JP5300779B2 - Area estimation apparatus, area estimation method, and program - Google Patents

Description

  The present invention relates to a technique for specifying the position of a user using a wireless communication device such as a wireless base station.

  Currently, a position detection system has been devised that estimates the position of a user using a name (hereinafter referred to as an area) given to a place as position information instead of coordinates (see, for example, Patent Document 1). The area where the user is located is estimated from the radio wave intensity measured between the radio communication terminal carried by the user and the radio base station arranged according to the area configuration. In such a position detection system, it is easy to realize a service using position information by using an area instead of coordinates as position information.

  Further, Non-Patent Document 1 discloses a method for constructing an area estimation model using a learning vector quantization method based on the radio wave intensity from a radio base station for each area in order to estimate the area where the user is located. Has been.

JP 2006-308361 A

Ogawa, Yoshino, Shimizu, “Examination of Location Detection Method Using Learning Function”, IEICE Technical Report, Network System IEICE technical report 102 (202) pp.13-18 20020711

In the above prior art, it is not necessary to calculate distances and coordinates assuming a radio wave propagation environment, and therefore, it is considered that the influence of multipath due to indoor structures or shielding objects can be considered to some extent. However, in supermarkets, shopping centers, and the like, there are people coming and going, and the propagation environment is not constant, and when people come and go, it is considered that the user's area estimation accuracy may be adversely affected.
The present invention has been made in view of the above points, and an object of the present invention is to provide an area estimation technique that enables a user to perform area estimation with high accuracy even in a place where people come and go.

  In order to solve the above problems, the present invention provides an area estimation for estimating an area where a radio communication terminal held by a user is located at a place where a plurality of radio base stations are arranged and a plurality of areas are configured. An area estimation model storage unit that stores a representative vector, which is a radio wave intensity vector representing an area, for each area of the plurality of areas, and a predetermined value between each wireless base station and the wireless communication terminal. Radio wave intensity measured at a time interval is acquired via a communication network, and stored in a storage means as a radio wave intensity vector, and a plurality of radio wave intensity vectors measured at the predetermined time interval, A measurement vector combining means for obtaining a measurement vector in which the maximum value of a plurality of elements at the same element position is arranged at each element position; and the measurement vector Dimension reduction means for excluding the element from the measurement vector when there is an element that is less than the minimum value of the corresponding element in the representative vector stored in the area estimation model storage means, and the area Referring to the estimated model storage means, obtain a representative vector that maximizes the degree of coincidence with the measurement vector processed by the dimension reduction means, and set the area corresponding to the representative vector as the area where the wireless communication terminal is located An area estimation apparatus having an output area estimation output means can be configured.

  The place may be a store, and the wireless base station may be arranged for each item in the store.

  Further, the area estimation output means may obtain the degree of coincidence by calculating a distance between the representative vector and the measurement vector.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the area estimation technique which made it possible to perform a user's area estimation with high precision also in the place where there is traffic.

1 is an overall configuration diagram of an area estimation system in an embodiment of the present invention. It is a flowchart which shows the process outline | summary in embodiment of this invention. It is a functional block diagram of an area estimation apparatus. It is a flowchart for demonstrating the construction method of an area estimation model. It is a figure which shows the example of arrangement | positioning and area structure of a wireless base station. It is a figure which shows the example of the data of an area estimation model. It is a flowchart for demonstrating operation | movement of an area estimation apparatus. It is a figure which shows the result of having performed area estimation using each field intensity vector.

  Embodiments of the present invention will be described below with reference to the drawings.

(Overview)
First, a system configuration and an overview of processing according to the embodiment of the present invention will be described. FIG. 1 shows an overall configuration diagram of an area estimation system in the present embodiment.

  As shown in FIG. 1, the area estimation system according to the present embodiment includes a plurality of radio base stations 1, a radio communication terminal 2 carried by a user who is an area estimation target, and an area estimation that is an apparatus according to the present invention. Device 10. In the present embodiment, each radio base station 1 and the area estimation device 10 are connected via the communication network 3. Further, FIG. 1 shows an area estimation model construction apparatus 20 that executes an area estimation model construction process.

  The wireless base station 1 is a wireless communication device that is finely arranged for each product or product type (hereinafter referred to as an item) in, for example, a supermarket or a shopping center (hereinafter referred to as a store). It is a device that transmits and receives signals to and from the communication terminal 2 wirelessly.

  The wireless communication terminal 2 is a device such as a mobile phone that can communicate with a device connected to the communication network 3 via the wireless base station 1. In the present embodiment, a wireless tag or the like may be used as the wireless communication terminal 2.

  The radio communication terminal 2 receives a signal (radio wave) including the ID of the radio base station 1 from the radio base station 1, measures the intensity of the received radio wave, and uses the measured radio wave intensity as its own ID and radio base station. It has a function of transmitting to the area estimation apparatus 10 together with the ID of 1.

  In the present embodiment, the radio communication terminal 2 measures the radio wave intensity in estimating the user area, but the radio base station 1 may measure the radio wave intensity. In this case, the radio base station 1 receives a signal including the ID of the radio communication terminal 2 from the radio communication terminal 2, measures the intensity of the received radio wave, and uses the measured radio wave intensity as its own ID and radio communication terminal. It transmits to area estimation apparatus 10 with 2 ID.

  The area estimation apparatus 10 includes an area estimation model storage unit 11 that stores area estimation model data. Based on the radio wave intensity measured between the radio base station 1 and the radio communication terminal 2, the area estimation model data is stored. By using it, it has a function of performing area estimation.

  FIG. 2 shows an outline of processing in the present embodiment. As shown in FIG. 2, the processing in this embodiment is roughly divided into an area estimation model construction stage (step 100) and an area estimation stage for estimating the area where the user is located (step 200). There is.

  At the stage of area estimation model construction in step 100, a plurality of radio base stations 1 are arranged as described above, and an area is configured so that it is possible to guess which item the user is interested in. Then, the radio field intensity from each radio base station 1 is measured for each area, and an area estimation model is constructed by the method disclosed in Non-Patent Document 1. And the data of this area estimation model are stored in the area estimation model storage part 11 in the area estimation apparatus 10 shown in FIG.

  In the area estimation model construction stage, radio wave intensity may be measured using an arbitrary measuring device. Furthermore, the area estimation model calculation process may be performed using an apparatus such as an arbitrary computer. As an example, an area estimation model construction apparatus 20 is shown in FIG. Of course, the area estimation apparatus 10 may have an area estimation model construction function.

  In the area estimation stage of step 200 for estimating the area in the store where the user is located, the area estimation device 10 acquires the radio field strength between the radio communication terminal 2 carried by the user and each radio base station 1. Then, referring to the data of the area estimation model stored in the area estimation model storage unit 11, the area where the user is located is estimated.

  The data of the area estimation model is composed of representative vectors representing the area calculated from the radio wave intensity vector measured for each area, and the area estimation apparatus 10 An area having a radio field intensity vector with a high degree of coincidence is estimated as an area where the user is located.

  Here, the radio field intensity vector is a vector whose element is a radio field intensity value between the radio base station 1 arranged according to the item and the radio communication terminal 2 carried by the user. It is a vector with the number of dimensions.

  In the present embodiment, the radio field intensity between the radio base station 1 and the radio communication terminal 2 is measured a plurality of times at short time intervals, and the propagation in the best possible state from the values measured a plurality of times. The values in the environment are selected and used. In other words, a vector obtained by arranging (combining) the maximum value of the radio field intensity with the radio base station for each radio base station among the radio base stations is measured as the radio field intensity vector. This makes it possible to correct the radio field intensity by shifting the measurement time with respect to the propagation path of the radio wave disturbed by traffic.

  Further, in the present embodiment, area estimation is performed by excluding the vector axis of the radio base station that is below the minimum value of the radio field strength for each radio base station in the area estimation model in the measured radio field strength for each radio base station. With reference to the model, the area where the user is located is estimated from the degree of coincidence of the radio wave intensity vectors. The reason for performing such estimation is that the propagation path of radio waves is disturbed by the traffic of people between the radio communication terminal and the radio base station that are below the minimum radio field intensity in the area estimation model, and accurate area estimation is performed. It is because it is thought that it is obstructing. Hereinafter, embodiments of the present invention will be described in more detail.

(Device configuration)
FIG. 3 shows a functional configuration diagram of the area estimation apparatus 10 according to the embodiment of the present invention. As shown in FIG. 3, the area estimation apparatus 10 includes an area estimation model storage unit 11, a radio wave intensity acquisition unit 12, a measurement vector synthesis unit 13, a dimension reduction unit 14, an intervector distance calculation unit 15, and an area estimation result output unit 16. Have The processing content executed by each functional unit will be described in the area estimation processing description to be described later.

  The area estimation apparatus 10 can be realized by mounting a program for performing processing corresponding to each functional unit on a general computer including a storage unit such as a memory and a hard disk and a CPU. The program may be installed in the computer from a recording medium such as a portable memory or a disk, or may be downloaded from a server on a network to the computer and installed.

(About area estimation model construction)
Next, the construction method of the area estimation model will be described with reference to the flowchart of FIG. Note that the area estimation model construction method itself is an existing technology. Refer to Non-Patent Document 1 for details of the learning vector quantization method and the like used there.

First, in step 101 of FIG. 4, the wireless base station 1 is arranged according to the items in the store, and an area is configured so that it is possible to estimate which item the user is interested in.
FIG. 5 shows an example in which the wireless base station 1 is arranged in a part of a store and an area is configured. In FIG. 5, Pi (i = 1... 8) represents the radio base station 1. Aj (j = 1... 25) represents an area.
By adopting such an area configuration, for example, when the area estimation device 10 detects that a certain user is located in the area A20 for a certain period of time, the user is interested in the item facing the area A20. Can be estimated.
Further, for example, when the area estimation device 10 detects that a certain user is located in the area A13 and detects that the user is not located in the area A20 for a certain period of time, the user is directed to an item facing the area A20. It can be estimated that he is not interested.

In step 102 of FIG. 4, the area estimation model construction device 20 acquires the radio wave intensity measured for each area between each wireless base station 1 and the measurement point of the area. Radio wave intensity V _Aj at the measurement point with a certain area Aj is as follows, the radio wave intensity l _Pi at the measurement point from the wireless base station 1 (Pi particular radio base station) is represented as a vector of.
V _Aj = (l _P1 ,…, l _Pi ,…, l _Pn ) (n is the number of radio base stations)
Subsequently, in Step 103 of FIG. 4, the area estimation model construction device 20 constructs an area estimation model.

The area estimation model construction device 20 calculates a representative vector _WAj representing each area by a learning vector quantization method using the measured radio wave intensity vector for each area. A combination of an area and a representative vector representing the area becomes data of the area estimation model. An example of data of an area estimation model constructed based on the area configuration of FIG. 5 is shown in FIG. As shown in FIG. 6, for example, the representative vector WA20 of the area _A20 is as follows.
W _A20 = (l _P1 , l _P2 , l _P3 , l _P4 , l _P5 , l _P6 , l _P7 , l _P8 )
= (-44.125, -41.418, -42.711, -52.150, -42.832, -40.125 , -41.418, -46.711)
Here, the radio wave intensity attenuates according to the distance. Therefore, the radio field intensity from the radio base station P6 closest to the area A20 is the highest, and its value is l _P6 = −40.125 (dB). The radio field intensity from the radio base station P4 farthest from the area A20 is l _P4 = −52.150 (dB), and the value is the smallest among the radio field intensity receivable from the area A20.

  The area estimation model data obtained as described above is stored in the area estimation model storage unit 11 of the area estimation apparatus 10 (step 104).

(Area estimation processing)
Next, area estimation processing executed by the area estimation apparatus 10 will be described with reference to the flowchart of FIG. In the following, description will be given focusing on one wireless communication terminal, but it goes without saying that when there are a plurality of wireless communication terminals, the same operation is performed in each.

  In the present embodiment, the radio field intensity between the radio base station 1 and the radio communication terminal 2 is measured at a short time interval T. That is, for example, each wireless base station 1 transmits a signal including its own ID at intervals of T, and the wireless communication terminal 2 measures the radio wave intensity when receiving the signal. The wireless communication terminal 2 transmits its own ID, the ID of the wireless base station, and the measurement result (radio wave intensity) to the area estimation device 10 every time it measures. For example, the wireless communication terminal 2 transmits a signal including its own ID at intervals of T, and each wireless base station 1 measures the radio field intensity when receiving the signal. Each radio base station 1 transmits its own ID, the ID of the radio communication terminal 2 and the measurement result (radio wave intensity) to the area estimation device 10 every time it measures.

  By performing such radio field intensity measurement, in step 201 of FIG. 7, the radio field intensity acquisition unit 12 of the area estimation apparatus 10 communicates with the radio base station 1 at a time interval T for each radio base station. The radio field intensity with the terminal 2 is acquired via the communication network 3. The radio wave intensity data acquired by the radio wave intensity acquisition unit 12 is stored in a storage unit such as a memory as a radio wave intensity vector.

  For example, assume that users X, Y, and Z are located in areas A20, A5, and A13 in FIG. 5 at a certain timing (t = 1), respectively. At the next timing (t = 2), users X, Y, and Z are located in areas A20, A6, and A13, respectively, and at the next timing (t = 3), they are located in areas A20, A6, and A5. And At this time, the radio wave intensity vector V measured for the wireless communication terminal 1 carried by the user X at each timing is as follows.

V (X, t = 1) = (-43.243, -50.696, -68.338, -68.473, -42.404, -40.926, -40.956, -47.650)
V (X, t = 2) = (-44.112, -43.489, -73.335, -69.842, -43.256, -40.125, -41.815, -47.178)
V (X, t = 3) = (-44.548, -50.701, -70.402, -53.715, -43.684, -40.932, -40.590, -48.603)
Here, if the users Y and Z do not exist, the radio wave intensity vector measured at each timing is considered to show a value close to the representative vector _WA20 in FIG. However, since the users Y and Z are considered to disturb the radio wave propagation path from the radio base stations P2, P3, and P4 to the radio communication terminal 1 carried by the user X, the propagation intensity from the radio base station 1 is determined at each timing. The value weakened according to the position is measured by the users Y and Z at.

  Next, in step 202, the measurement vector combining unit 13 of the area estimation apparatus 10 receives radio waves between the radio base stations 1 based on a plurality of radio wave intensity vectors obtained at predetermined time intervals by the radio wave intensity acquisition unit 12. A measurement vector synthesis process for obtaining a radio wave intensity vector having the maximum value of the intensity as an element is performed. That is, the measurement vector combining unit 13 obtains a measurement vector in which a maximum value among a plurality of elements at the same element position is arranged at each element position among a plurality of radio wave intensity vectors measured at a predetermined time interval.

  A predetermined value can be used as the number of times of measurement used for measurement vector synthesis. In the following, as an example, a case where three measurement results are used in measurement vector synthesis is shown.

  For example, when the radio wave intensity vector V (X, t = 1,2,3) is measured as described above, the measurement vector combining unit 13 includes a plurality of measurement results at each position (axis) of the vector. Is obtained, and the maximum value is used as an element of the corresponding axis of the vector. Thereby, the vector shown below is obtained, and this is set as a measured radio wave intensity vector V (X) (referred to as a measurement vector) and stored in storage means such as a memory.

V (X) = (-43.243, -43.489, -68.338, -53.715, -42.404, -40.125, -40.590, -47.178)
Subsequently, in step 203 in FIG. 7, the dimension reduction unit 14 of the area estimation apparatus 10 refers to the area estimation model storage unit 11 and determines the radio field strength of each radio base station in the measurement vector obtained by the measurement vector synthesis process. Among them, the radio field intensity from the radio base station below the minimum value of the radio field intensity l _Pi (Pi is a specific radio base station) in the area estimation model is excluded from the measurement vector.

For example, if the measurement vector V (X) is synthesized as described above, the value from the radio base station P3 in V (X) (-68.338) is the minimum value of l _P3 area estimation model from (-53) Since it is low, the dimension of l _{P3 in} the measurement vector is not used for area estimation. This is expressed as follows.

V '(X) = (-43.243, -43.489, NA, -53.715, -42.404, -40.125, -40.590, -47.178)
The dimension reduction unit 14 stores the data of the measurement vector V ′ (X) indicating that the dimension of l _P3 is NA in a storage unit such as a memory.

Subsequently, in step 204 of FIG. 7, the inter-vector distance calculation unit 15 of the area estimation apparatus 10 refers to the area estimation model data stored in the area estimation model storage unit 11, and the dimension reduction unit 14 performs dimension reduction. A representative vector that maximizes the degree of coincidence between the measured vector V ′ (X) and the representative vector W _Aj (Aj is a specific area) is obtained. The inter-vector distance calculation unit 15 calculates the distance between the measurement vector V ′ (X) and the representative vector _WAj based on the following formula as the degree of coincidence, and obtains the representative vector that maximizes the calculated distance. .
| W _Aj -V '(X) |
For example, when the dimension-reduced measurement vector V ′ (X) is as described above, the distance to the representative vector W _A20 is 2.933, and W _A20 has the highest degree of coincidence with V ′ (X). The inter-distance calculation unit 15 outputs the representative vector W _A20 (or area identification information corresponding to W _A20 ).

Subsequently, in step 205, the area estimation result output unit 16 receives from the inter-vector distance calculation unit 15 the representative vector W _Aj having the closest inter-vector distance to the measurement vector V ′ (X), and receives the representative vector W _Aj. Is estimated as the area where the user is located, and identification information indicating the area Aj is output.

As described above, in this example (measurement vector V ′ (X) is the above value), representative vector _WA20 has the highest degree of coincidence with V ′ (X), and area A20 is estimated as the area where the user is located. Information indicating the area A20 is output.

  In addition, referring to the area estimation model data for each measured or synthesized radio wave intensity vector, calculating the distance between the vectors, and estimating the position of the user from the degree of coincidence results in the result shown in FIG. As shown in FIG. 8, it is possible to estimate with high accuracy by V ′ (X) obtained by the embodiment of the present invention.

(Summary of the embodiment, effects)
In the area estimation system described in the present embodiment, as shown in FIG. 5, a service using a user's position in a store or the like is realized by arranging wireless base stations finely according to items and configuring an area. Making it easy.

  In addition, by precisely arranging the radio base stations, it is possible to detect the position of a fine user, and it is possible to provide an area estimation method that is robust against traffic by using a large number of radio base stations.

  In the present embodiment, the area estimation apparatus 10 performs the measurement vector combining process (step 202 in FIG. 7) to combine the element having the maximum value from the radio wave intensity vectors measured a plurality of times. The measurement vector is to be obtained. By this processing, even if a specific radio wave propagation path is disturbed by traffic at a certain timing, the radio wave intensity of the radio wave propagation path can be measured at a different timing and reflected in the radio wave intensity vector. Therefore, disturbance of the radio wave propagation path is corrected, and the accuracy of area estimation can be improved. Thereby, an area estimation method that is robust against traffic of people can be provided.

  Furthermore, in the present embodiment, the area estimation apparatus 10 performs dimension reduction processing (step 203 in FIG. 7), so that each radio field intensity in the area estimation model among the measured radio field intensity for each radio base station. It is possible to use radio field intensities from other radio base stations, except for radio field intensities from radio base stations below the minimum value. This makes it possible to eliminate the adverse effects of using the radio field intensity measured from a radio base station where the propagation path of radio waves due to human traffic has been disturbed, and an area estimation method that is robust against human traffic. It becomes possible to provide.

  The present invention is not limited to the above-described embodiments, and various modifications and applications are possible within the scope of the claims.

DESCRIPTION OF SYMBOLS 1 Radio base station 2 Wireless communication terminal 3 Communication network 10 Area estimation apparatus 11 Area estimation model storage part 12 Electric wave intensity acquisition part 13 Measurement vector composition part 14 Dimension reduction part 15 Vector distance calculation part 16 Area estimation result output part 20 Area estimation Model building device

Claims (7)

An area estimation device for estimating an area where a wireless communication terminal held by a user is located at a place where a plurality of wireless base stations are arranged and a plurality of areas are configured,
Area estimation model storage means for storing a representative vector, which is a radio wave intensity vector representing an area, for each area of the plurality of areas;
Radio wave intensity acquisition means for acquiring radio field intensity measured at predetermined time intervals between each radio base station and the radio communication terminal via a communication network, and storing the radio wave intensity vector in a storage means as a radio wave intensity vector;
Measurement vector combining means for obtaining a measurement vector in which the maximum value of a plurality of elements at the same element position is arranged at each element position in the plurality of radio wave intensity vectors measured at the predetermined time interval;
Of the elements included in the measurement vector, when there is an element that is less than the minimum value of the corresponding element in the representative vector stored in the area estimation model storage unit, the dimension reduction unit excludes the element from the measurement vector When,
Referring to the area estimation model storage means, obtain a representative vector that maximizes the degree of coincidence with the measurement vector processed by the dimension reduction means, and the wireless communication terminal locates the area corresponding to the representative vector An area estimation apparatus comprising: area estimation output means for outputting as an area.   The area estimation apparatus according to claim 1, wherein the place is a store, and the wireless base station is arranged for each item in the store.   The area estimation apparatus according to claim 1, wherein the area estimation output unit calculates the degree of coincidence by calculating a distance between the representative vector and the measurement vector. An area estimation method executed by an area estimation apparatus for estimating an area where a wireless communication terminal held by a user is located at a place where a plurality of wireless base stations are arranged and a plurality of areas are configured,
The area estimation device includes an area estimation model storage unit that stores a representative vector, which is a radio wave intensity vector representing an area, for each area of the plurality of areas,
A radio wave intensity acquisition step of acquiring radio field intensity measured at a predetermined time interval between each radio base station and the radio communication terminal via a communication network and storing it in a storage means as a radio field intensity vector;
A measurement vector combining step for obtaining a measurement vector in which a maximum value of a plurality of elements at the same element position is arranged at each element position in the plurality of radio wave intensity vectors measured at the predetermined time interval;
The dimension reduction step of excluding the element from the measurement vector when there is an element that is less than the minimum value of the corresponding element in the representative vector stored in the area estimation model storage unit among the elements included in the measurement vector When,
Referring to the area estimation model storage means, obtain a representative vector that maximizes the degree of coincidence with the measurement vector processed by the dimension reduction means, and the wireless communication terminal locates the area corresponding to the representative vector An area estimation output step comprising: an area estimation output step for outputting as an area.   The area estimation method according to claim 4, wherein the place is a store, and the wireless base station is arranged for each item in the store.   The area according to claim 4 or 5, wherein, in the area estimation output step, the area estimation device calculates the degree of coincidence by calculating a distance between the representative vector and the measurement vector. Estimation method.   The program for functioning a computer as each means of the area estimation apparatus of any one of Claims 1 thru | or 3.

Images