JP2017143472A - Device, program and method for estimating user's movement object - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device or the like capable of estimating a movement object by taking the stay/movement of a portable terminal carried by a user into consideration regardless of position information in which spatial granularity is rough and time intervals are not constant.SOLUTION: The device includes: communication history storage means for storing a plurality of communication logs obtained by associating a time with position information in each portable terminal; first movement speed calculation means for calculating a first movement speed based on the position information and a lapse time about a prescribed fixation time window including a communication log in each communication log; second movement speed calculation means for calculating a second movement speed based on the position information and a lapse time about a fluctuation time window in which a past time is set longer by as much as a prescribed condition than the fixation time window in each communication log; a movement object reference table obtained by associating the range of the first movement speed with the range of the second movement speed in each object; and movement object estimation means for selecting a movement object including both the first movement speed and the second movement speed by using the movement object reference table.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a technique for estimating a user's behavior using a communication history in a portable terminal possessed by the user.

  Mobile terminals such as smartphones and tablets are equipped with positioning functions such as GPS (Global Positioning System) and acceleration sensors. The mobile terminal transmits the position information as a history to the server of the service provider, so that the server can derive an action history such as a moving speed and a staying time at the current position of the user.

  On the other hand, when the portable terminal activates a positioning function such as GPS or its application constantly or periodically, the consumption of the battery charge is accelerated. In addition, since each mobile terminal transmits position information to the server, there is a possibility that an increase in the amount of communication packets transmitted from the mobile terminal and network congestion due to the communication packets may occur.

  2. Description of the Related Art Conventionally, there is a technique for estimating a user's significant sphere using position information of a base station under the control of each mobile terminal that can be acquired by a communication equipment device as a communication provider side (see, for example, Patent Document 1). According to this technology, there is no need to activate the positioning function of the mobile terminal. However, the base station position information alone has disadvantages that the spatial granularity is coarse and the time interval is not constant. Therefore, according to this technique, for each mobile terminal, the stay / movement is determined by clustering position information of a large number of base stations. As a result, the accuracy of the “stay” position is lowered.

  There is also a technology on the telecommunications carrier side that estimates the significant sphere by using an agglomerative clustering method called “Leader Algorithm” for the history of location information of base stations under the control of a mobile terminal ( For example, refer nonpatent literature 1).

  Furthermore, as a communication carrier side, there is also a technology for setting an upper limit value (for example, three times) for the number of base station switching in each portable terminal and determining a series of time zones not exceeding the upper limit value as “stay”. (See Non-Patent Document 2, for example).

  Furthermore, there is a technology on the telecommunications carrier side that divides a time-series communication log into predetermined time windows and determines stay / movement based on probability distributions in position information of a plurality of base stations for each time window ( For example, see Patent Document 2). When it is determined as “stay”, the center of gravity of a plurality of pieces of position information in each time window is used as a representative point, and clustering is performed into a set having a distance close to the representative point. A position obtained from the cluster is defined as a stay position.

  Furthermore, there is a technique for determining a stay position from a communication log (see, for example, Non-Patent Document 3). According to this technique, when the position information of the time-series communication log is continuously outside the range of the threshold D, the same cluster is set immediately before the communication log that first falls outside the range of the threshold D. . However, even if communication logs that are outside the range of the threshold D appear continuously, if a communication log included in the immediately preceding cluster occurs before the predetermined number is exceeded, the communication log that is outside the range of the threshold D is excluded. Treat as a value. Even if different clusters occur successively, if the distance between the centers of the clusters is within D / 3, these clusters are regarded as the same cluster. Then, the center of gravity of the cluster is set as the stay position.

  Further, the distance from all communication logs included in a time range that is traced back by a predetermined time mintime from the time when the communication log determined to be stayed is calculated, and the distance from each communication log is within the threshold D range. For example, there is a technique for determining that they are the same “stay” (see Non-Patent Document 4, for example). Here, when it is determined that the stays are the same, the distance between the communication logs is obtained as a brute force, and the location of the communication log that minimizes the total distance is defined as the “stay position”. Further, communication logs generated thereafter are set to the same stay position if the distance from the stay position is within D / 2. On the other hand, if this condition is not met, the communication log up to the immediately preceding communication log is set as “stay”, and the communication log is sequentially processed starting from the new communication log.

  In addition to the technology for determining stay / movement as described above, there is also a technology for estimating a moving object (walking, bicycle, bus, train, etc.) that the user is likely to board from communication logs (for example, patents). Reference 3). According to this technology, as a communication carrier side, the movement probability, movement distance and movement time calculated from the transition of the connected base station are compared with the average speed and standard deviation for each movement object, and the movement object is presume.

  There is also a technique for estimating a user's moving object by acquiring positions at intervals of X seconds and calculating average speed, maximum speed, and maximum acceleration every Y (X <Y) seconds (for example, Patent Document 4). reference). According to this technique, stay / movement is determined from the average speed, and the bicycle or high-speed vehicle is used from the maximum speed. Here, when it is determined that the vehicle is a high-speed vehicle, it is determined from the maximum acceleration whether the vehicle is a passenger car or a railway.

JP 2012-85095 A JP 2014-116808 A JP 2012-272316 A JP 2010-19811 A

S. Isaacman, R. Becker, R. Caceres, SG Kobourov, M. Martonosi, J. Rowland, and A. Varshavsky, "Identifying Important Places in People's Lives from Cellular Network Data", Proc. Of the 9th International Conference on Pervasive Computing, pp.133-151, 2011 MA Bayir, M. Demirbas, and N. Eagle, `` Mobility profiler: A framework for discovering mobility profiles of cell phone users '', Proc. Of the International Conference on Pervasive and Mobile Computing, vol.6, no.4, pp. 435--454, 2010 J. H. Kang, W. Welbourne, B. Stewart, G. Borriello, "Extracting places from traces of locations", ACM SIGMOBILE Mobile Computing and Communications Review, vol. 9, no. 3, July 2005 J. Liu, O. Wolfson, H. Yin, "Extracting Semantic Location from Outdoor Positioning Systems", Proc. Of the 7th International Conference on Mobile Data Management (MDM 2006), pp. 73, 2006

According to the techniques described in Non-Patent Document 1 and Patent Document 1, a change in the base station that occurs when moving between the staying locations affects the accuracy of the staying location.
In addition, according to the techniques described in Patent Document 2 and Non-Patent Documents 3 and 4, when the state is not “stay”, only the “moving” state is set, and the moving speed and the moving distance are not considered at all.
Furthermore, according to the technique described in Non-Patent Document 2, since the location information is not taken into consideration, for mobile terminals that do not generate communication during movement, the staying place before moving and the staying place after moving are distinguished. Can not do it. In addition, when base stations are densely arranged in a narrow range, the staying time is excessively divided.
Furthermore, according to the technique described in Patent Document 2, when there is location information of a plurality of base stations that satisfy the stay condition for each time window, the positioning using all the base stations is performed and the stay is performed. The positions related to are consolidated in one place. As a result, when the mobile terminal moves between a plurality of base stations included in one time window, it may be determined that the user is staying.
Furthermore, according to the techniques described in Patent Documents 2 and 3 and Non-Patent Documents 3 and 4, distance is used as a determination criterion for stay / movement. The longer this distance is, the less likely it is to be determined to be moving (easy to be determined as stay), and the higher accuracy to determine that it is moving (the accuracy of the determination result as stay is lower). Also, the shorter the distance, the easier it is to determine that it is moving, and the lower the accuracy of determining that it is moving. Therefore, in these techniques, the stay / movement is determined after determining one distance at which the stay and movement can be determined in a well-balanced manner.

  Here, according to the techniques described in Patent Documents 3 and 4, in order to estimate the user's moving object, the speed and acceleration calculated according to the distance moved in a predetermined time range are used. It is not determined until “stay / movement” of the user's position. Therefore, even for the same moving object, the moving object that repeats staying / moving in a short time and the moving object that moves for a long time are distinguished and determined. For example, even if it is a moving object “train”, it can be assumed that an “ordinary” train is determined as a moving object “automobile” when attention is paid to the average speed.

  In particular, according to the technique described in Patent Document 4, the speed needs to be as accurate as possible. For this reason, it is difficult to estimate a moving object with a communication log of position information with a coarse spatial granularity and a fixed time interval, such as the position information of a base station.

  For example, it is assumed that a mobile terminal of a user who is on a moving object traveling at a speed of 60 km / h in an area sparsely arranged with a distance between base stations of 2 km communicates once every 30 seconds. The speed measured here is calculated from the base station position and time difference in the preceding and following communication logs. Depending on the surrounding environment between the mobile terminal and the base station, the connected base station also changes, and an accurate speed cannot be calculated.

  Therefore, the present invention provides an apparatus and a program that can estimate a moving object in consideration of the stay / movement of a mobile terminal possessed by a user even if the position information is spatially coarse and the time interval is not constant. And to provide a method.

According to the present invention, an apparatus for estimating a moving object of a user carrying a mobile terminal,
Communication history storage means for storing a plurality of communication logs in which time and position information are associated with each portable terminal;
For each communication log, for a predetermined fixed time window including the communication log, first movement speed calculation means for calculating a first movement speed based on position information and elapsed time;
A second moving speed calculating means for calculating a second moving speed based on the position information and the elapsed time for a variable time window in which the past time is set longer than the fixed time window by a predetermined condition for each communication log;
A moving object reference table in which a first moving speed range and a second moving speed range are associated with each moving object;
And a moving object estimation means for selecting a moving object including both the first moving speed and the second moving speed using a moving object reference table.

According to another embodiment of the device of the invention,
In the moving object reference table, for each moving object, a speed difference range or a ratio range between the first moving speed and the second moving speed is associated,
If the moving object cannot be selected because either one of the first moving speed and the second moving speed is not included, the moving object estimation means uses the moving object reference table to determine the first moving speed. It is also preferable to select a moving object including a speed difference range or a ratio range between the second moving speed and the second moving speed.

According to another embodiment of the device of the invention,
The fixed time window includes from the communication log of a predetermined time to the communication log of the determination target time, centering on the communication log of the determination target time,
The variable time window preferably includes at least the communication log of the latest time in the fixed time window.

According to another embodiment of the device of the invention,
The location information is the location information of the base station accessed by the mobile device,
The communication log is also preferably time and position information collected by the communication equipment device every time the mobile terminal accesses the wide area network.

According to another embodiment of the device of the invention,
It is also preferable to further include estimation timing control means for controlling the moving object estimation means to function when position information of a plurality of base stations is included in the variable time window.

According to another embodiment of the device of the invention,
The location information is measured by the positioning function installed in the mobile device.
The communication log is also preferably time and position information related to communication received from the mobile terminal.

According to another embodiment of the device of the invention,
The second moving speed calculation means, for each communication log,
Calculate the stay movement ratio by dividing the number of communication logs included in the variable time window by the number of logs including position information within a predetermined distance from the position information at the current time,
When the stay movement ratio is equal to or greater than the predetermined ratio, it is determined as “stay”, and the next time fluctuation time window is set to be long so as to include the current time fluctuation time window,
In other cases, it is also preferable to determine “move” and set the fluctuation time window of the next time as a fixed time window.

According to another embodiment of the device of the invention,
The first moving speed calculating means and the second moving speed calculating means select the first position information and the second position information that are the longest distance in the fixed time window or the variable time window,
The distance between the first position information and the second position information is the time or time between the first time when the first position information is detected and the second time when the second position information is detected. It is also preferable to calculate the moving speed by dividing by the window range.

According to another embodiment of the device of the invention,
The first moving speed calculating means and the second moving speed calculating means determine the distance between the position information of the oldest time and the position information of the latest time in the fixed time window or the variable time window within the range of the time window. It is also preferable to calculate the moving speed by dividing.

According to the present invention, there is provided a program for causing a computer mounted on an apparatus for estimating a moving object of a user possessing a mobile terminal to function,
Communication history storage means for storing a plurality of communication logs in which time and position information are associated with each portable terminal;
For each communication log, for a predetermined fixed time window including the communication log, first movement speed calculation means for calculating a first movement speed based on position information and elapsed time;
A second moving speed calculating means for calculating a second moving speed based on the position information and the elapsed time for a variable time window in which the past time is set longer than the fixed time window by a predetermined condition for each communication log;
A moving object reference table in which a first moving speed range and a second moving speed range are associated with each moving object;
The computer is caused to function as a moving object estimation unit that selects a moving object including both the first moving speed and the second moving speed using the moving object reference table.

According to the present invention, there is provided a moving object estimation method for an apparatus for estimating a moving object of a user who has a portable terminal,
The device is
For each portable terminal, a communication history storage unit that stores a plurality of communication logs in which time and position information are associated,
A moving object reference table in which a first moving speed range and a second moving speed range are associated with each moving object;
The device is
For each communication log, for a predetermined fixed time window including the communication log, the first moving speed is calculated based on the position information and the elapsed time, and the past time is set longer than the fixed time window by a predetermined condition. A first step of calculating a second moving speed based on position information and elapsed time for a time window;
And a second step of selecting a moving object including both the first moving speed and the second moving speed using the moving object reference table.

  According to the apparatus, the program, and the method of the present invention, the moving object is estimated in consideration of the stay / movement of the mobile terminal possessed by the user even if the positional information has a coarse spatial granularity and the time interval is not constant. be able to.

It is explanatory drawing showing stay / movement of the moving target object which the user who possessed the portable terminal has boarded. It is a functional block diagram of the communication equipment apparatus in this invention. It is explanatory drawing showing the communication log of the portable terminal accumulate | stored in the communication equipment apparatus. It is explanatory drawing showing the communication log accumulate | stored in the communication log | history storage part which merged the base station position information. It is explanatory drawing showing the fixed time window of a 1st moving speed calculation part. It is explanatory drawing showing transition of a fixed time window. It is explanatory drawing showing the fluctuation | variation time window of a 2nd moving speed calculation part. It is explanatory drawing showing transition of a fluctuation | variation time window. It is explanatory drawing showing the movement target object reference table in this invention. It is explanatory drawing which controlled the moving target object estimation part by the estimation timing control part.

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

  FIG. 1 is an explanatory diagram showing stay / movement of a moving object on which a user carrying a mobile terminal is boarded.

  According to FIG. 1, the user who has the portable terminal 2 gets on / off the moving object such as a train, a bus, an automobile, and a bicycle, including walking, and repeats staying / moving. The mobile terminal 2 is a smartphone, a tablet, a mobile phone, or a wearable device. According to FIG. 1, the user who possesses the mobile terminal 2 enters a “stay” state at home, A station, B station and work place, and “moves” between these “stay” states by boarding a moving object. Is connected in the state. Note that the mobile terminal 2 can communicate with any of the base stations 3 at any position.

  Each base station 3 arranged in the wide area wireless communication network (mobile phone network) acquires a communication log for each mobile terminal 2 under its control. The communication log is a history of the time (date / time) at which communication occurred. The “time” is recorded by various communications such as a call, mail transmission / reception, access to a Web site, communication between an installed application and a server, and application / content upload / download. The wide area wireless communication network is operated by a communication carrier, and is a wireless access network such as LTE (Long Term Evolution), WiMAX (Worldwide Interoperability for Microwave Access), or 3G (3rd Generation). These communication logs are collected in the communication facility apparatus 1 that is operated and managed by the communication carrier.

  FIG. 2 is a functional configuration diagram of the communication equipment apparatus according to the present invention.

  The communication equipment device 1 of the present invention can estimate the moving object of the user who has the mobile terminal 2. Of course, the apparatus for estimating a moving object in the present invention is not required to be the communication equipment apparatus 1 of the wide area wireless communication network, and may be any apparatus that accumulates at least time-series communication logs.

  2, the communication history storage unit 101, the first movement speed calculation unit 111, the second movement speed calculation unit 112, the movement object reference table 102, and the movement object estimation. Unit 12 and an estimation timing control unit 13. These functional components are realized by executing a program that causes a computer mounted on the communication equipment to function. Further, the processing flow of these functional components can also be understood as a moving object estimation method of the apparatus.

[Communication history storage unit 101]
The communication history accumulation unit 101 accumulates a plurality of communication logs in which time and position information are associated with each portable terminal.

  FIG. 3 is an explanatory diagram illustrating a communication log of the mobile terminal accumulated in the communication facility device.

The communication log is obtained by arranging time and base station ID in time series for each terminal ID (IDentifier).
Communication log [terminal ID, time, base station ID]
The “time” may be at the start or disconnection of communication. The communication log may further be associated with radio quality information such as radio wave signal strength (RSSI) and round trip delay time (RTD) between the terminal and the base station.

The base station ID may be a latitude / longitude based on the base station ID, or a latitude / longitude acquired by a positioning function (for example, GPS) of the terminal.
<Latitude and longitude based on base station ID>
The position information is the latitude and longitude of the base station 3 accessed by the mobile terminal 2. This communication log is recorded by the communication equipment device 1 when the mobile terminal 2 accesses the wide area communication carrier network. Therefore, as described above as the prior art, the mobile terminal does not need to transmit the position information based on its positioning function to the communication equipment device 1.

  The location information set as the base station ID has coarse spatial granularity and the time interval is not constant. Here, “the spatial granularity is coarse” means that the (geographic) distance in the real space between the position information is relatively long. Further, “the time interval is not constant” means that the time interval between the times associated with the position information varies depending on the communication timing.

<Latitude and longitude based on the positioning function (GPS) of the mobile terminal>
The position information is measured by a positioning function installed in the mobile terminal. This communication log needs to be transmitted by the mobile terminal 2 itself to the communication equipment device 1.

  According to FIG. 3, the terminal a is moving while handing over from the base station 001 to 003. Here, it is assumed that the distance between the base stations 001 <-> 002 and the distance between the base stations 002 <-> 003 are 2 km. At this time, when the terminal a is handed over from the base station 001 to 002, the moving speed can be calculated from the relationship between the distance between the base stations 001 <-> 002 and the time of the communication log.

The following communication log is recorded for the terminal a.
Log 1 [terminal a, 12:10:00, base station 001]
Log 2 [terminal a, 12:10:30, base station 001]
Here, the terminal a itself is moving, but the communicating base station 001 has not changed. Therefore, the moving speed is calculated as 0km.
Log 3 [terminal a, 12:11:00, base station 001]
Log 4 [terminal a, 12:11:30, base station 002]
Here, due to the movement of the terminal a, the base station 001 is changing from the communicating base station 001 within 30 seconds. Further, the distance between the base station 001 and the base station 002 is 2 km. Therefore, the moving speed is calculated as 240 km / h by the following formula.
2km / h / (30 seconds / 3600 seconds) = 240km / h
According to the communication log of FIG. 3, when the base station changes, it moves at 240 km / h, and when the base station is the same, it is recorded that it stops at 0 km. The moving speed here is averaged over a relatively long time, whereby a moving speed close to the actual condition can be obtained.

  FIG. 4 is an explanatory diagram showing communication logs stored in the communication history storage unit, which are merged with base station position information.

  According to FIG. 4, base station location information is stored separately. In this case, the latitude and longitude of the base station are associated with each base station ID. According to FIG. 4, for example, it can be understood that the position information of the base station 001 is 37.21 degrees latitude and 139.31 longitude. The base station position information may be stored in advance in the communication facility device 1 or may be received from each base station.

When the base station ID is included in the communication log, the latitude and longitude of the base station are merged and recorded in the base station ID. For example, the base station position information of FIG. 4 is associated with the communication log of FIG.
According to FIG. 4, position information is associated as follows.
Log 1 [terminal a, 2014/7/31 20:11, location information 37.21, 139.31]
Log 2 [terminal a, 2014/7/31 20:12, location information 37.22, 139.32]
...

[First moving speed calculation unit 111]
For each communication log, the first moving speed calculation unit 111 calculates a first moving speed based on the position information and the elapsed time for a predetermined fixed time window including the communication log.

FIG. 5 is an explanatory diagram illustrating a fixed time window of the first moving speed calculation unit.
FIG. 6 is an explanatory diagram showing transition of a fixed time window.

The “fixed time window” includes a communication log at a predetermined time to a communication log at a determination target time (determination target communication log) centering on a communication log at a determination target time (current time to be determined). is there.
According to FIG. 6, “predetermined time” and “determination target time” in the fixed time window are set to 3 minutes, for example. In this case, a communication log (determination target communication log) in 3 minutes before and after the determination target time (current time), that is, 6 minutes (determination target time) is set as a fixed time window. In the case of a fixed time window, as shown in FIG. 6, the determination target time progresses and progresses in the same time window (fixed time window) of three minutes before and after.

  According to FIG. 5, for each of the communication logs No. 1 to No. 15, the determination target communication log in the fixed time window is represented by an arrow. The moving speed is calculated for each fixed time window.

  The first moving speed calculation unit 111 calculates the first moving speed for each fixed time window. Here, either the moving speed based on the longest distance or the moving speed based on the longest time is calculated.

<Movement speed based on the longest distance>
In the fixed time window, first position information and second position information that are the longest distance are selected. Then, the distance between the first position information and the second position information is determined between the first time at which the first position information is detected and the second time at which the second position information is detected. Divide by time or time window range to calculate moving speed.
First moving speed = | first position information−second position information |
/ | First time-second time | or | time window range |

(Example: Communication log No. 6 in Fig. 5)
From the viewpoint of the base station 002 at the determination target time, the fixed time window includes the base stations 001 and 004. In this case, the base station 002 <-> 004 having the longest distance is selected. This longest distance is divided by the time or time window range between the first time (20:16) when the base station 002 is detected and the second time (20:18) when the base station 004 is detected. Thus, the first moving speed is calculated. When a plurality of base stations having the longest distance are included, it is also preferable to select a combination having the shortest time.
(Example: Communication log No. 8 in Fig. 5)
From the viewpoint of the base station 004 at the determination target time, the fixed time window includes base stations 001, 002, 005, and 006. In this case, the base station 002 <-> 006 having the longest distance is selected. At this time, the first moving speed is divided by the time between the first time (20:16) when the base station 002 is detected and the second time (20:20) when the base station 006 is detected. Is calculated.

<Movement speed based on the longest time>
The moving speed is calculated by dividing the distance between the position information of the oldest time and the position information of the latest time by the range of the time window.
First moving speed = | Position information of the oldest time−Position information of the latest time |
/ | Oldest time-latest time | or | range of fixed time window |

(Example: Communication log No. 7 in Fig. 5)
From the viewpoint of the base station 001 at the determination target time, the fixed time window includes base stations 002, 004, and 005. In this case, the distance between the position information of the base station 001 at the oldest time and the position information of the base station 005 at the oldest time is expressed as the first time (20:15) when the base station 001 is detected, and the base station The first moving speed is calculated by dividing 005 by the time from the second time (20:19) at which it was detected.
(Example: Communication log No. 8 in Fig. 5)
From the viewpoint of the base station 004 at the determination target time, the fixed time window includes base stations 001, 002, 005, and 006. In this case, the distance between the position information of the base station 002 at the earliest time and the position information of the base station 006 at the latest time is the first time (20:16) when the base station 002 is detected, and the base station The first moving speed is calculated by dividing 006 by the time from the second time (20:20) when the 006 is detected.

[Second moving speed calculation unit 112]
The second moving speed calculation unit 112 calculates a second moving speed based on the position information and the elapsed time for a communication time log in which the past time is set longer than the fixed time window by a predetermined condition for each communication log.

FIG. 7 is an explanatory diagram illustrating a variation time window of the second moving speed calculation unit.
FIG. 8 is an explanatory diagram showing transition of the fluctuation time window.

  The “variable time window” includes at least the communication log of the latest time in the fixed time window. That is, for the determination target communication log, the latest time in the variable time window is the same as the latest time in the fixed time window. On the other hand, the oldest time of the variable time window is set to be the same as the oldest time of the fixed time window or longer than the oldest time of the fixed time window by a predetermined condition in the past.

The second moving speed calculation unit 112 calculates the second moving speed by executing as follows.
<S0: Reset of time window>
<S1: Judgment of stay / movement of variable time window>
<S2: Calculation of Second Movement Speed of Fluctuating Time Window>

<S0: Reset of time window>
The second moving speed calculation unit 112 resets to the same time window as the fixed time window in the first fluctuation time window.

<S1: Judgment of stay / movement of variable time window>
The second moving speed calculation unit 112 calculates a “stay movement ratio” obtained by dividing the number of logs including position information within a predetermined distance from the position information at the current time by the number of communication logs included in the variable time window.
The “predetermined distance” is a distance of position information (for example, base station position information) that can be regarded as a stay, and is set to, for example, 5 km in the following embodiment.
Then, the second movement speed calculation unit 112 determines “stay” when the stay movement ratio is equal to or greater than the predetermined ratio, and the variation time window of the next time includes the variation time window of the current time. Set a longer time.
On the other hand, when the stay movement ratio is lower than the predetermined ratio, it is determined as “movement” and the next time variation time window is set as a fixed time window.

  Here, the predetermined ratio with respect to the stay movement ratio can be set to 1, for example. That is, if it is determined to move even once within the fluctuation time window, it is determined to be “move”. That is, it can be determined that the lower the stay movement ratio, the higher the degree of “movement”. The staying movement ratio derived by the variable time window can determine the staying movement state of the user even if it is a communication log based on position information with a coarse spatial granularity and a time interval that is not constant.

  In addition, when no other communication log is included in the fluctuation time window, it is preferable to determine that the state of the user of the mobile terminal 2 is “stay”.

  Also in FIG. 7, as in FIG. 5, for each of the communication logs No. 1 to No. 15, the determination target communication log in the variable time window is represented by an arrow.

(Example: Communication log No. 3 in Fig. 7)
The variable time window includes determination target communication logs No. 1, No. 2, No. 4, and No. 5. Both the base stations 001 and 002 related to the determination target communication log are located within the determination distance from the position of the base station 003 related to the communication log No. 3.
That is, among the four base stations included in the determination target communication log, four are located within the determination distance.
As a result, for the determination target communication log No. 3, the stay movement ratio is 1 (= 4/4), and the state of the user of the mobile terminal 2 is determined as “stay”.
The next time fluctuation time window is set to have a long past time so as to include the current time fluctuation time window.

(Example: Communication log No. 5 in Fig. 7)
The variable time window includes judgment target communication logs No. 1, No. 2, No. 3, No. 4, No. 6, and No. 7. Both of the base stations 002 and 003 related to the determination target communication log are located within the determination distance from the position of the base station 001 related to the communication log No. 5.
That is, among the positions of the six base stations included in the determination target communication log, six are located within the determination distance.
As a result, for the determination target communication log No. 5, the stay movement ratio is 1 (= 6/6), and the state of the user of the mobile terminal 2 is determined as “stay”.
The next time fluctuation time window is set to have a long past time so as to include the current time fluctuation time window.

(Example: Communication log No. 6 in Fig. 7)
The variable time window includes judgment target communication logs No. 1, No. 2, No. 3, No. 4, No. 5, No. 7, and No. 8. Both the base stations 001 and 003 related to the determination target communication log are located within the determination distance from the position of the base station 002 related to communication log No. 6, but the base station 004 Is outside.
That is, among the positions of the seven base stations included in the determination target communication log, six are located within the determination distance.
As a result, regarding the determination target communication log No. 6, the stay movement ratio is 0.857 (= 6/7), and the state of the user of the mobile terminal 2 is determined as “movement”.
Then, the fluctuation time window of the next time is set to a fixed time window.

(Example: Communication log No. 7 in Fig. 7)
The variable time window includes determination target communication logs No. 5, No. 6, No. 8, and No. 9. The base station 002 related to the determination target communication log is located within the determination distance from the position of the base station 001 related to communication log No. 7, but the base stations 004 and 005 are located outside the determination distance from the position of the base station 001. is there.
That is, two of the positions of the four base stations included in the determination target communication log are located within the determination distance.
As a result, for the determination target communication log No. 7, the stay movement ratio is 0.5 (= 2/4), and the state of the user of the mobile terminal 2 is determined as “movement”.
Then, the fluctuation time window of the next time is set to a fixed time window.

(Example: Communication log No. 11 in Fig. 7)
The variable time window includes determination target communication logs No. 9, No. 10, No. 12, and No. 13. Both of the base stations 005 and 007 related to the determination target communication log are located within the determination distance from the position of the base station 006 related to the communication log No. 11.
That is, among the four base stations included in the determination target communication log, four are located within the determination distance.
As a result, for the determination target communication log No. 11, the stay movement ratio is 1 (= 4/4), and the state of the user of the mobile terminal 2 is determined as “stay”.
The next time fluctuation time window is set to have a long past time so as to include the current time fluctuation time window.

(Example: Communication log No. 12 in Fig. 7)
The variable time window includes determination target communication logs No. 9, No. 10, No. 11, No. 13, and No. 14. The base stations 005 and 007 related to the determination target communication log are located within the determination distance from the position of the base station 006 related to the communication log No. 12.
That is, out of the five base station positions included in the determination target communication log, five are located within the determination distance.
As a result, regarding the determination target communication log No. 12, the stay movement ratio is 1 (= 5/5), and the state of the user of the mobile terminal 2 is determined as “stay”.
The next time fluctuation time window is set to have a long past time so as to include the current time fluctuation time window.

<S2: Calculation of Second Movement Speed of Fluctuating Time Window>
The second moving speed calculation unit 112 calculates the second moving speed for each variation time window. Here, there are a moving speed based on the longest distance and a moving speed based on the longest time.
Note that this is exactly the same as the calculation method of the second moving speed calculation unit 111 described above.

<Movement speed based on the longest distance>
The first position information and the second position information that are the longest distance are selected in the variable time window. Then, the distance between the first position information and the second position information is determined between the first time at which the first position information is detected and the second time at which the second position information is detected. Divide by time to calculate movement speed.
Second moving speed = | first position information−second position information |
/ | First time-second time | or | time window range |

<Movement speed based on the longest time>
The moving speed is calculated by dividing the distance between the position information of the oldest time and the position information of the latest time by the range of the time window.
Second moving speed = | position information of the oldest time−position information of the latest time |
/ | Oldest time-latest time | or range of variable time window

  For the first movement speed calculation unit 111 and the second movement speed calculation unit 112, for example, in the comparison target communication log, the difference (distance) in position information is 1 km, the fixed time window is 3 minutes, and the variable time window Is 5 minutes. In this case, the first moving speed of the fixed time window is calculated to be 20 km or more, and the second moving speed of the variable time window is calculated to be 12 km or more.

[Movement object reference table 102]
The moving object reference table 102 is a table in which a first moving speed range and a second moving speed range are associated with each moving object.
The moving object reference table is associated with a “speed difference range” or “ratio range” between the first moving speed and the second moving speed for each moving object. Also good.
The “speed difference range” is represented by “minimum speed to maximum speed”.
Speed difference = the difference between the minimum value of the first moving speed and the minimum value of the second moving speed, or
The difference between the maximum value of the first moving speed and the maximum value of the second moving speed is expressed by “minimum ratio to maximum ratio”.
Ratio = first moving speed / second moving speed or second moving speed / first moving speed

  FIG. 9 is an explanatory diagram showing a moving object reference table in the present invention.

  According to FIG. 9, it can be understood that there is a difference between the first moving speed of the fixed time window and the second moving speed of the variable time window, for example, even if it is a train, depending on whether it is normal or fast. . That is, for ordinary or rapid trains, even if the difference in moving speed in a short time in a fixed time window is small, the difference in moving speed in a long time in a variable time window is large.

[Movement object estimation unit 12]
The moving object estimation unit 12 uses the moving object reference table 102 to calculate the first movement speed calculated by the first movement speed calculation unit 111 and the first movement speed calculated by the second movement speed calculation unit 112. A moving object including both of the two moving speeds is selected. Here, the moving object reference table 102 may be compared with the average speed or the median speed at the moving speed of each of the fixed time window and the variable time window.

Here, since the fluctuation time window is longer than the fixed time window by the stay time, even if the communication log has the same speed in the fixed time window, if the difference in the speed of the fluctuation time window is large, it is more It can be understood that it was influenced by the stop.
For example, comparing the moving speed of a bus with the moving speed of a private car, the speed difference is small. However, while a private car stays only with a signal, a bus stays at a stop in addition to the signal. For this reason, the difference between the moving speed of the fixed time window and the moving speed of the variable time window is larger for the bus than for the private car.
Therefore, the moving object according to the stay / movement characteristics can be estimated by comparing the moving speed of the fixed time window and the moving speed of the fluctuation time with the moving object reference table.

  Further, the moving object estimation unit 12 may not be able to select a moving object because either one of the first moving speed and the second moving speed is not included. In this case, the moving object reference table 102 is used to select a moving object including the “speed difference range” or “ratio range” between the first moving speed and the second moving speed. Is also preferable.

  For example, if the fixed time window is 26 km / h and the variable time window is 9 km / h, the fixed time window corresponds to “bus”, but the variable time window does not correspond. The speed difference is 17km / h (= 26km / h-9km / h). At this time, referring to the moving object reference table 102, the speed difference of 17 km / h corresponds to a speed difference of 15 km / h to 20 km / h in the moving object “bus”. In this case, the moving object is estimated to be a “bus”.

[Estimated timing controller 13]
The estimation timing control unit 13 controls the moving object estimation unit 12 to function when position information of a plurality of base stations is included in the variation time window.

  FIG. 10 is an explanatory diagram in which the moving object estimation unit is controlled by the estimation timing control unit.

(Example: Communication log No. 3 in Fig. 7)
The variable time window includes determination target communication logs No. 1, No. 2, No. 4, and No. 5. Only the base station 001 is included in the determination target communication log. Therefore, in the communication log No. 3, the moving object is not estimated.
(Example: Communication log No. 4 in Fig. 7)
The variable time window includes judgment target communication logs No. 1, No. 2, No. 3, No. 5, and No. 6. The determination target communication log includes base stations 001 and 002. Therefore, in the communication log No. 4, the moving object is estimated.
The estimation timing control unit 13 can always execute the estimation process of the moving object at a timing when the user's moving object changes while reducing the processing load.

  As described above in detail, according to the apparatus, program, and method of the present invention, the stay / movement of the mobile terminal possessed by the user is taken into account even if the location information has a coarse spatial granularity and the time interval is not constant. Thus, the moving object can be estimated.

  Various changes, modifications, and omissions of the various embodiments of the present invention described above within the scope of the technical idea and the viewpoint 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 any limitation. The present invention is limited only by the claims and the equivalents thereof.

1 Communication equipment (moving object estimation device)
DESCRIPTION OF SYMBOLS 101 Communication history storage part 102 Moving object reference table 111 1st moving speed calculation part 112 2nd moving speed calculation part 12 Moving object estimation part 13 Estimation timing control part 2 Portable terminal 3 Base station

Claims (11)

An apparatus for estimating a moving object of a user having a mobile terminal,
Communication history storage means for storing a plurality of communication logs in which time and position information are associated with each portable terminal;
For each communication log, for a predetermined fixed time window including the communication log, first movement speed calculation means for calculating a first movement speed based on position information and elapsed time;
A second moving speed calculating means for calculating a second moving speed based on position information and elapsed time for a variable time window in which the past time is set longer than the fixed time window by a predetermined condition for each communication log;
A moving object reference table in which a first moving speed range and a second moving speed range are associated with each moving object;
An apparatus comprising: a moving object estimation means for selecting a moving object including both the first moving speed and the second moving speed using the moving object reference table. In the moving object reference table, for each moving object, a speed difference range or a ratio range between the first moving speed and the second moving speed is associated,
When the moving object estimation unit cannot select the moving object because one of the first moving speed and the second moving speed is not included, the moving object estimation unit uses the moving object reference table to The apparatus according to claim 1, wherein a moving object including a range of a speed difference or a range of a ratio between the moving speed and the second moving speed is selected. The fixed time window includes a communication log at a predetermined time from a communication log at a predetermined time centered on a communication log at a determination target time,
The apparatus according to claim 1, wherein the variable time window includes at least a communication log of the latest time in the fixed time window. The location information is location information of a base station accessed by the mobile terminal,
4. The communication log according to any one of claims 1 to 3, wherein the communication log is time and position information collected by a communication equipment device every time the portable terminal accesses a wide area communication carrier network. Equipment. 5. The apparatus according to claim 4, further comprising estimation timing control means for controlling the moving object estimation means to function when position information of a plurality of base stations is included in the variation time window. Equipment. The location information is measured by a positioning function installed in the mobile terminal,
The apparatus according to claim 1, wherein the communication log is time and position information related to communication received from the mobile terminal. The second moving speed calculation means, for each communication log,
Calculate the stay movement ratio by dividing the number of communication logs included in the variable time window by the number of logs including position information within a predetermined distance from the position information at the current time,
When the stay movement ratio is a predetermined ratio or more, it is determined as “stay”, and the next time fluctuation time window is set to have a long past time so as to include the current time fluctuation time window,
7. The apparatus according to claim 1, wherein, in other cases, it is determined as “move”, and the variation time window of the next time is set to a fixed time window. The first moving speed calculating means and the second moving speed calculating means select the first position information and the second position information that are the longest distance in the fixed time window or the variable time window,
The distance between the first position information and the second position information is the time or time between the first time when the first position information is detected and the second time when the second position information is detected. The apparatus according to claim 1, wherein the moving speed is calculated by dividing by a window range. The first moving speed calculating means and the second moving speed calculating means determine the distance between the position information of the oldest time and the position information of the latest time in the fixed time window or the variable time window within the range of the time window. The apparatus according to claim 1, wherein the moving speed is calculated by division. A program for causing a computer mounted on an apparatus for estimating a moving object of a user possessing a mobile terminal to function,
Communication history storage means for storing a plurality of communication logs in which time and position information are associated with each portable terminal;
For each communication log, for a predetermined fixed time window including the communication log, first movement speed calculation means for calculating a first movement speed based on position information and elapsed time;
A second moving speed calculating means for calculating a second moving speed based on position information and elapsed time for a variable time window in which the past time is set longer than the fixed time window by a predetermined condition for each communication log;
A moving object reference table in which a first moving speed range and a second moving speed range are associated with each moving object;
A program that causes a computer to function as moving object estimation means for selecting a moving object that includes both the first moving speed and the second moving speed using the moving object reference table. A moving object estimation method for a device that estimates a moving object of a user who has a portable terminal,
The device is
For each portable terminal, a communication history storage unit that stores a plurality of communication logs in which time and position information are associated,
A moving object reference table in which a first moving speed range and a second moving speed range are associated with each moving object;
The device is
For each communication log, for a predetermined fixed time window including the communication log, the first moving speed is calculated based on the position information and the elapsed time, and the past time is set longer than the fixed time window by a predetermined condition. A first step of calculating a second moving speed based on position information and elapsed time for a time window;
And a second step of selecting a moving object including both the first moving speed and the second moving speed using the moving object reference table. Method.

Images