienced radio degradation of a predetermined threshold or more at the radio degradation location a.
Mobile terminals 16 to 18: Experienced radio degradation of a predetermined threshold value or more in the radio degradation location c.

  FIG. 3 is a map showing a place where radio deterioration equal to or greater than a predetermined threshold should be ideally detected.

  Originally, positioning is ideally performed as shown in FIG. However, in reality, as shown in FIGS. 1 and 2, the position information obtained by positioning is not the same location as the wireless degradation location (a, i, or c). This is due to positioning errors.

  As described above, according to the prior art, even if radio degradation exceeding a predetermined threshold occurs in one area due to the influence of positioning error, it is possible to specify the actual radio degradation area. difficult. In addition, when there are a plurality of radio degradation areas, it is more difficult to specify where the mobile terminal experienced the radio degradation.

  In addition, in any of the above-described conventional techniques (including Patent Documents 1 to 11 and Non-Patent Documents 2 to 3 described above), the same mobile terminal acquires a plurality of pieces of position information that are continuous at the same location. It is assumed that. Therefore, it is not assumed that one mobile terminal has acquired position information only a few times such as once or twice. In particular, according to the technique described in Patent Document 10, the correction frequency distribution cannot be created with a small number of position information such as one or two times.

  According to the above-described example, the area of radio degradation is specified. However, the same problem arises even when specifying an area that satisfies a predetermined condition without being limited to radio communication quality. That is, the positioning error is affected.

  Therefore, according to the present invention, even if only a small number of location information of each mobile terminal including a positioning error can be accumulated, the actual target area can be determined by using the location information of a large number of mobile terminals. An object of the present invention is to provide a database server, system, program, and method that can be specified as accurately as possible.

According to the present invention, there is provided a database server that has position information storage means for storing position information of a mobile terminal, and identifies a target area,
A plurality of pieces of position information within a predetermined position range are classified into one or more clusters on the basis of the similarity using the base station identifier of the connection destination / adjacent base station and the radio quality value for each base station as elements. Clustering means;
Each cluster has an area extraction means for extracting a confidence ellipse area based on a predetermined confidence level value from a plurality of pieces of position information included in the cluster.

According to another embodiment of the database server of the present invention,
Regarding the area extraction means, the confidence ellipse area estimates the range of the representative value of the population based on the two-dimensional normal distribution, and the confidence level value is the probability that the representative value of the population is included inside the ellipse. It is also preferable.

According to another embodiment of the database server of the present invention,
The position information storage means stores the position information in association with the positioning accuracy,
It is also preferable to further include a map output means for outputting a map image in which the confidence ellipse area to be superimposed on the map image is visually identified based on the positioning accuracy and based on the number of terminals.

According to another embodiment of the database server of the present invention,
The location information storage means preferably stores location information of the mobile terminal that has reached a predetermined condition.

According to another embodiment of the database server of the present invention,
For the location information storage means, as a predetermined condition,
(Trigger 1) When the wireless communication quality falls below a predetermined threshold,
(Trigger 2) When predetermined data communication is executed, or
(Trigger 3) It is also preferable that the location information related to the mobile terminal when the data communication amount in a predetermined period is equal to or greater than a predetermined threshold is accumulated.

According to the present invention, there is provided a system having the aforementioned database server and a mobile terminal that transmits location information to the database server,
The mobile terminal
Positioning means for acquiring current position information when a predetermined condition is reached;
It is characterized by comprising position information transmitting means for transmitting the position information and the base station identifier of the connected / adjacent base station when the position information is measured to the database server.

According to another embodiment of the system of the present invention,
The positioning method of the mobile terminal is
(Trigger 1) When the wireless communication quality falls below a predetermined threshold,
(Trigger 2) When predetermined data communication is executed,
(Trigger 3) When the amount of data communication in a predetermined period exceeds a predetermined threshold, or
(Trigger 4) When a predetermined user operation is executed,
It is also preferable to measure the position information.

According to the present invention, there is provided a program that has a location information storage unit that stores location information of a mobile terminal, and that causes a computer to function to specify a target area.
A plurality of pieces of position information within a predetermined position range are classified into one or more clusters on the basis of the similarity using the base station identifier of the connection destination / adjacent base station and the radio quality value for each base station as elements. Clustering means;
For each cluster, the computer is caused to function as area extraction means for extracting a confidence ellipse area based on a predetermined confidence level value from a plurality of pieces of position information included in the cluster.

According to the present invention, there is provided a target area display method in an apparatus for specifying a target area having a position information storage unit that stores position information of a mobile terminal,
A plurality of pieces of position information within a predetermined position range are classified into one or more clusters on the basis of the similarity using the base station identifier of the connection destination / adjacent base station and the radio quality value for each base station as elements. A first step;
A second step of extracting, for each cluster, a confidence ellipse area based on a predetermined confidence level value from a plurality of pieces of position information included in the cluster;
It is characterized by having.

  According to the database server, system, program and method of the present invention, even if only a small number of location information of each mobile terminal including a positioning error is accumulated, based on the location information of a large number of mobile terminals, The target area can be specified.

It is a map showing the positional information notified from the mobile terminal when radio | wireless degradation more than a predetermined threshold value generate | occur | produced. It is a map showing the place where the radio | wireless degradation more than a predetermined threshold has actually occurred. It is a map showing the place where the radio degradation more than a predetermined threshold should be detected ideally. It is a system configuration diagram in the present invention. It is a sequence diagram in the present invention. It is a table of the positional information accumulate | stored in the database server. It is a dendrogram when the group average method is used. It is explanatory drawing showing a 2-dimensional normal distribution and the 95% confidence ellipse area of the representative (average) value of a population. FIG. 8 is a map representing only position information classified as [cluster-a] in FIG. 7. FIG. 8 is a map representing only position information classified as [cluster-a] in FIG. 7. FIG. 8 is a map representing only position information classified as [Cluster-U] in FIG. 7. FIG. 10 is a map in which a confidence ellipse calculated from the position information of [cluster-a] in FIG. 9 is superimposed. FIG. 10 is a map on which a confidence ellipse calculated from position information of [Cluster A] in FIG. 9 is superimposed. FIG. 10 is a map on which a confidence ellipse calculated from position information of [Cluster-U] in FIG. 9 is superimposed. It is the map which superimposed the radio | wireless degradation place of FIG. 2, and the reliability ellipse of FIGS. It is a functional block diagram of the mobile terminal and database server in this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 4 is a system configuration diagram according to the present invention.

  FIG. 4 is a multiple representation of FIG. 1 described above. According to FIG. 4, a plurality of base stations 3 are arranged, and each base station 3 communicates with a mobile terminal 1 located in the service area. All the base stations 3 are connected to the mobile communication network of the communication carrier. The mobile communication network may be a mobile phone network that covers a wide area, for example, and the mobile terminal 1 may be a mobile phone, for example.

  The service areas of the plurality of base stations 3 are arranged in a superimposed manner so that a dead zone does not occur. However, when a building such as a building is interposed between the mobile terminal and the base station, radio waves are cut off, resulting in a dead zone. As a carrier, we want to detect such dead zones as much as possible. In the following description, the “target area” to be specified by the present invention will be described as the “wireless degradation” area in the communication carrier, but it is of course not limited thereto.

  According to FIG. 4, wireless communication quality is represented on a two-dimensional map (area map). The higher the value of the z-axis representing the wireless communication quality, the better the wireless communication quality. According to FIG. 4, the wireless communication quality is represented hierarchically in a pyramid shape with the base station 3 as the center. This indicates that the closer the position of the mobile terminal 1 is to the base station 3, the better the wireless communication quality.

  Further, according to FIG. 4, a database server (LAS-DB (Location Information and Available System Information Data Base)) 2 that stores location information of mobile terminals is connected to the mobile communication network. The database server 2 can create an area map specifying the target area by receiving the position information and the wireless communication quality from the mobile terminal 1.

The mobile terminal 1 such as a mobile phone is equipped with a positioning unit and can measure the current position. Examples of the positioning unit include the following.
“GPS positioning method”: Positioning by positioning radio waves from GPS satellites “Multiple base station positioning method”: Positioning by communication radio waves with multiple neighboring base stations “Hybrid positioning method”: GPS positioning method + multiple base station positioning method

  Further, the mobile terminal 1 measures the wireless communication quality by the wireless communication interface unit for the base station 3. Examples of the wireless communication quality include reception quality, number of retransmissions or data error rate, throughput, and delay time. The reception quality includes, for example, RSS (Received Signal Strength), C / I (carrier to interference wave ratio), and Ec / Io (average energy per PN chip versus total received power spectral density).

“Good wireless communication quality” is when the received signal strength is high, the signal-to-interference noise ratio is large, the number of retransmissions is small, the data error rate is low, the throughput is high, or the delay time is short.
“Poor wireless communication quality” is when the received signal strength is low, the signal-to-interference noise ratio is small, the number of retransmissions is large, the data error rate is high, the throughput is low, or the delay time is large.

  According to FIG. 4, when the wireless communication quality deteriorates below a predetermined threshold, the mobile terminal 1 measures and stores the position at that time. And when the wireless communication quality returns to a good state, the mobile terminal 1 transmits the position information when the wireless communication quality is deteriorated to the database server 2 together with the wireless communication quality.

  As a communication carrier, by creating an area map as shown in FIG. 4, the wireless communication quality corresponding to the position of the mobile terminal can be grasped, which can be used for facility design and service quality improvement.

  FIG. 5 is a sequence diagram in the present invention.

According to FIG. 5, it is assumed that the mobile terminal 1 has moved from the service area of the base station 3 to the dead zone.
(S11) The mobile terminal 1 acquires the current position information using a trigger as reaching a predetermined condition. Here, the predetermined conditions include, for example, the following depending on the use of the target area to be extracted.
(Trigger 1) The wireless communication quality is equal to or less than a predetermined threshold (Trigger 2) The predetermined data communication is executed (Trigger 3) The data communication amount in the predetermined period is equal to or higher than the predetermined threshold (Trigger 4) The specified user operation was executed

  According to FIG. 2, since it is intended to extract a location where wireless communication quality is degraded, (Trigger 1) is applied. When the mobile terminal 1 enters the dead zone and the wireless communication quality falls below a predetermined threshold, the mobile terminal 1 activates the positioning unit and measures the current position.

  As another use of the target area to be extracted, (trigger 2) is applied when trying to extract a place where a lot of “passing communication” is performed by the mobile game terminal. “Passing communication” refers to a function in which a portable game machine automatically and instantaneously communicates with a game machine of another player that affects the progress of the game using a wireless LAN. When the passing communication is executed, the mobile terminal 1 activates the positioning unit and measures the current position. Also, (trigger 2) is applied when trying to extract a place where a predetermined application is used.

  Furthermore, when trying to extract an area with a large amount of data communication, (trigger 3) is applied.

  Further, as another application of the target area to be extracted, (trigger 4) is applied when trying to extract a place where a lot of specific photographs are taken. The mobile terminal 1 activates the positioning unit and measures the current position when a specific photograph is taken by the camera using the application.

(S12) It is also preferable that the mobile terminal 1 further acquires the positioning accuracy in the acquired position information. The positioning accuracy may be set in three stages as follows, for example.
(High accuracy) Positioning by GPS positioning method (Medium) Positioning by hybrid positioning method (Poor accuracy) Positioning by multiple base station positioning method

Further, when only the GPS positioning method is used, “good” or “bad” positioning accuracy may be determined based on an estimated mean square error value (RMSE (Root Mean Square Error)). The mean square error value is obtained by evaluating a difference from the true value X, and is represented by the following expression, for example.
RMSE = √ (Σ [i = 1, n] (X−xi) ^ 2 / n)

(S13) Thereafter, when the mobile terminal 1 further moves and communication with the base station 3 is restored, [location information, wireless communication quality information, supplementary base station information, wireless quality information with the base station, positioning accuracy] ] Is sent to the database server 2.

(S21) The database server 2 receives, for example, the following information from the mobile terminal 1.
Location information (latitude / longitude)
Wireless communication quality information (for example, reception strength)
Supplementary base station information (base station identifier of connected base station and adjacent base station)
Positioning accuracy The position information is stored in a position information storage unit.

  Here, according to the embodiment of FIG. 5, the location information and the like of the mobile terminal is measured and detected by the mobile terminal and transmitted to the database server 2. However, the database server of the present invention is not limited to receiving position information and the like from the mobile terminal 1. In other words, the database server of the present invention only needs to store position information of a large number of mobile terminals that may include positioning errors. The location information and the like of the mobile terminal may be acquired and accumulated by a telecommunications carrier network.

  According to S11, the positioning unit in the mobile terminal 1 has been described to acquire position information by using a method such as “GPS positioning method”, “multiple base station positioning method”, and “hybrid positioning method”. On the other hand, in some cases, the radio base station measures the arrival time and arrival direction information of the radio wave from the mobile phone, and the service node in the telecommunications carrier network uses the information to determine the position of the mobile terminal 1. (For example, refer to Patent Document 15). The location information of the mobile terminal used by the present invention is not limited to that measured by the mobile terminal 1, and location information measured in the communication carrier network can also be used. In this case, it is not necessary for the mobile terminal 1 to have a positioning unit. In this case, it is also preferable that the positioning trigger is issued from the telecommunications carrier network side, for example, a service node. This includes that the service node monitors the communication quality between the base station and the mobile terminal and triggers that the quality deteriorates below a predetermined threshold.

If the location information of the mobile terminal is acquired and accumulated by the telecommunications carrier network, the trigger based on the use of the target area is, for example, the following three, except for (Trigger 4) described above: Become.
(Trigger 1) The wireless communication quality has become a predetermined threshold value or less (Trigger 2) The predetermined data communication has been executed (Trigger 3) The data communication amount in the predetermined period has become the predetermined threshold value or more

  FIG. 6 is a table of position information stored in the database server.

  According to FIG. 6, 18 position information logs are accumulated, and it can be understood that each log is measured by a different terminal (terminal ID). For example, for log number 1, latitude and longitude are acquired by terminal ID1, and that terminal (ID = 1) is connected to connection base station A with reception strength 20 and adjacent base station B with reception strength 17; An adjacent base station C with a reception intensity of 7 is visible.

  The log of the positional information accumulated as shown in FIG. 6 is divided for each predetermined range. Here, the “predetermined range” indicates a four-sided range of several hundred meters to several kilometers. For example, according to FIGS. 1 to 3, the entire map represents one predetermined range, and 18 pieces of position information are accumulated therein.

(S22) The database server 2 classifies the plurality of pieces of position information within the predetermined position range into one or more clusters based on the similarity with the base station identifier of the connection destination / adjacent base station as an element. Here, it is preferable that the element for calculating the degree of similarity further includes a radio quality value for each base station together with the base station identifier of the connection destination / adjacent base station. The position information (18 pieces) included in each predetermined range in FIG. 6 is classified according to the similarity between the connected base station / adjacent base station and the radio quality information (for example, reception strength).

  FIG. 7 is a dendrogram when the group average method is used.

  A dendrogram is a representation of a tree diagram showing how individuals (logs) are grouped into clusters for cluster analysis. According to FIG. 7, the numbers arranged in the vertical direction mean log numbers (see FIG. 6), and the horizontal axis means the similarity between the connected base station / adjacent base station and its radio quality information (for example, reception strength). (The closer to the left, the higher the degree of similarity, and the closer to the right, the lower the degree of similarity).

  According to FIG. 7, the hierarchy is cut by a broken line meaning a predetermined distance. Thereby, it is divided into three clusters. The meaning of the “cluster” is characterized in that it indicates where the position information is actually measured without depending on the positioning result (position).

  For example, according to FIG. 2, the location information 7 of the mobile terminal identifier 7 is measured at [Radio degradation location a]. However, the location information 7 of the mobile terminal identifier 7 represents a location farther than [Radio degradation location a] due to the influence of positioning error. When viewed from [Radio Degradation Location A], the location information 7 of the mobile terminal identifier 7 is located farther than the location information 15 of the mobile terminal identifier 15 measured at [Radio Degradation Location A].

  However, the position information 7 is measured at [Wireless degradation location a]. Therefore, the identifiers of the connection destination base station and the adjacent base station and the radio quality value (for example, reception strength) between them are based on [radio degradation place a]. Therefore, cluster analysis is performed to calculate and compare the degree of similarity using these identifiers and radio quality values. Accordingly, the position information 7 has a high degree of similarity with other position information (for example, position information 3) measured at [wireless deterioration place a]. That is, as shown in FIG. 7, it can be classified as position information measured by [wireless degradation location a].

(S23) Next, for each cluster, the database server 2 extracts a confidence ellipse area based on a predetermined confidence level value (preferably a plurality of confidence level values) from a plurality of pieces of position information included in the cluster. As described above, the communication carrier arranges the plurality of base stations 3 so that their service areas are overlapped so that a dead zone does not occur. However, radio waves are blocked by the presence of a building such as a building between the mobile terminal and the base station, and a dead zone may occur in some areas. That is, it is preferable to consider that the clustered position information is acquired in the partial dead zone (area). Therefore, assuming that the mother representative (average) position is the area, as described above, a range where the mother representative (average) position exists with a predetermined probability is calculated as an ellipse. The database server 2 is reliable not only for each cluster but also for each positioning accuracy in the location information (that is, for each location information within the same cluster and for each accuracy of “good”, “medium”, “bad”). It is also preferable to calculate an elliptical area.

Here, a method for calculating the confidence ellipse area will be described.
[Calculation method of trusted ellipse area]
The position information measured by the mobile terminal is represented by two dimensions. Since the positioning information includes the positioning error as described above, the positioning data is based on the observation that it follows a two-dimensional normal distribution (multivariate normal distribution). Here, the confidence ellipse area of the representative (average) value of the population is estimated with the confidence level value β. It is referred to as “100 · (1-β))% confidence ellipse area”. The reliability “100 · (1-β)%” is the probability that the representative (average) value of the population is included in the ellipse. At this time, it is also preferable to determine the β value to be used from the dispersion value of the plurality of pieces of position information falling within a predetermined range and the positioning accuracy in the position information. For example, when the positioning accuracy is “good” (when the ratio of good is large or when only good is collected), the variance value is large, and a relatively small β value is used. On the other hand, when the positioning accuracy is “high” and the variance value is small, a relatively large β value is used. Even when the positioning accuracy is “bad” (when the ratio of bad is large or when only bad is collected), a relatively large β value is used as the variance value is large.

  FIG. 8 is an explanatory diagram showing a two-dimensional normal distribution and a 95% confidence ellipse area of the representative (average) value of the population.

The trust ellipse is calculated by the following equation.
(1 / n) * F (2, φ, β) = (V (ly) * X ^ 2-2 * C (lx, ly) * X * Y + V (lx) * Y ^ 2) /
2 * (V (lx) * V (ly) -C (lx, ly) ^ 2)
lx: Longitude direction representative (average) value of location information measured by mobile terminal
ly: representative () value in the latitudinal direction of position information measured by the mobile terminal X = x−lx
Y = y-ly
V (lx): Variance of lx (sample average in the longitude direction) V (ly): Variance of ly (sample average in the latitude direction) C (lx, ly): Covariance of lx and ly F (2, φ, β ): Follows the first degree of freedom 2 and the second degree of freedom φ (residual degree of freedom)
Value of 100β% point in F distribution

(S24) Then, the database server 2 outputs a map image on which the confidence ellipse area is superimposed.

When the position information storage unit 21 stores the positioning information in association with the positioning accuracy, the map image is superimposed so that the confidence ellipse area to be superimposed on the map image can be visually identified based on the positioning accuracy. Can be output. The positioning accuracy in the confidence ellipse area may be, for example, an average value of the positioning accuracy in a plurality of pieces of position information included in the confidence ellipse area. For example, it is also preferable to change the display format of the confidence ellipse according to the positioning accuracy as follows.
[Dashed line] / [Solid line]: [Number of position information: small] / [Number of position information: Many]
[Thin line] / [Thick line]: [Positioning accuracy: Low] / [Positioning accuracy: High]
[Dashed line / thin line] Number of position information: Low Positioning accuracy: Low [Dashed line / thick line] Number of position information: Small Positioning accuracy: High [Solid line / thin line] Number of position information: High Positioning precision: Low [Solid line / thick line] Number of position information : Multi-positioning accuracy: High This enables the observer to understand the positioning accuracy, including a simple glance at the map.

FIG. 9 is a map representing only position information classified as [cluster-a] in FIG.
FIG. 10 is a map representing only the position information classified as [cluster-a] in FIG.
FIG. 11 is a map representing only the position information classified as [Cluster-U] in FIG.

FIG. 12 is a map on which a confidence ellipse calculated from the position information of [cluster-a] in FIG. 9 is superimposed.
FIG. 13 is a map in which a confidence ellipse calculated from the position information of [Cluster A] in FIG. 9 is superimposed.
FIG. 14 is a map on which a confidence ellipse calculated from the position information of [Cluster-U] in FIG. 9 is superimposed.

  FIG. 15 is a map in which the wireless degradation location of FIG. 2 and the confidence ellipses of FIGS. According to FIG. 15, it can be understood that the actual wireless degradation locations are narrowed down by the trust ellipse.

  FIG. 16 is a functional configuration diagram of the mobile terminal and the database server in the present invention.

  According to FIG. 16, the mobile terminal 1 includes a wireless communication interface unit 101 that communicates with the base station 3 through the air, and a positioning unit 102 that calculates a current position based on positioning radio waves received from the GPS satellite 4 or the base station. Have. In addition, the mobile terminal 1 further includes a positioning activation unit 11, a positioning accuracy acquisition unit 12, and a position information transmission unit 13. These functional components are realized by executing a program that causes a computer mounted on the mobile terminal to function.

  The positioning activation unit 11 activates the positioning unit 102 using the fact that the predetermined condition has been reached as a trigger to acquire the current position information (performs the same processing as S11 in FIG. 5). For example, the trigger may be that the wireless communication quality has deteriorated below a predetermined threshold.

  The positioning accuracy acquisition unit 12 further acquires the positioning accuracy in the acquired position information (executes the same process as S12 in FIG. 5). For example, “good” or “bad” positioning accuracy is determined based on a positioning method or a mean square error value of GPS. The positioning accuracy is output to the position information transmission unit 13.

  The location information transmission unit 13 transmits the location information, supplementary base station identifier (connection / adjacent base station identifier), and positioning accuracy to the database server 2 for each terminal (execute processing similar to S13 in FIG. 5). ).

  The database server 2 includes a communication interface unit 201 connected to a network (a mobile communication network according to FIG. 16) and a display unit 202 that displays a map to a supervisor.

  Further, the database server 2 includes a position information storage unit 21, a clustering unit 22, an area extraction unit 23, and a map output unit 24. These functional components are realized by executing a program that causes a computer mounted on the server to function.

  The position information storage unit 21 stores [position information, wireless communication quality information, supplementary base station information, wireless quality information with a base station, positioning accuracy] of the mobile terminal 1. (The same processing as S21 in FIG. 5 is executed).

  The clustering unit 22 classifies a plurality of pieces of position information within a predetermined position range into one or more clusters based on the similarity using the base station identifiers of the connection destination / adjacent base stations as elements (S22 in FIG. 5). The same processing is executed). Here, it is preferable that the element for calculating the similarity further includes a radio quality value for each base station together with the base station identifier of the connection destination / adjacent base station.

  For each cluster, the area extraction unit 23 extracts a confidence ellipse area based on a predetermined confidence level value from a plurality of pieces of position information included in the cluster (executes processing similar to S23 in FIG. 5). At this time, a β value that is set in advance according to the positioning accuracy and the dispersion value of the position information is used.

  The map output unit 24 outputs a map image in which the confidence ellipse area is superimposed (executes processing similar to S24 in FIG. 5). Here, it is also preferable that the database server 2 clearly shows the confidence ellipse area to be superimposed on the map image on the map so that it can be visually identified based on the positioning accuracy. For example, the display format of the confidence ellipse can be changed according to the average value of the positioning accuracy of the position information included in the cluster. As a result, the map image superimposed so as to be visually identifiable can be output to the viewer.

  As described above in detail, according to the database server, system, program, and method of the present invention, even when only a small number of location information of each mobile terminal including positioning errors can be accumulated, By using the location information of the mobile terminal, the actual target area can be specified as accurately as possible.

  According to the present invention, a target area can be specified for an event that occurs in a specific area. For example, this is not limited to the investigation of the degradation area of the wireless communication quality, and can be used for marketing. For example, a place where a lot of “passing communication” is performed by the mobile game terminal can be extracted. In this case, it is preferable to process each game title. Similarly, it can also be used for extracting a place where a predetermined application is used. Furthermore, it can also be used for extracting an area where data communication exceeding a predetermined amount of data is performed within a predetermined time. Furthermore, as another application of the target area to be extracted, a place where a lot of specific photographs are taken can be extracted. In this case, it is preferable to process each specific photograph analyzed as “Tokyo Tower” or “Rainbow Bridge” by the image recognition process.

  Various changes, modifications, and omissions of the above-described various embodiments of the present invention can be easily made by those skilled in the art. The above description is merely an example, and is not intended to be restrictive. The invention is limited only as defined in the following claims and the equivalents thereto.

  The database server, system, program, and method of the present invention can be used for communication area countermeasures and marketing, for example, because the target area can be specified for an event that occurs in a specific area.

DESCRIPTION OF SYMBOLS 1 Mobile terminal 101 Wireless communication interface part 102 Positioning part 11 Positioning start part 12 Positioning accuracy acquisition part 13 Position information transmission part 2 Database server 201 Communication interface part 202 Display part 21 Position information storage part 22 Clustering part 23 Area extraction part 24 Map output Part 3 Base station 4 GPS satellite

Claims (9)

A database server that has location information storage means for storing location information of a mobile terminal and identifies a target area,
A plurality of pieces of position information within a predetermined position range are classified into one or more clusters on the basis of the similarity using the base station identifier of the connection destination / adjacent base station and the radio quality value for each base station as elements. Clustering means;
A database server comprising, for each cluster, area extraction means for extracting a confidence ellipse area based on a predetermined confidence level value from a plurality of pieces of position information included in the cluster.   Regarding the area extracting means, the confidence ellipse area estimates a range of representative values of a population based on a two-dimensional normal distribution, and the confidence level value includes a representative value of the population within an ellipse. The database server according to claim 1, wherein the database server is a probability. The location information storage means further stores the location information in association with positioning accuracy,
The system further comprises map output means for outputting a map image superimposed so that the confidence ellipse area to be superimposed on the map image can be visually identified based on the positioning accuracy and based on the number of the terminals. The database server according to claim 1 or 2.   The database server according to any one of claims 1 to 3, wherein the location information storage means stores location information of a mobile terminal that has reached a predetermined condition. For the location information storage means, as the predetermined condition,
(Trigger 1) When the wireless communication quality falls below a predetermined threshold,
(Trigger 2) When predetermined data communication is executed, or
(Trigger 3) The database server according to claim 4, wherein the location information relating to the mobile terminal when the amount of data communication during a predetermined period is equal to or greater than a predetermined threshold is accumulated. A system comprising the database server according to any one of claims 1 to 5 and a mobile terminal that transmits location information to the database server,
The mobile terminal
Positioning means for acquiring current position information when a predetermined condition is reached;
A system for identifying a target area, comprising: position information transmitting means for transmitting the position information and a base station identifier of a connection / adjacent base station when the position information is measured to the database server. . The positioning means of the mobile terminal is
(Trigger 1) When the wireless communication quality falls below a predetermined threshold,
(Trigger 2) When predetermined data communication is executed,
(Trigger 3) When the amount of data communication in a predetermined period exceeds a predetermined threshold, or
(Trigger 4) When a predetermined user operation is executed,
The system according to claim 6, wherein the position information is measured. A program that has a location information storage unit that stores location information of a mobile terminal and causes a computer to function to identify a target area,
A plurality of pieces of position information within a predetermined position range are classified into one or more clusters on the basis of the similarity using the base station identifier of the connection destination / adjacent base station and the radio quality value for each base station as elements. Clustering means;
A program for a database server, which causes a computer to function as area extraction means for extracting a confidence ellipse area based on a predetermined confidence level value from a plurality of pieces of position information included in each cluster for each cluster. A target area display method in an apparatus for identifying a target area, having a position information storage unit that stores position information of a mobile terminal,
A plurality of pieces of position information within a predetermined position range are classified into one or more clusters on the basis of the similarity using the base station identifier of the connection destination / adjacent base station and the radio quality value for each base station as elements. A first step;
A second step of extracting, for each cluster, a confidence ellipse area based on a predetermined confidence level value from a plurality of pieces of position information included in the cluster;
An object area display method characterized by comprising:

Images