JP2013015418A - Position estimating server, position estimating method, and position estimating program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a position estimating server, a position estimating method, and a position estimating program that make possible simple and yet accurate estimation of the position of an access point by using information of uneven accuracy, collected from a terminal mounted with no special device.SOLUTION: A position estimating server 100 for estimating the position of an access point 30 of wireless LAN comprises a database that holds reported data reported from a terminal 20 in association with a specific access point 30, including information on the position of the terminal 20 and on a propagation route, and a position estimator that estimates the specific access point 30 with reference to the database, with priority given to data more accurate regarding the terminal position and less in propagation route variation out of the reported data. This makes possible simple and yet accurate estimation of the position of the access point 30 by using information of uneven accuracy obtained from the terminal 20 mounted with no special device.

Description

  The present invention relates to a position estimation server, a position estimation method, and a position estimation program that estimate the position of a wireless LAN access point using data collected from a terminal.

  In recent years, wireless LANs have been used due to their convenience and ease of installation of access points (APs) against the backdrop of widespread use of mobile terminals and wireless LANs in homes, offices, public facilities, restaurants, etc. Attention has been focused on location estimation and services using the obtained location information.

  Along with this, an enormous number of APs are being placed in various places, and by accurately grasping AP position information, (1) high-precision user position estimation (2) grasping the occurrence of wireless LAN interference It can be expected to help. However, on the other hand, wireless LAN APs can be freely installed and removed by users. Unlike mobile phone base stations and other wireless broadband base stations, operators can accurately locate all AP locations. Can't figure out.

  In such an environment, the Locky.jp project (see Non-Patent Document 1) is known as a system for grasping the AP position of a wireless LAN. Locky.jp records the BSSID (Basic Service Set Identifier) and RSSI (Received Signal Strength Indicator) of the AP received by the PC on which the software called “Locky Stumbler” is installed. To report to. Then, the server estimates the position of the AP based on such information.

  On the other hand, a technique for estimating the position with high accuracy using a terminal equipped with a special device has also been proposed. According to Non-Patent Document 2, the position information of the AP is regarded as a probability density function, and by using the Particle Filter based on the information collected from the user terminal equipped with the directional antenna, more accurate position estimation is realized. ing.

“Locky.jp project”, [on line], test release on December 16, 2005, [search June 20, 2011], Internet <URL: http://locky.jp/> Hiroyuki Suzuki, Seigo Ito, Nobuo Kawaguchi, “Position Estimation Method for Multiple Wireless LAN Base Stations Using Particle Filter”, Information Processing Society of Japan, Research Report, November 16, 2006

  However, in order to perform AP position estimation using the above-described method, since the user's moving direction information affects the accuracy of final position estimation, the user terminal can use an optical sensor or a directional antenna in addition to GPS. Must be installed. Although GPS devices and the like are already widely installed in mobile phones and the like, devices equipped with optical sensors and directional antennas are rare at present, and it is difficult to estimate the position of an AP with a general terminal. .

  The present invention has been made in view of such circumstances, and a position that can estimate the position of an access point simply and with high accuracy using information collected from a terminal that does not have a special device and having a variation in accuracy. An object is to provide an estimation server, a position estimation method, and a position estimation program.

  (1) In order to achieve the above object, a position estimation server of the present invention is a position estimation server for estimating the position of a wireless LAN access point, includes information on the position of a terminal and a propagation path, and has a specific access A database holding report data reported by a terminal in association with a point, referring to the database, and focusing on data with high terminal position accuracy and small fluctuations in propagation path conditions among the report data, A position estimation unit that estimates the position of the specific access point.

  As described above, the position estimation server of the present invention processes information to estimate the position of the access point with reference to accuracy and reliability. As a result, the position of the access point can be estimated simply and with high accuracy using position information that varies in accuracy by a terminal not equipped with a special device.

  (2) In the position estimation server of the present invention, the position estimation unit is configured to propagate the report data associated with the specific access point based on the fact that the received radio wave intensity is not within the abnormal value range. The position of the specific access point is estimated with emphasis on data with small fluctuations in road conditions. Thereby, since the position of the access point is estimated with emphasis on the report data whose received radio wave intensity is not within the range of the abnormal value, the accuracy of the estimated position can be improved.

  (3) In the position estimation server of the present invention, the position estimation unit may include a predicted value and an instantaneous value that incorporate a fluctuation of the received radio wave intensity over a certain period from the report data associated with the specific access point. The position of the specific access point is estimated with emphasis on data with small fluctuations in propagation path conditions, based on the fact that the difference is within a predetermined range. Thereby, since the position of the access point is estimated with emphasis on data with small fluctuations in the received radio wave intensity, the accuracy of the estimated position can be improved.

  (4) In addition, the position estimation server of the present invention determines whether or not the report data satisfies a condition provided by a criterion that places importance on data with high terminal position accuracy and small propagation path condition fluctuation. A determination unit; an extraction unit that extracts report data determined to satisfy the condition; and an estimated position calculation unit that calculates an estimated position of the specific access point using the extracted report data. It is characterized by that. In this way, since only data satisfying the conditions is used, the position of the access point can be estimated simply and with high accuracy.

  (5) Further, the position estimation server of the present invention includes a weight calculation unit that calculates weights for the report data on the basis of emphasizing data with high terminal position accuracy and small propagation path condition fluctuation. And an estimated position calculating unit that calculates an estimated position of the specific access point using the weighting. In this way, the accuracy of estimation of the position of the access point is improved using weighting. Further, all the obtained samples can be used, and the accuracy of position estimation can be improved.

  (6) Further, in the position estimation server of the present invention, the weight calculation unit calculates a difference between the predicted value and the instantaneous value that incorporates the fluctuation of the received radio wave intensity over a certain period, and propagates according to a ratio with a predetermined threshold. It is characterized by quantitatively evaluating the fluctuation state of the road and calculating the weighting of the data of each terminal position and received radio wave intensity. Thus, since weighting is calculated according to the magnitude of fluctuations in received radio wave intensity, highly accurate position estimation is possible.

  (7) Further, in the position estimation server of the present invention, the weight calculation unit calculates a weighting matrix using data of terminal position accuracy and propagation path fluctuation status of each report data, and the estimated position calculation unit includes: The estimated position of the specific access point is calculated using the terminal position and received radio wave intensity data of each report data and the calculated weighting matrix. In this way, by calculating the estimated position of the access point using the weighting matrix, the position of the access point can be estimated simply and with high accuracy.

  (8) The position estimation method of the present invention is a position estimation method for estimating the position of a wireless LAN access point, includes information on the position of the terminal and a propagation path, and is associated with a specific access point. The position of the specific access point is estimated by focusing on data with high terminal position accuracy and small fluctuations in propagation path conditions among the report data, referring to a database holding report data reported by It is characterized by that. As a result, the position of the access point can be estimated simply and with high accuracy using position information that varies in accuracy by a terminal not equipped with a special device.

  (9) A position estimation program according to the present invention is a position estimation program for estimating the position of a wireless LAN access point by causing a data management server to execute the program, and includes information on a terminal position and a propagation path. , Referring to a database that holds report data reported by a terminal in association with a specific access point, out of the report data, focusing on data with high terminal position accuracy and small fluctuations in propagation path conditions, The position of the specific access point is estimated. As a result, the position of the access point can be estimated simply and with high accuracy using position information that varies in accuracy by a terminal not equipped with a special device.

  According to the present invention, it is possible to estimate the position of an access point simply and with high accuracy using information with a variation in accuracy collected from a terminal not equipped with a special device.

1 is a schematic diagram showing a wireless LAN system according to the present invention. It is a block diagram which shows the functional structure of the management server (location estimation server) which concerns on 1st Embodiment. 3 is a sequence chart showing an operation of the wireless LAN system according to the first embodiment. It is the schematic which shows the operation | movement which excludes an abnormal value based on the statistical distribution of RSSI. It is a graph which shows the time fluctuation | variation of RSSI. It is a block diagram which shows the functional structure of the management server (location estimation server) which concerns on 2nd Embodiment. 6 is a sequence chart showing an operation of the wireless LAN system according to the second embodiment.

  Next, embodiments of the present invention will be described with reference to the drawings.

[Wireless LAN system]
FIG. 1 is a schematic diagram showing a wireless LAN system 10. In the example illustrated in FIG. 1, a wireless LAN system 10 is configured by a user terminal 20, an access point 30, a base station 40, and a management server (position estimation server) 100. In the wireless LAN system 10, the user terminal 20 transmits the access point ID (SSID) and the received signal strength indicator (RSSI) received as sensing results from the user terminal 20 as report data, and the management server 100. Collects and estimates the position of each access point.

  The user terminal 20 is equipped with a wireless LAN interface and can communicate with each access point. Further, the user terminal 20 is equipped with a GPS and can detect the position of the user terminal 20. A balloon attached to the user terminal 20 in FIG. 1 represents report data, and the report data indicates that the SSID, RSSI, terminal location, time, and the like are included. The user terminal 20 transmits report data to the management server 20 via the base station 40. The management server 20 that has received the report data stores and holds the report data.

  The management server 100 uses the data collected from the user terminal 20 to estimate the position of the wireless LAN access point. The management server 100 estimates the position of a specific access point by placing importance on the terminal position and propagation loss data with high terminal position accuracy and small propagation path condition fluctuation among the data for the specific access point. At that time, the position is estimated with reference to error factors included in each data used for calculating the estimated position. Error factors include (1) GPS positioning error of the user position and (2) RSSI fluctuation.

  By referring to the degree of these error factors, the position is estimated with emphasis on reliable report data. As a result, it is possible to prevent the estimated position of the access point, which is the final solution, from fluctuating due to information including more errors, and to derive a solution with higher accuracy.

  By using the obtained location information of the access point, for example, the mapping result of the access point as shown in FIG. 1 can be provided as a location information service at low cost. As a mapping result, the ID of the access point and its position can be displayed on the screen of the user terminal 20 together with the map in the vicinity.

[First Embodiment]
(Management server configuration)
FIG. 2 is a block diagram illustrating a functional configuration of the management server 100. The management server 100 includes a database 110, a position estimation unit 115, and a mapping unit 150. The position estimation unit 115 includes a determination unit 120, an extraction unit 130, and an estimated position calculation unit 140.

  The database 110 holds report data reported from the user terminal 20. That is, as the report data, the position of the terminal, its position accuracy, and RSSI data for the terminal are stored in association with the specific access point 30. Note that RSSI and propagation loss are equivalent, and either may be used.

  The position estimation unit 115 refers to the database 110 and estimates the position of the specific access point 30 by focusing on data with high terminal position accuracy and small fluctuations in propagation path conditions among report data. For example, the position of the specific access point 30 can be estimated based on the fact that the RSSI is not in the range of the abnormal value among the report data associated with the specific access point 30. Thereby, since the position of the access point 30 is estimated with emphasis on report data whose RSSI is not within the range of abnormal values, the accuracy of the estimated position can be improved.

  The position of the specific access point is determined based on the fact that the difference between the predicted value and the instantaneous value within the predetermined range of the report data associated with the specific access point 30 is within a predetermined range. It may be estimated. Thereby, since the position of the access point 30 is estimated with emphasis on data with small fluctuations in RSSI, the accuracy of the estimated position can be improved. Specific examples of such criteria will be described later.

  The determination unit 120 determines whether or not the report data satisfies a condition provided by a criterion that places importance on data with high terminal position accuracy and small propagation path condition variation. For example, it is determined whether at least one of terminal position accuracy and propagation path condition data satisfies a predetermined condition.

  The extraction unit 130 extracts report data determined to satisfy the condition. The estimated position calculation unit 140 calculates the estimated position of the specific access point 30 using the extracted data. As described above, since only the data satisfying the conditions is extracted and used, the position of the access point 30 can be simply and accurately estimated.

  The mapping unit 150 reflects the estimated position of the access point in the map information and outputs it. For example, when receiving a request for searching a surrounding access point from the terminal, map data displaying the access point together with the specific information is returned to the terminal.

  Note that the position estimation process as described above is a kind of triangulation, and in this case, the distance between the user terminal and the access point is important. The propagation distance can be calculated by calculating the propagation loss from the measured RSSI and applying a known model. However, since the wireless environment constantly changes, there is a possibility that a propagation loss that is different from the actual distance is measured. Therefore, a statistical distribution of the measured RSSI (that is, propagation loss) is calculated by the management server 100, and data including RSSI that is determined to be significantly different from the actual is not used for position estimation processing.

(Management server operation)
Next, the operation of the management server 100 configured as described above will be described. FIG. 3 is a sequence chart showing the operation of the management server 100. As shown in FIG. 3, the user terminal 20 performs sensing associated with the access point 30, and reports the sensing result to the management server 100 as report data (step S101).

  The management server 100 that has received the report data stores the report data in the database 110 (step S102). Such transmission / reception and storage of report data are repeated (steps S103 to S106). For example, the report data may be transmitted when time has elapsed, a location has been moved, processing has been completed, or the like.

  Next, triggered by some event, the management server 100 starts the process of estimating the position of the access point 30 (step S111). The management server 100 determines whether each data held in the database 110 satisfies a predetermined condition (step S112). If the predetermined condition is satisfied, the report data is extracted (step S113), and if the condition is not satisfied, the data is not extracted.

  Then, the estimated position of the access point 30 is calculated using the extracted data (step S114), and the position estimation process is completed (step S115). The estimated position of the access point 30 obtained in this way is used for processing such as mapping.

  As described above, the position of the access point 30 can be estimated simply and with high accuracy using position information that varies in accuracy by a terminal not equipped with a special device. The above operation can be executed by a program.

(Abnormal value detection of RSSI)
FIG. 4 is a schematic diagram illustrating an operation for eliminating an abnormal value based on a statistical distribution of RSSI. FIG. 4 shows a database and a histogram in the management server. The reported data is histogrammed based on RSSI to eliminate abnormal values.

  The management server 100 stores data of the terminal position “Xi, Yi, Zi” and the received radio wave intensity “Ri” (or propagation loss “Li”) in the database 110 in association with the access point ID “AP1”. ing. Then, when performing position estimation, the management server 100 refers to the database 110 and histograms the RSSI obtained for the specific access point 30 that is the position estimation target. In the histogram, for example, data not included within 90% is not used as an abnormal value in the position estimation process. Thereby, it is possible to estimate the position of the access point using only reliable RSSI data.

(Abnormal value detection from long interval fluctuation)
The RSSI value fluctuates at high speed even when the surrounding environment is stationary, but fluctuates even more when the user moves. On the other hand, when RSSI is observed in a certain long section rather than instantaneously, the fluctuation is reduced, and the fluctuation according to the surrounding environment can be stably captured. FIG. 5 is a graph showing the temporal variation of RSSI. In FIG. 5, the instantaneous fluctuation of RSSI and the fluctuation | variation which smoothed it in a fixed area are shown with the graph.

As shown in FIG. 5, it is possible to determine whether or not a large fluctuation has occurred instantaneously by comparing the value of the instantaneous fluctuation with the value smoothed in a certain period. In this way, only data included in the allowable range ± R from the smoothed value RSSI _ave is adopted as a reliable value. As a result, greatly fluctuating data can be excluded from the position estimation processing data.

[Second Embodiment]
(Management server configuration)
In the above embodiment, the report data is selected and used to estimate the position of the access point 30 depending on whether or not the condition is satisfied. However, the report data may be weighted to estimate the position of the access point 30. . FIG. 6 is a block diagram illustrating a functional configuration of the management server 200. The management server 200 includes a database 110, a position estimation unit 215, and a mapping unit 150. The position estimation unit 215 includes a weight calculation unit 220 and an estimated position calculation unit 240. Among these, the database 110 and the mapping unit 150 have the same functions as in the above embodiment.

  The weighting calculation unit 220 calculates the weighting of each terminal position and RSSI data based on a predetermined reference using at least one of terminal position accuracy and propagation path condition data. For example, the difference between the value obtained by performing statistical processing on the fluctuation of RSSI for a certain period and the instantaneous value is calculated, and the fluctuation state of the propagation path is quantitatively evaluated based on the ratio with the threshold provided separately. Thus, the weighting of each terminal position and RSSI data is calculated. That is, as will be described later, the weighting matrix is calculated using the data of the accuracy of the terminal position and the state of fluctuation of the propagation path for each measurement. In this way, the accuracy of estimation of the position of the access point 30 can be improved using weighting. Further, all the obtained samples can be used, and the accuracy of position estimation can be improved.

  The estimated position calculation unit 240 calculates an estimated position of a specific access point using the calculated weighting. The estimated position calculation unit 240 calculates the estimated position of a specific access point using the terminal position and RSSI for each report data and the calculated weighting matrix.

(Management server operation)
Next, the operation of the management server 200 configured as described above will be described. FIG. 7 is a sequence chart showing the operation of the management server 200. As shown in FIG. 7, the user terminal 20 performs sensing associated with the access point 30, and reports the sensing result to the management server 200 as report data (step S201). The management server 200 that has received the report data stores the report data in the database 110 (step S202). Such transmission / reception and storage of report data are repeated (steps S203 to S206).

  Next, triggered by some event, the management server 200 starts a process for estimating the position of the access point 30 (step S211). The management server 200 calculates a weight using each data held in the database 110 (step S212). The weighting can be calculated as a weighting matrix, for example, and details will be described later. Then, the estimated position of the access point 30 is calculated using the obtained weighting (step S213), and the position estimation process is completed (step S214). The estimated position of the access point 30 obtained in this way is used for processing such as mapping. The above operation can be executed by a program.

(Calculation of weight)
In the data reported from the user to the management server 200, there are cases where the accuracy of the location information of the user terminal itself is high, but the reliability of the RSSI value is low, or vice versa. Since these can be considered comprehensively, positioning can be performed with high accuracy without reducing the amount of data used for the position estimation process more than necessary.

The management server 200 calculates the position (x, y, z) of the access point using Expression (1). Here, (xi, yi, zi) is position information of the sensed location.

In order to obtain the solution of equation (1), equation (2) using a matrix is used.

Here, Δx = [x, y, z] ^T , Δr = [ri], and G is a design matrix. In order to weight the report data from the user to be used based on the reliability, a weighting matrix W is introduced to derive Equation (3).

Here, the matrix W is defined by Formula (4).

  DOPi is DOP (Dilution Of Precision) at the user's current location, and Si is the number of captured satellites. DOP is a value determined by the geometrical arrangement of the user and the GPS satellite, and the larger the value, the greater the error. The number S of captured satellites is the number of GPS satellites captured by the user terminal 20, and at least four are necessary.

  As the number S of captured satellites increases, the accuracy of the position information calculated from them increases. These values can be acquired from a general GPS receiver. Therefore, by reflecting these in the form of equation (4), it is possible to reduce the weight of the position information that can be determined to contain many errors.

On the other hand, ki is defined by the following equation, and is a ratio between a difference between RSSIi and a value predicted by long-term fluctuation (RSSI _ave ) and a threshold R set to determine a normal value.

Accordingly, it is possible to reduce the weight of information including RSSI having a large deviation from RSSI _ave , that is, RSSI indicating an abnormal value. From the above, when the DOP value is large, that is, when the reliability of the location information of the included user is low, the weight is low, and also when the RSSI fluctuation is large compared to the previous fluctuation and the reliability is low. By giving a low weight, the influence on the derived solution can be relatively reduced.

10 wireless LAN system 20 user terminal 30 access point 40 base station 100 management server (location estimation server)
110 Database 115 Position estimation unit 120 Determination unit 130 Extraction unit 140 Estimated position calculation unit 150 Mapping unit 200 Management server 215 Position estimation unit 220 Weight calculation unit 240 Estimated position calculation unit

Claims (9)

A position estimation server for estimating the position of a wireless LAN access point,
A database containing terminal location and propagation path information and holding report data reported by the terminal in association with a specific access point;
A position estimation unit that refers to the database and estimates the position of the specific access point by focusing on data with high terminal position accuracy and small fluctuations in propagation path conditions among the report data. A position estimation server characterized by   The position estimation unit is based on the fact that the received radio wave intensity is not in the range of the abnormal value among the report data associated with the specific access point, and emphasizes data with small fluctuations in propagation path conditions, The position estimation server according to claim 1, wherein the position of a specific access point is estimated.   The position estimation unit, based on the report data associated with the specific access point, the difference between the predicted value and the instantaneous value that incorporates the fluctuation of the received radio wave intensity over a certain period is within a predetermined range, The position estimation server according to claim 1, wherein the position of the specific access point is estimated with emphasis on data with small fluctuations in propagation path conditions. A determination unit that determines whether or not the report data satisfies a condition set by a criterion of emphasizing data with high terminal position accuracy and small fluctuations in propagation path conditions;
An extraction unit for extracting report data determined to satisfy the condition;
The position estimation server according to any one of claims 1 to 3, further comprising: an estimated position calculation unit that calculates an estimated position of the specific access point using the extracted report data. . A weight calculation unit that calculates the weight of the report data on the basis of high priority on data with high terminal position accuracy and small fluctuations in propagation path conditions;
The position estimation server according to claim 1, further comprising: an estimated position calculation unit that calculates an estimated position of the specific access point using the calculated weighting.   The weighting calculation unit calculates a difference between a predicted value and a instantaneous value that incorporates a fluctuation of a received radio wave intensity over a certain period, and quantitatively evaluates a fluctuation state of a propagation path according to a ratio with a predetermined threshold, 6. The position estimation server according to claim 5, wherein weighting of the terminal position and the received radio wave intensity is calculated. The weight calculation unit calculates a weighting matrix using data of terminal position accuracy and propagation path fluctuation status of each report data,
The estimated position calculation unit calculates an estimated position of the specific access point using the terminal position and received radio wave intensity data of each report data and the calculated weighting matrix. The position estimation server according to claim 5 or 6. A position estimation method for estimating a position of a wireless LAN access point,
Refers to a database that contains information on the location and propagation path of the terminal and holds report data reported by the terminal in association with a specific access point. Among the report data, the accuracy of the terminal location is high, and the propagation path status A position estimation method, wherein the position of the specific access point is estimated with emphasis on data with small fluctuations. A position estimation program for estimating the position of a wireless LAN access point by causing a server for data management to execute,
Refers to a database that contains information on the location and propagation path of the terminal and holds report data reported by the terminal in association with a specific access point. Among the report data, the accuracy of the terminal location is high, and the propagation path status A position estimation program for estimating the position of the specific access point by focusing on data with small fluctuations.

Images